Added 3D to 4D (tensor) transformation #19

mmrahorovic · 2021-01-15T14:23:04Z

[change_3d_tensors_to_4d]: 3D to 4D transformation of tensors. This transformation was needed to make QuartzNet compatible with other transformations (like LowerConvsToMatMul).
[test_4d_conversion]: added test cases for the 3D to 4D (tensor) transformation.

This pull request (PR) contains the changes from PR #17.

…m2col]: added/modified several test cases for (non-)square images and kernels

…ol]: added/modified several test cases for (non-)square images and kernels

…instead of integer as required by im2col node

[test_general_transformation]: changed the type of the kernel_size attribute to list of integers

…d kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution.

…in tests

…lutions in tests" This reverts commit 3ff449c

…mages and kernels and non-equal padding." This reverts commit 15e34ed.

This reverts commit c524020.

…on value must be equal along each axis. [test_im2col] added several test cases in case dilations = 2 (a.o. cases where image and kernel are 2D and 1D, with and without padding, and with stride = 2). [lower_convs_to_matmul] added support for dilation value. Dilation value must be equal along each axis. [test_conv_lowering] added test case for dilations = 2 for 2D and 1D images and kernels, with and without padding, and with stride = 2.

[test_conv_lowering] added more dilation values to test cases. Dilation values of {1, 2, 3, 4} are tested.

…on-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim [test_conv_lowering] changed function call to compute_conv_output_dim

…on-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion.

[lower_convs_to_matmul]: minor fix with assumption on kernel dimension

…D tensors to 4D and changes the nodes accordingly [test_4d_conversion]: test function for 3D to 4D tensor transformation

…ensors to 4D. [test_4d_conversion]: added a test case for the 3D to 4D transformation.

…et). [test_4d_conversion]: added test case for 3D to 4D transform.

[test_4d_conversion]: changed the test case for an invalid graph.

[lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated).

…is instantiated.

…tions.

…onvolutions

[test_conv_lowering]: merged test functions into one test function.

[test_4d_converions]: rearranged code.

…mmrahorovic/finn-base into feature/3d_to_4d_transformation

maltanar · 2021-02-06T12:20:07Z

src/finn/transformation/change_3d_tensors_to_4d.py

+        # Initializers
+        initializer_names = []  # list with all initializers
+        initializers = {}  # list with initializers from Conv, Mul, and Add nodes
+        for i in model.graph.initializer:
+            init_name = i.name
+            initializer_names.append(init_name)
+            if "Conv" in init_name:
+                init_dim = i.dims
+                init_dtype = i.data_type
+                initializers[init_name] = [init_dtype]
+                initializers[init_name].append(init_dim)
+            elif init_name[0:4] == "Mul_":
+                init_dim = i.dims
+                if len(i.dims) == 3:
+                    init_dtype = i.data_type
+                    initializers[init_name] = [init_dtype]
+                    initializers[init_name].append(init_dim)
+            elif "Add" in init_name:
+                init_dim = i.dims
+                if len(i.dims) == 3:
+                    init_dtype = i.data_type
+                    initializers[init_name] = [init_dtype]
+                    initializers[init_name].append(init_dim)


This part seems problematic to me. Is the intention to extract a list of initializers for particular node types? Currently you're checking the start of the initializer tensor name for this but that may not be reliable. A better way would be to build a list of nodes based on op_type in ["Conv","Mul","Add"], then use ModelWrapper.get_initializer() on the inputs of each node.

maltanar · 2021-02-06T12:23:40Z

src/finn/transformation/change_3d_tensors_to_4d.py

+        # Value infos
+        for t in model.graph.value_info:
+            tensor_name = t.name
+            if tensor_name in initializer_names:
+                continue
+            else:
+                tensor_type = t.type.tensor_type.elem_type
+                tensor_shape = model.get_tensor_shape(tensor_name)
+                tensor_names[tensor_name] = [tensor_type]
+                tensor_names[tensor_name].append(tensor_shape)
+
+        # Outputs
+        for t in model.graph.output:
+            tensor_name = t.name
+            tensor_type = t.type.tensor_type.elem_type
+            tensor_shape = model.get_tensor_shape(tensor_name)
+            tensor_names[tensor_name] = [tensor_type]
+            tensor_names[tensor_name].append(tensor_shape)


This bit (plus the bit that extracts the initializers above) could potentially be simplified by iterating on the list returned from ModelWrapper.get_all_tensor_names() -- a list of all input, output and value_info tensor names in the graph. You can then iterate over this list to check whether it needs to be considered for 3D->4D or not.

maltanar · 2021-02-06T12:25:54Z

Also one more general thing: for future PRs, can you please ensure you have merged dev into the PR branch for any outstanding PRs, especially if there are conflicts?

…nsformation

* Modified set_nodeattr to allow using it on repeated fields (#18) * [base]: changed how the floats, ints, strings, tensors, graphs and sparse_tensors field of AttributeProto is set. * [Core] restrict attributes to tested types Co-authored-by: Yaman Umuroglu <[email protected]> * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Support for non-square input images and kernels in LowerConvsToMatMul transformation (#16) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst Co-authored-by: Yaman Umuroglu <[email protected]> * Added support for dilation value = 2 for 1D and 2D images/kernels (#17) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] added support for dilations = 2 for 2D and 1D images. Dilation value must be equal along each axis. [test_im2col] added several test cases in case dilations = 2 (a.o. cases where image and kernel are 2D and 1D, with and without padding, and with stride = 2). [lower_convs_to_matmul] added support for dilation value. Dilation value must be equal along each axis. [test_conv_lowering] added test case for dilations = 2 for 2D and 1D images and kernels, with and without padding, and with stride = 2. * [lower_convs_to_matmul] removed old assertion [test_conv_lowering] added more dilation values to test cases. Dilation values of {1, 2, 3, 4} are tested. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim [test_conv_lowering] changed function call to compute_conv_output_dim * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * [im2col]: minor style adjustment. [test_conv_lowering]: merged test functions into one test function. * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Support for non-square input images and kernels in LowerConvsToMatMul transformation (#16) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst Co-authored-by: Yaman Umuroglu <[email protected]> Co-authored-by: Yaman Umuroglu <[email protected]> * Update quantavgpool2d.py (#22) Add "Copyright (c) 2021 Xilinx, Inc" heather * [batchnorm_to_affine]: epsilon value is now read out from the attributes. (#21) [test_batchnorm_to_affine]: added a test case for various epsilon values. * Added 3D to 4D (tensor) transformation (#19) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] added support for dilations = 2 for 2D and 1D images. Dilation value must be equal along each axis. [test_im2col] added several test cases in case dilations = 2 (a.o. cases where image and kernel are 2D and 1D, with and without padding, and with stride = 2). [lower_convs_to_matmul] added support for dilation value. Dilation value must be equal along each axis. [test_conv_lowering] added test case for dilations = 2 for 2D and 1D images and kernels, with and without padding, and with stride = 2. * [lower_convs_to_matmul] removed old assertion [test_conv_lowering] added more dilation values to test cases. Dilation values of {1, 2, 3, 4} are tested. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim [test_conv_lowering] changed function call to compute_conv_output_dim * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: minor fix with assumption on kernel dimension [lower_convs_to_matmul]: minor fix with assumption on kernel dimension * [change_3d_tensors_to_4d]: added new transformation that transforms 3D tensors to 4D and changes the nodes accordingly [test_4d_conversion]: test function for 3D to 4D tensor transformation * [change_3d_tensors_to_4d]: added new transformation that changes 3D tensors to 4D. [test_4d_conversion]: added a test case for the 3D to 4D transformation. * [change_3d_tensors_to_4d]: added 3D to 4D transformation (for QuartzNet). [test_4d_conversion]: added test case for 3D to 4D transform. * [change_3d_tensors_to_4d]: changed how an invalid graph is handled. [test_4d_conversion]: changed the test case for an invalid graph. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [im2col]: minor style change and changed the way how a square kernel is instantiated. * [test_conv_lowering]: merged tests for depthwise and standard convolutions. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * [im2col]: minor style adjustment. [test_conv_lowering]: merged test functions into one test function. * [change_3d_tensors_to_4d]: style fixes and comments. [test_4d_converions]: rearranged code. * [Transform] check invalid node list length * [change_3d_tensors_to_4d]: rearranged the code to make it more readable. Co-authored-by: Yaman Umuroglu <[email protected]> * [Docs] update tutorials * [Util] experimental: fast_mode data packing for binary (#24) * Generic partitioning feature (#23) * Add basic partitioning functionality * Mount build dir within docker container * Support for non-linear models and multi in/out partitions * Remove dataflowpartition custom op from finn-base * Fix temporary build dir for CI * Fix docstring * [create_generic_partitions]: minor modification, removed redundant output value_info entries. (#26) * [extend_partition]: added a new transformation ExtendPartition. (#27) [test_extend_partition]: added a test case for the new transformation. * Added support for non-equal strides along different axes (#25) * [im2col]: added support for non-equal strides along different axes and cleaned up the code. [lower_convs_to_matmul]: added support for non-equal strides along different axes and cleaned up the code. [test_conv_lowering]: added test case for non-equal strides along different axes. * [im2col]: minor fix. [test_im2col]: added test case for non-equal strides along different axes. * Changes for supporting vitis_hls (#28) * [Refactor] split up RTL/HLS-related utils * [Util] rename to CallHLS and allow specifying vivado_hls/vitis_hls * [Util] more flexible stream naming in rtlsim_multi_io * Changes for supporting non-equal dilation (#29) * added support for non-equal dilation value along (H, W) dimension * added test cases for non-equal dilation configurations * appending dilation value along dummy dimension correctly (i.e. with a '1') * changed tensor sparsity annotation for consistency * Support infer_datatype for flatten layer (#30) * Support infer_datatype for flatten layer * [InferDT] add more identity op types for datatype inference * [Lint] fix linting issues Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Create python-publish.yml * Add ZCU111 board to part map (#32) * Update AUTHORS.rst Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]>

* finn-base v0.0.2 (#34) * Modified set_nodeattr to allow using it on repeated fields (#18) * [base]: changed how the floats, ints, strings, tensors, graphs and sparse_tensors field of AttributeProto is set. * [Core] restrict attributes to tested types Co-authored-by: Yaman Umuroglu <[email protected]> * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Support for non-square input images and kernels in LowerConvsToMatMul transformation (#16) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst Co-authored-by: Yaman Umuroglu <[email protected]> * Added support for dilation value = 2 for 1D and 2D images/kernels (#17) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] added support for dilations = 2 for 2D and 1D images. Dilation value must be equal along each axis. [test_im2col] added several test cases in case dilations = 2 (a.o. cases where image and kernel are 2D and 1D, with and without padding, and with stride = 2). [lower_convs_to_matmul] added support for dilation value. Dilation value must be equal along each axis. [test_conv_lowering] added test case for dilations = 2 for 2D and 1D images and kernels, with and without padding, and with stride = 2. * [lower_convs_to_matmul] removed old assertion [test_conv_lowering] added more dilation values to test cases. Dilation values of {1, 2, 3, 4} are tested. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim [test_conv_lowering] changed function call to compute_conv_output_dim * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * [im2col]: minor style adjustment. [test_conv_lowering]: merged test functions into one test function. * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Support for non-square input images and kernels in LowerConvsToMatMul transformation (#16) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst Co-authored-by: Yaman Umuroglu <[email protected]> Co-authored-by: Yaman Umuroglu <[email protected]> * Update quantavgpool2d.py (#22) Add "Copyright (c) 2021 Xilinx, Inc" heather * [batchnorm_to_affine]: epsilon value is now read out from the attributes. (#21) [test_batchnorm_to_affine]: added a test case for various epsilon values. * Added 3D to 4D (tensor) transformation (#19) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] added support for dilations = 2 for 2D and 1D images. Dilation value must be equal along each axis. [test_im2col] added several test cases in case dilations = 2 (a.o. cases where image and kernel are 2D and 1D, with and without padding, and with stride = 2). [lower_convs_to_matmul] added support for dilation value. Dilation value must be equal along each axis. [test_conv_lowering] added test case for dilations = 2 for 2D and 1D images and kernels, with and without padding, and with stride = 2. * [lower_convs_to_matmul] removed old assertion [test_conv_lowering] added more dilation values to test cases. Dilation values of {1, 2, 3, 4} are tested. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim [test_conv_lowering] changed function call to compute_conv_output_dim * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: minor fix with assumption on kernel dimension [lower_convs_to_matmul]: minor fix with assumption on kernel dimension * [change_3d_tensors_to_4d]: added new transformation that transforms 3D tensors to 4D and changes the nodes accordingly [test_4d_conversion]: test function for 3D to 4D tensor transformation * [change_3d_tensors_to_4d]: added new transformation that changes 3D tensors to 4D. [test_4d_conversion]: added a test case for the 3D to 4D transformation. * [change_3d_tensors_to_4d]: added 3D to 4D transformation (for QuartzNet). [test_4d_conversion]: added test case for 3D to 4D transform. * [change_3d_tensors_to_4d]: changed how an invalid graph is handled. [test_4d_conversion]: changed the test case for an invalid graph. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [im2col]: minor style change and changed the way how a square kernel is instantiated. * [test_conv_lowering]: merged tests for depthwise and standard convolutions. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * [im2col]: minor style adjustment. [test_conv_lowering]: merged test functions into one test function. * [change_3d_tensors_to_4d]: style fixes and comments. [test_4d_converions]: rearranged code. * [Transform] check invalid node list length * [change_3d_tensors_to_4d]: rearranged the code to make it more readable. Co-authored-by: Yaman Umuroglu <[email protected]> * [Docs] update tutorials * [Util] experimental: fast_mode data packing for binary (#24) * Generic partitioning feature (#23) * Add basic partitioning functionality * Mount build dir within docker container * Support for non-linear models and multi in/out partitions * Remove dataflowpartition custom op from finn-base * Fix temporary build dir for CI * Fix docstring * [create_generic_partitions]: minor modification, removed redundant output value_info entries. (#26) * [extend_partition]: added a new transformation ExtendPartition. (#27) [test_extend_partition]: added a test case for the new transformation. * Added support for non-equal strides along different axes (#25) * [im2col]: added support for non-equal strides along different axes and cleaned up the code. [lower_convs_to_matmul]: added support for non-equal strides along different axes and cleaned up the code. [test_conv_lowering]: added test case for non-equal strides along different axes. * [im2col]: minor fix. [test_im2col]: added test case for non-equal strides along different axes. * Changes for supporting vitis_hls (#28) * [Refactor] split up RTL/HLS-related utils * [Util] rename to CallHLS and allow specifying vivado_hls/vitis_hls * [Util] more flexible stream naming in rtlsim_multi_io * Changes for supporting non-equal dilation (#29) * added support for non-equal dilation value along (H, W) dimension * added test cases for non-equal dilation configurations * appending dilation value along dummy dimension correctly (i.e. with a '1') * changed tensor sparsity annotation for consistency * Support infer_datatype for flatten layer (#30) * Support infer_datatype for flatten layer * [InferDT] add more identity op types for datatype inference * [Lint] fix linting issues Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Create python-publish.yml * Add ZCU111 board to part map (#32) * Update AUTHORS.rst Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]> * Update python-publish.yml * Update python-publish.yml * Lint Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]>

* [create_generic_partitions]: minor modification, removed redundant output value_info entries. (#26) * [extend_partition]: added a new transformation ExtendPartition. (#27) [test_extend_partition]: added a test case for the new transformation. * Added support for non-equal strides along different axes (#25) * [im2col]: added support for non-equal strides along different axes and cleaned up the code. [lower_convs_to_matmul]: added support for non-equal strides along different axes and cleaned up the code. [test_conv_lowering]: added test case for non-equal strides along different axes. * [im2col]: minor fix. [test_im2col]: added test case for non-equal strides along different axes. * Changes for supporting vitis_hls (#28) * [Refactor] split up RTL/HLS-related utils * [Util] rename to CallHLS and allow specifying vivado_hls/vitis_hls * [Util] more flexible stream naming in rtlsim_multi_io * Changes for supporting non-equal dilation (#29) * added support for non-equal dilation value along (H, W) dimension * added test cases for non-equal dilation configurations * appending dilation value along dummy dimension correctly (i.e. with a '1') * changed tensor sparsity annotation for consistency * Support infer_datatype for flatten layer (#30) * Support infer_datatype for flatten layer * [InferDT] add more identity op types for datatype inference * [Lint] fix linting issues Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Create python-publish.yml * Add ZCU111 board to part map (#32) * Update AUTHORS.rst * GHA for PyPI, sdist only (#35) * finn-base v0.0.2 (#34) * Modified set_nodeattr to allow using it on repeated fields (#18) * [base]: changed how the floats, ints, strings, tensors, graphs and sparse_tensors field of AttributeProto is set. * [Core] restrict attributes to tested types Co-authored-by: Yaman Umuroglu <[email protected]> * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Support for non-square input images and kernels in LowerConvsToMatMul transformation (#16) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst Co-authored-by: Yaman Umuroglu <[email protected]> * Added support for dilation value = 2 for 1D and 2D images/kernels (#17) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] added support for dilations = 2 for 2D and 1D images. Dilation value must be equal along each axis. [test_im2col] added several test cases in case dilations = 2 (a.o. cases where image and kernel are 2D and 1D, with and without padding, and with stride = 2). [lower_convs_to_matmul] added support for dilation value. Dilation value must be equal along each axis. [test_conv_lowering] added test case for dilations = 2 for 2D and 1D images and kernels, with and without padding, and with stride = 2. * [lower_convs_to_matmul] removed old assertion [test_conv_lowering] added more dilation values to test cases. Dilation values of {1, 2, 3, 4} are tested. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim [test_conv_lowering] changed function call to compute_conv_output_dim * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * [im2col]: minor style adjustment. [test_conv_lowering]: merged test functions into one test function. * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Support for non-square input images and kernels in LowerConvsToMatMul transformation (#16) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst Co-authored-by: Yaman Umuroglu <[email protected]> Co-authored-by: Yaman Umuroglu <[email protected]> * Update quantavgpool2d.py (#22) Add "Copyright (c) 2021 Xilinx, Inc" heather * [batchnorm_to_affine]: epsilon value is now read out from the attributes. (#21) [test_batchnorm_to_affine]: added a test case for various epsilon values. * Added 3D to 4D (tensor) transformation (#19) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] added support for dilations = 2 for 2D and 1D images. Dilation value must be equal along each axis. [test_im2col] added several test cases in case dilations = 2 (a.o. cases where image and kernel are 2D and 1D, with and without padding, and with stride = 2). [lower_convs_to_matmul] added support for dilation value. Dilation value must be equal along each axis. [test_conv_lowering] added test case for dilations = 2 for 2D and 1D images and kernels, with and without padding, and with stride = 2. * [lower_convs_to_matmul] removed old assertion [test_conv_lowering] added more dilation values to test cases. Dilation values of {1, 2, 3, 4} are tested. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim [test_conv_lowering] changed function call to compute_conv_output_dim * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: minor fix with assumption on kernel dimension [lower_convs_to_matmul]: minor fix with assumption on kernel dimension * [change_3d_tensors_to_4d]: added new transformation that transforms 3D tensors to 4D and changes the nodes accordingly [test_4d_conversion]: test function for 3D to 4D tensor transformation * [change_3d_tensors_to_4d]: added new transformation that changes 3D tensors to 4D. [test_4d_conversion]: added a test case for the 3D to 4D transformation. * [change_3d_tensors_to_4d]: added 3D to 4D transformation (for QuartzNet). [test_4d_conversion]: added test case for 3D to 4D transform. * [change_3d_tensors_to_4d]: changed how an invalid graph is handled. [test_4d_conversion]: changed the test case for an invalid graph. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [im2col]: minor style change and changed the way how a square kernel is instantiated. * [test_conv_lowering]: merged tests for depthwise and standard convolutions. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * [im2col]: minor style adjustment. [test_conv_lowering]: merged test functions into one test function. * [change_3d_tensors_to_4d]: style fixes and comments. [test_4d_converions]: rearranged code. * [Transform] check invalid node list length * [change_3d_tensors_to_4d]: rearranged the code to make it more readable. Co-authored-by: Yaman Umuroglu <[email protected]> * [Docs] update tutorials * [Util] experimental: fast_mode data packing for binary (#24) * Generic partitioning feature (#23) * Add basic partitioning functionality * Mount build dir within docker container * Support for non-linear models and multi in/out partitions * Remove dataflowpartition custom op from finn-base * Fix temporary build dir for CI * Fix docstring * [create_generic_partitions]: minor modification, removed redundant output value_info entries. (#26) * [extend_partition]: added a new transformation ExtendPartition. (#27) [test_extend_partition]: added a test case for the new transformation. * Added support for non-equal strides along different axes (#25) * [im2col]: added support for non-equal strides along different axes and cleaned up the code. [lower_convs_to_matmul]: added support for non-equal strides along different axes and cleaned up the code. [test_conv_lowering]: added test case for non-equal strides along different axes. * [im2col]: minor fix. [test_im2col]: added test case for non-equal strides along different axes. * Changes for supporting vitis_hls (#28) * [Refactor] split up RTL/HLS-related utils * [Util] rename to CallHLS and allow specifying vivado_hls/vitis_hls * [Util] more flexible stream naming in rtlsim_multi_io * Changes for supporting non-equal dilation (#29) * added support for non-equal dilation value along (H, W) dimension * added test cases for non-equal dilation configurations * appending dilation value along dummy dimension correctly (i.e. with a '1') * changed tensor sparsity annotation for consistency * Support infer_datatype for flatten layer (#30) * Support infer_datatype for flatten layer * [InferDT] add more identity op types for datatype inference * [Lint] fix linting issues Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Create python-publish.yml * Add ZCU111 board to part map (#32) * Update AUTHORS.rst Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]> * Update python-publish.yml * Update python-publish.yml * Lint Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]> Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]> Co-authored-by: jalezeta <[email protected]>

* Add default Verilator args to disable warnings, force inlining * Update AUTHORS.rst * Create python-publish.yml * Add ZCU111 board to part map (#32) * Update AUTHORS.rst * GHA for PyPI, sdist only (#35) * finn-base v0.0.2 (#34) * Modified set_nodeattr to allow using it on repeated fields (#18) * [base]: changed how the floats, ints, strings, tensors, graphs and sparse_tensors field of AttributeProto is set. * [Core] restrict attributes to tested types Co-authored-by: Yaman Umuroglu <[email protected]> * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Support for non-square input images and kernels in LowerConvsToMatMul transformation (#16) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst Co-authored-by: Yaman Umuroglu <[email protected]> * Added support for dilation value = 2 for 1D and 2D images/kernels (#17) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] added support for dilations = 2 for 2D and 1D images. Dilation value must be equal along each axis. [test_im2col] added several test cases in case dilations = 2 (a.o. cases where image and kernel are 2D and 1D, with and without padding, and with stride = 2). [lower_convs_to_matmul] added support for dilation value. Dilation value must be equal along each axis. [test_conv_lowering] added test case for dilations = 2 for 2D and 1D images and kernels, with and without padding, and with stride = 2. * [lower_convs_to_matmul] removed old assertion [test_conv_lowering] added more dilation values to test cases. Dilation values of {1, 2, 3, 4} are tested. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim [test_conv_lowering] changed function call to compute_conv_output_dim * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * [im2col]: minor style adjustment. [test_conv_lowering]: merged test functions into one test function. * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Support for non-square input images and kernels in LowerConvsToMatMul transformation (#16) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst Co-authored-by: Yaman Umuroglu <[email protected]> Co-authored-by: Yaman Umuroglu <[email protected]> * Update quantavgpool2d.py (#22) Add "Copyright (c) 2021 Xilinx, Inc" heather * [batchnorm_to_affine]: epsilon value is now read out from the attributes. (#21) [test_batchnorm_to_affine]: added a test case for various epsilon values. * Added 3D to 4D (tensor) transformation (#19) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] added support for dilations = 2 for 2D and 1D images. Dilation value must be equal along each axis. [test_im2col] added several test cases in case dilations = 2 (a.o. cases where image and kernel are 2D and 1D, with and without padding, and with stride = 2). [lower_convs_to_matmul] added support for dilation value. Dilation value must be equal along each axis. [test_conv_lowering] added test case for dilations = 2 for 2D and 1D images and kernels, with and without padding, and with stride = 2. * [lower_convs_to_matmul] removed old assertion [test_conv_lowering] added more dilation values to test cases. Dilation values of {1, 2, 3, 4} are tested. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim [test_conv_lowering] changed function call to compute_conv_output_dim * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: minor fix with assumption on kernel dimension [lower_convs_to_matmul]: minor fix with assumption on kernel dimension * [change_3d_tensors_to_4d]: added new transformation that transforms 3D tensors to 4D and changes the nodes accordingly [test_4d_conversion]: test function for 3D to 4D tensor transformation * [change_3d_tensors_to_4d]: added new transformation that changes 3D tensors to 4D. [test_4d_conversion]: added a test case for the 3D to 4D transformation. * [change_3d_tensors_to_4d]: added 3D to 4D transformation (for QuartzNet). [test_4d_conversion]: added test case for 3D to 4D transform. * [change_3d_tensors_to_4d]: changed how an invalid graph is handled. [test_4d_conversion]: changed the test case for an invalid graph. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [im2col]: minor style change and changed the way how a square kernel is instantiated. * [test_conv_lowering]: merged tests for depthwise and standard convolutions. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * [im2col]: minor style adjustment. [test_conv_lowering]: merged test functions into one test function. * [change_3d_tensors_to_4d]: style fixes and comments. [test_4d_converions]: rearranged code. * [Transform] check invalid node list length * [change_3d_tensors_to_4d]: rearranged the code to make it more readable. Co-authored-by: Yaman Umuroglu <[email protected]> * [Docs] update tutorials * [Util] experimental: fast_mode data packing for binary (#24) * Generic partitioning feature (#23) * Add basic partitioning functionality * Mount build dir within docker container * Support for non-linear models and multi in/out partitions * Remove dataflowpartition custom op from finn-base * Fix temporary build dir for CI * Fix docstring * [create_generic_partitions]: minor modification, removed redundant output value_info entries. (#26) * [extend_partition]: added a new transformation ExtendPartition. (#27) [test_extend_partition]: added a test case for the new transformation. * Added support for non-equal strides along different axes (#25) * [im2col]: added support for non-equal strides along different axes and cleaned up the code. [lower_convs_to_matmul]: added support for non-equal strides along different axes and cleaned up the code. [test_conv_lowering]: added test case for non-equal strides along different axes. * [im2col]: minor fix. [test_im2col]: added test case for non-equal strides along different axes. * Changes for supporting vitis_hls (#28) * [Refactor] split up RTL/HLS-related utils * [Util] rename to CallHLS and allow specifying vivado_hls/vitis_hls * [Util] more flexible stream naming in rtlsim_multi_io * Changes for supporting non-equal dilation (#29) * added support for non-equal dilation value along (H, W) dimension * added test cases for non-equal dilation configurations * appending dilation value along dummy dimension correctly (i.e. with a '1') * changed tensor sparsity annotation for consistency * Support infer_datatype for flatten layer (#30) * Support infer_datatype for flatten layer * [InferDT] add more identity op types for datatype inference * [Lint] fix linting issues Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Create python-publish.yml * Add ZCU111 board to part map (#32) * Update AUTHORS.rst Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]> * Update python-publish.yml * Update python-publish.yml * Lint Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]> * fix broken logo in README * Extend Change3DTo4DTensors() to more layers (#31) * Support infer_datatype for flatten layer * 3d_to_4d support for additional nodes * Bugfix * Add ZCU111 to part map * Fix ZCU111 part * Separate PRs * Bugfix, new test case * [Docker] pin pyverilator version * [Util] pyverilate_stitched_ip: disable_common_warnings optional Co-authored-by: Yaman Umuroglu <[email protected]> Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]>

* Update AUTHORS.rst * Create python-publish.yml * Add ZCU111 board to part map (#32) * Update AUTHORS.rst * GHA for PyPI, sdist only (#35) * finn-base v0.0.2 (#34) * Modified set_nodeattr to allow using it on repeated fields (#18) * [base]: changed how the floats, ints, strings, tensors, graphs and sparse_tensors field of AttributeProto is set. * [Core] restrict attributes to tested types Co-authored-by: Yaman Umuroglu <[email protected]> * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Support for non-square input images and kernels in LowerConvsToMatMul transformation (#16) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst Co-authored-by: Yaman Umuroglu <[email protected]> * Added support for dilation value = 2 for 1D and 2D images/kernels (#17) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] added support for dilations = 2 for 2D and 1D images. Dilation value must be equal along each axis. [test_im2col] added several test cases in case dilations = 2 (a.o. cases where image and kernel are 2D and 1D, with and without padding, and with stride = 2). [lower_convs_to_matmul] added support for dilation value. Dilation value must be equal along each axis. [test_conv_lowering] added test case for dilations = 2 for 2D and 1D images and kernels, with and without padding, and with stride = 2. * [lower_convs_to_matmul] removed old assertion [test_conv_lowering] added more dilation values to test cases. Dilation values of {1, 2, 3, 4} are tested. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim [test_conv_lowering] changed function call to compute_conv_output_dim * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * [im2col]: minor style adjustment. [test_conv_lowering]: merged test functions into one test function. * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Support for non-square input images and kernels in LowerConvsToMatMul transformation (#16) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst Co-authored-by: Yaman Umuroglu <[email protected]> Co-authored-by: Yaman Umuroglu <[email protected]> * Update quantavgpool2d.py (#22) Add "Copyright (c) 2021 Xilinx, Inc" heather * [batchnorm_to_affine]: epsilon value is now read out from the attributes. (#21) [test_batchnorm_to_affine]: added a test case for various epsilon values. * Added 3D to 4D (tensor) transformation (#19) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] added support for dilations = 2 for 2D and 1D images. Dilation value must be equal along each axis. [test_im2col] added several test cases in case dilations = 2 (a.o. cases where image and kernel are 2D and 1D, with and without padding, and with stride = 2). [lower_convs_to_matmul] added support for dilation value. Dilation value must be equal along each axis. [test_conv_lowering] added test case for dilations = 2 for 2D and 1D images and kernels, with and without padding, and with stride = 2. * [lower_convs_to_matmul] removed old assertion [test_conv_lowering] added more dilation values to test cases. Dilation values of {1, 2, 3, 4} are tested. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim [test_conv_lowering] changed function call to compute_conv_output_dim * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: minor fix with assumption on kernel dimension [lower_convs_to_matmul]: minor fix with assumption on kernel dimension * [change_3d_tensors_to_4d]: added new transformation that transforms 3D tensors to 4D and changes the nodes accordingly [test_4d_conversion]: test function for 3D to 4D tensor transformation * [change_3d_tensors_to_4d]: added new transformation that changes 3D tensors to 4D. [test_4d_conversion]: added a test case for the 3D to 4D transformation. * [change_3d_tensors_to_4d]: added 3D to 4D transformation (for QuartzNet). [test_4d_conversion]: added test case for 3D to 4D transform. * [change_3d_tensors_to_4d]: changed how an invalid graph is handled. [test_4d_conversion]: changed the test case for an invalid graph. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [im2col]: minor style change and changed the way how a square kernel is instantiated. * [test_conv_lowering]: merged tests for depthwise and standard convolutions. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * [im2col]: minor style adjustment. [test_conv_lowering]: merged test functions into one test function. * [change_3d_tensors_to_4d]: style fixes and comments. [test_4d_converions]: rearranged code. * [Transform] check invalid node list length * [change_3d_tensors_to_4d]: rearranged the code to make it more readable. Co-authored-by: Yaman Umuroglu <[email protected]> * [Docs] update tutorials * [Util] experimental: fast_mode data packing for binary (#24) * Generic partitioning feature (#23) * Add basic partitioning functionality * Mount build dir within docker container * Support for non-linear models and multi in/out partitions * Remove dataflowpartition custom op from finn-base * Fix temporary build dir for CI * Fix docstring * [create_generic_partitions]: minor modification, removed redundant output value_info entries. (#26) * [extend_partition]: added a new transformation ExtendPartition. (#27) [test_extend_partition]: added a test case for the new transformation. * Added support for non-equal strides along different axes (#25) * [im2col]: added support for non-equal strides along different axes and cleaned up the code. [lower_convs_to_matmul]: added support for non-equal strides along different axes and cleaned up the code. [test_conv_lowering]: added test case for non-equal strides along different axes. * [im2col]: minor fix. [test_im2col]: added test case for non-equal strides along different axes. * Changes for supporting vitis_hls (#28) * [Refactor] split up RTL/HLS-related utils * [Util] rename to CallHLS and allow specifying vivado_hls/vitis_hls * [Util] more flexible stream naming in rtlsim_multi_io * Changes for supporting non-equal dilation (#29) * added support for non-equal dilation value along (H, W) dimension * added test cases for non-equal dilation configurations * appending dilation value along dummy dimension correctly (i.e. with a '1') * changed tensor sparsity annotation for consistency * Support infer_datatype for flatten layer (#30) * Support infer_datatype for flatten layer * [InferDT] add more identity op types for datatype inference * [Lint] fix linting issues Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Create python-publish.yml * Add ZCU111 board to part map (#32) * Update AUTHORS.rst Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]> * Update python-publish.yml * Update python-publish.yml * Lint Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]> * Inference cost: JSON export, parameter for output onnx * Added support for cost estimation for upsampling Co-authored-by: Yaman Umuroglu <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]> Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]>

* [Op] add skeleton for Quant custom op * [Op] more comments on Quant op * [InferDT] add MaxPool to list of identity ops for datatype inf. * [Transform] temporarily exclude Quant ops from const folding * [Analysis] add skeleton for inference_cost, limited conv support * [Analysis] support inference_cost_matmul * [Deps] add clize * [Util] add inference_cost util with entrypoint * [Docker] add clize * [Infra] add entrypoint for inference_cost * Preliminary implementation of quant operation * Fixed wrong numpy type in quant op * Made Quant pre-commit compatible * Changed docker base image to PyTorch and added Brevitas dockerfile and entrypoint * Implemented preliminary test for Quant operation * Reworked comments for Quant operation * Readability improvements for Quant operation * Created preliminary Quant op documentation * Move Quant op documentation * Made documentation of quant attributes more precise. * Created skeleton for Trunc QNNX operation * [DataType] add experimental SCALED(U)INT DataTypes * [Quant] use new scaled int dtypes for dt inference * [TypInf] use non-integer instead of float check in dt inference * [ConstFold] make op exclusion optional * [Analysis] deal with dyn matmul, add some zero-cost op types * [Analysis] make inference_cost optionally sparse-weights-aware * [Util] pass sparsity option to analysis pass, kwargs only for clize * [Analysis] add more free ops to inference_cost * [Quant] fallback to float32 for failed dtype inference * Inference cost: JSON export, parameter for output onnx * Removed FINN dependency from documentation * Added support for executing qonnx custom_ops. * Add support for Bipolar and Binary FINN datatype for Quant op. (#41) * Add support for Bipolar and Binary FINN datatype for Quant op. * [Quant] custom qnt execution for bipolar until Brevitas bug resolved Co-authored-by: Yaman Umuroglu <[email protected]> * Added support for cost estimation for upsampling (#37) * Update AUTHORS.rst * Create python-publish.yml * Add ZCU111 board to part map (#32) * Update AUTHORS.rst * GHA for PyPI, sdist only (#35) * finn-base v0.0.2 (#34) * Modified set_nodeattr to allow using it on repeated fields (#18) * [base]: changed how the floats, ints, strings, tensors, graphs and sparse_tensors field of AttributeProto is set. * [Core] restrict attributes to tested types Co-authored-by: Yaman Umuroglu <[email protected]> * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Support for non-square input images and kernels in LowerConvsToMatMul transformation (#16) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst Co-authored-by: Yaman Umuroglu <[email protected]> * Added support for dilation value = 2 for 1D and 2D images/kernels (#17) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] added support for dilations = 2 for 2D and 1D images. Dilation value must be equal along each axis. [test_im2col] added several test cases in case dilations = 2 (a.o. cases where image and kernel are 2D and 1D, with and without padding, and with stride = 2). [lower_convs_to_matmul] added support for dilation value. Dilation value must be equal along each axis. [test_conv_lowering] added test case for dilations = 2 for 2D and 1D images and kernels, with and without padding, and with stride = 2. * [lower_convs_to_matmul] removed old assertion [test_conv_lowering] added more dilation values to test cases. Dilation values of {1, 2, 3, 4} are tested. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim [test_conv_lowering] changed function call to compute_conv_output_dim * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * [im2col]: minor style adjustment. [test_conv_lowering]: merged test functions into one test function. * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Support for non-square input images and kernels in LowerConvsToMatMul transformation (#16) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst Co-authored-by: Yaman Umuroglu <[email protected]> Co-authored-by: Yaman Umuroglu <[email protected]> * Update quantavgpool2d.py (#22) Add "Copyright (c) 2021 Xilinx, Inc" heather * [batchnorm_to_affine]: epsilon value is now read out from the attributes. (#21) [test_batchnorm_to_affine]: added a test case for various epsilon values. * Added 3D to 4D (tensor) transformation (#19) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020. * [im2col] added support for dilations = 2 for 2D and 1D images. Dilation value must be equal along each axis. [test_im2col] added several test cases in case dilations = 2 (a.o. cases where image and kernel are 2D and 1D, with and without padding, and with stride = 2). [lower_convs_to_matmul] added support for dilation value. Dilation value must be equal along each axis. [test_conv_lowering] added test case for dilations = 2 for 2D and 1D images and kernels, with and without padding, and with stride = 2. * [lower_convs_to_matmul] removed old assertion [test_conv_lowering] added more dilation values to test cases. Dilation values of {1, 2, 3, 4} are tested. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim [test_conv_lowering] changed function call to compute_conv_output_dim * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: minor fix with assumption on kernel dimension [lower_convs_to_matmul]: minor fix with assumption on kernel dimension * [change_3d_tensors_to_4d]: added new transformation that transforms 3D tensors to 4D and changes the nodes accordingly [test_4d_conversion]: test function for 3D to 4D tensor transformation * [change_3d_tensors_to_4d]: added new transformation that changes 3D tensors to 4D. [test_4d_conversion]: added a test case for the 3D to 4D transformation. * [change_3d_tensors_to_4d]: added 3D to 4D transformation (for QuartzNet). [test_4d_conversion]: added test case for 3D to 4D transform. * [change_3d_tensors_to_4d]: changed how an invalid graph is handled. [test_4d_conversion]: changed the test case for an invalid graph. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [im2col]: minor style change and changed the way how a square kernel is instantiated. * [test_conv_lowering]: merged tests for depthwise and standard convolutions. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * [im2col]: minor style adjustment. [test_conv_lowering]: merged test functions into one test function. * [change_3d_tensors_to_4d]: style fixes and comments. [test_4d_converions]: rearranged code. * [Transform] check invalid node list length * [change_3d_tensors_to_4d]: rearranged the code to make it more readable. Co-authored-by: Yaman Umuroglu <[email protected]> * [Docs] update tutorials * [Util] experimental: fast_mode data packing for binary (#24) * Generic partitioning feature (#23) * Add basic partitioning functionality * Mount build dir within docker container * Support for non-linear models and multi in/out partitions * Remove dataflowpartition custom op from finn-base * Fix temporary build dir for CI * Fix docstring * [create_generic_partitions]: minor modification, removed redundant output value_info entries. (#26) * [extend_partition]: added a new transformation ExtendPartition. (#27) [test_extend_partition]: added a test case for the new transformation. * Added support for non-equal strides along different axes (#25) * [im2col]: added support for non-equal strides along different axes and cleaned up the code. [lower_convs_to_matmul]: added support for non-equal strides along different axes and cleaned up the code. [test_conv_lowering]: added test case for non-equal strides along different axes. * [im2col]: minor fix. [test_im2col]: added test case for non-equal strides along different axes. * Changes for supporting vitis_hls (#28) * [Refactor] split up RTL/HLS-related utils * [Util] rename to CallHLS and allow specifying vivado_hls/vitis_hls * [Util] more flexible stream naming in rtlsim_multi_io * Changes for supporting non-equal dilation (#29) * added support for non-equal dilation value along (H, W) dimension * added test cases for non-equal dilation configurations * appending dilation value along dummy dimension correctly (i.e. with a '1') * changed tensor sparsity annotation for consistency * Support infer_datatype for flatten layer (#30) * Support infer_datatype for flatten layer * [InferDT] add more identity op types for datatype inference * [Lint] fix linting issues Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Create python-publish.yml * Add ZCU111 board to part map (#32) * Update AUTHORS.rst Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]> * Update python-publish.yml * Update python-publish.yml * Lint Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]> * Inference cost: JSON export, parameter for output onnx * Added support for cost estimation for upsampling Co-authored-by: Yaman Umuroglu <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]> Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]> * [Quant] fix edge case for 0d input array execution * Added trunc support for MobileNet, needs some fixing. * Made trunc datatype inference pass-through. * Addressed numpy deprecation warning in Truc op exectution. * Updated Quant op to support rounding mode. * Added support for new Trunc op. * Added support for static datatype inference to the Trunc op. * Resolved left over ToDo in Trunc op. * Remove scaledInt from Trunc op and replace with float32. * Moved RemoveIdentityOps transformation from FINN to finn-base. * Fixed a bug in RemoveIdentityOps, where an identity op would not get removed if it was preceded by another identity op. * Moved GemmToMatMul from FINN to finn-base. * Added support for recognizing MultiThreshold node in sign discovery. * Moved RemoveEmptyPadding transformation from finn to finn-base. * Updated Quant op documentation to add the rounding_mode attribute. * Added documentation for trunc op and small adjustments for Quant op documentation. * Added preliminary support for BinaryQuant node. * Updated datatype inference for Quant and BinaryQuant to remove scaledInt types. * Fixed bug in datatype inference for Quant node. * Added support for merging into non-linear graphs for _remove_node_and_rewire function. * Added another op to sign_preserving_ops. * Updated Trunc op documentation. * Extended graph based signed inference to more quantization nodes. * Renamed BinaryQuant to BipolarQuant. * Renamed BinaryQuant to BipolarQuant for Trunc op Signed inference. * Updated BipolarQuant op documentation and moved documentation of QONNX ops into their own folder. * Made remove_node_and_rewire function part of the publicly facing api. * Adapt test_remove_identity_ops to new datatype system. * Made padding removal more reliable. * Moved ExtractBiasFromConv transformation from FINN to finn-base. * Fixed changed import for test_remove_identity_ops test. * Extended MultiThreshold data type inference to adjust for float scale/bias. * Removed test_brevitas_quant_onnx_export_and_exec test, since test_QONNX_to_FINN in FINN already tests for the same. * Fixed spelling typos in QONNX op documentation. * Specified rounding operation in Quant and Trunc documentation. * Updated version description for QONNX ops documentation. * Removed dependency on onnxoptimizer. * Refactored check for float. * Refactored get_internal_dtype to get_integer_datatype. * Disabled constant folding for Quant and BipolarQuant by default. * Added a warning if an unsupported ONNX version is encountered during the GemmToMatMul transformation. * Moved functionality of RemoveEmptyPadding into RemoveIdentityOps. * Made float32 default output for Trunc node and removed sign inference from node. Co-authored-by: Yaman Umuroglu <[email protected]> Co-authored-by: Yaman Umuroglu <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]> Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]>

* GHA for PyPI, sdist only (#35) * finn-base v0.0.2 (#34) * Modified set_nodeattr to allow using it on repeated fields (#18) * [base]: changed how the floats, ints, strings, tensors, graphs and sparse_tensors field of AttributeProto is set. * [Core] restrict attributes to tested types Co-authored-by: Yaman Umuroglu <[email protected]> * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Support for non-square input images and kernels in LowerConvsToMatMul transformation (#16) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst Co-authored-by: Yaman Umuroglu <[email protected]> * Added support for dilation value = 2 for 1D and 2D images/kernels (#17) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col] added support for dilations = 2 for 2D and 1D images. Dilation value must be equal along each axis. [test_im2col] added several test cases in case dilations = 2 (a.o. cases where image and kernel are 2D and 1D, with and without padding, and with stride = 2). [lower_convs_to_matmul] added support for dilation value. Dilation value must be equal along each axis. [test_conv_lowering] added test case for dilations = 2 for 2D and 1D images and kernels, with and without padding, and with stride = 2. * [lower_convs_to_matmul] removed old assertion [test_conv_lowering] added more dilation values to test cases. Dilation values of {1, 2, 3, 4} are tested. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim [test_conv_lowering] changed function call to compute_conv_output_dim * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * [im2col]: minor style adjustment. [test_conv_lowering]: merged test functions into one test function. * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Support for non-square input images and kernels in LowerConvsToMatMul transformation (#16) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst Co-authored-by: Yaman Umuroglu <[email protected]> Co-authored-by: Yaman Umuroglu <[email protected]> * Update quantavgpool2d.py (#22) Add "Copyright (c) 2021 Xilinx, Inc" heather * [batchnorm_to_affine]: epsilon value is now read out from the attributes. (#21) [test_batchnorm_to_affine]: added a test case for various epsilon values. * Added 3D to 4D (tensor) transformation (#19) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col] added support for dilations = 2 for 2D and 1D images. Dilation value must be equal along each axis. [test_im2col] added several test cases in case dilations = 2 (a.o. cases where image and kernel are 2D and 1D, with and without padding, and with stride = 2). [lower_convs_to_matmul] added support for dilation value. Dilation value must be equal along each axis. [test_conv_lowering] added test case for dilations = 2 for 2D and 1D images and kernels, with and without padding, and with stride = 2. * [lower_convs_to_matmul] removed old assertion [test_conv_lowering] added more dilation values to test cases. Dilation values of {1, 2, 3, 4} are tested. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim [test_conv_lowering] changed function call to compute_conv_output_dim * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: minor fix with assumption on kernel dimension [lower_convs_to_matmul]: minor fix with assumption on kernel dimension * [change_3d_tensors_to_4d]: added new transformation that transforms 3D tensors to 4D and changes the nodes accordingly [test_4d_conversion]: test function for 3D to 4D tensor transformation * [change_3d_tensors_to_4d]: added new transformation that changes 3D tensors to 4D. [test_4d_conversion]: added a test case for the 3D to 4D transformation. * [change_3d_tensors_to_4d]: added 3D to 4D transformation (for QuartzNet). [test_4d_conversion]: added test case for 3D to 4D transform. * [change_3d_tensors_to_4d]: changed how an invalid graph is handled. [test_4d_conversion]: changed the test case for an invalid graph. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [im2col]: minor style change and changed the way how a square kernel is instantiated. * [test_conv_lowering]: merged tests for depthwise and standard convolutions. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * [im2col]: minor style adjustment. [test_conv_lowering]: merged test functions into one test function. * [change_3d_tensors_to_4d]: style fixes and comments. [test_4d_converions]: rearranged code. * [Transform] check invalid node list length * [change_3d_tensors_to_4d]: rearranged the code to make it more readable. Co-authored-by: Yaman Umuroglu <[email protected]> * [Docs] update tutorials * [Util] experimental: fast_mode data packing for binary (#24) * Generic partitioning feature (#23) * Add basic partitioning functionality * Mount build dir within docker container * Support for non-linear models and multi in/out partitions * Remove dataflowpartition custom op from finn-base * Fix temporary build dir for CI * Fix docstring * [create_generic_partitions]: minor modification, removed redundant output value_info entries. (#26) * [extend_partition]: added a new transformation ExtendPartition. (#27) [test_extend_partition]: added a test case for the new transformation. * Added support for non-equal strides along different axes (#25) * [im2col]: added support for non-equal strides along different axes and cleaned up the code. [lower_convs_to_matmul]: added support for non-equal strides along different axes and cleaned up the code. [test_conv_lowering]: added test case for non-equal strides along different axes. * [im2col]: minor fix. [test_im2col]: added test case for non-equal strides along different axes. * Changes for supporting vitis_hls (#28) * [Refactor] split up RTL/HLS-related utils * [Util] rename to CallHLS and allow specifying vivado_hls/vitis_hls * [Util] more flexible stream naming in rtlsim_multi_io * Changes for supporting non-equal dilation (#29) * added support for non-equal dilation value along (H, W) dimension * added test cases for non-equal dilation configurations * appending dilation value along dummy dimension correctly (i.e. with a '1') * changed tensor sparsity annotation for consistency * Support infer_datatype for flatten layer (#30) * Support infer_datatype for flatten layer * [InferDT] add more identity op types for datatype inference * [Lint] fix linting issues Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Create python-publish.yml * Add ZCU111 board to part map (#32) * Update AUTHORS.rst Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]> * Update python-publish.yml * Update python-publish.yml * Lint Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]> * fix broken logo in README * Extend Change3DTo4DTensors() to more layers (#31) * Support infer_datatype for flatten layer * 3d_to_4d support for additional nodes * Bugfix * Add ZCU111 to part map * Fix ZCU111 part * Separate PRs * Bugfix, new test case * Verilator: force inlining, disable common warnings (#33) * Add default Verilator args to disable warnings, force inlining * Update AUTHORS.rst * Create python-publish.yml * Add ZCU111 board to part map (#32) * Update AUTHORS.rst * GHA for PyPI, sdist only (#35) * finn-base v0.0.2 (#34) * Modified set_nodeattr to allow using it on repeated fields (#18) * [base]: changed how the floats, ints, strings, tensors, graphs and sparse_tensors field of AttributeProto is set. * [Core] restrict attributes to tested types Co-authored-by: Yaman Umuroglu <[email protected]> * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Support for non-square input images and kernels in LowerConvsToMatMul transformation (#16) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst Co-authored-by: Yaman Umuroglu <[email protected]> * Added support for dilation value = 2 for 1D and 2D images/kernels (#17) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col] added support for dilations = 2 for 2D and 1D images. Dilation value must be equal along each axis. [test_im2col] added several test cases in case dilations = 2 (a.o. cases where image and kernel are 2D and 1D, with and without padding, and with stride = 2). [lower_convs_to_matmul] added support for dilation value. Dilation value must be equal along each axis. [test_conv_lowering] added test case for dilations = 2 for 2D and 1D images and kernels, with and without padding, and with stride = 2. * [lower_convs_to_matmul] removed old assertion [test_conv_lowering] added more dilation values to test cases. Dilation values of {1, 2, 3, 4} are tested. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim [test_conv_lowering] changed function call to compute_conv_output_dim * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * [im2col]: minor style adjustment. [test_conv_lowering]: merged test functions into one test function. * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Support for non-square input images and kernels in LowerConvsToMatMul transformation (#16) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst Co-authored-by: Yaman Umuroglu <[email protected]> Co-authored-by: Yaman Umuroglu <[email protected]> * Update quantavgpool2d.py (#22) Add "Copyright (c) 2021 Xilinx, Inc" heather * [batchnorm_to_affine]: epsilon value is now read out from the attributes. (#21) [test_batchnorm_to_affine]: added a test case for various epsilon values. * Added 3D to 4D (tensor) transformation (#19) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col] added support for dilations = 2 for 2D and 1D images. Dilation value must be equal along each axis. [test_im2col] added several test cases in case dilations = 2 (a.o. cases where image and kernel are 2D and 1D, with and without padding, and with stride = 2). [lower_convs_to_matmul] added support for dilation value. Dilation value must be equal along each axis. [test_conv_lowering] added test case for dilations = 2 for 2D and 1D images and kernels, with and without padding, and with stride = 2. * [lower_convs_to_matmul] removed old assertion [test_conv_lowering] added more dilation values to test cases. Dilation values of {1, 2, 3, 4} are tested. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim [test_conv_lowering] changed function call to compute_conv_output_dim * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: minor fix with assumption on kernel dimension [lower_convs_to_matmul]: minor fix with assumption on kernel dimension * [change_3d_tensors_to_4d]: added new transformation that transforms 3D tensors to 4D and changes the nodes accordingly [test_4d_conversion]: test function for 3D to 4D tensor transformation * [change_3d_tensors_to_4d]: added new transformation that changes 3D tensors to 4D. [test_4d_conversion]: added a test case for the 3D to 4D transformation. * [change_3d_tensors_to_4d]: added 3D to 4D transformation (for QuartzNet). [test_4d_conversion]: added test case for 3D to 4D transform. * [change_3d_tensors_to_4d]: changed how an invalid graph is handled. [test_4d_conversion]: changed the test case for an invalid graph. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [im2col]: minor style change and changed the way how a square kernel is instantiated. * [test_conv_lowering]: merged tests for depthwise and standard convolutions. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * [im2col]: minor style adjustment. [test_conv_lowering]: merged test functions into one test function. * [change_3d_tensors_to_4d]: style fixes and comments. [test_4d_converions]: rearranged code. * [Transform] check invalid node list length * [change_3d_tensors_to_4d]: rearranged the code to make it more readable. Co-authored-by: Yaman Umuroglu <[email protected]> * [Docs] update tutorials * [Util] experimental: fast_mode data packing for binary (#24) * Generic partitioning feature (#23) * Add basic partitioning functionality * Mount build dir within docker container * Support for non-linear models and multi in/out partitions * Remove dataflowpartition custom op from finn-base * Fix temporary build dir for CI * Fix docstring * [create_generic_partitions]: minor modification, removed redundant output value_info entries. (#26) * [extend_partition]: added a new transformation ExtendPartition. (#27) [test_extend_partition]: added a test case for the new transformation. * Added support for non-equal strides along different axes (#25) * [im2col]: added support for non-equal strides along different axes and cleaned up the code. [lower_convs_to_matmul]: added support for non-equal strides along different axes and cleaned up the code. [test_conv_lowering]: added test case for non-equal strides along different axes. * [im2col]: minor fix. [test_im2col]: added test case for non-equal strides along different axes. * Changes for supporting vitis_hls (#28) * [Refactor] split up RTL/HLS-related utils * [Util] rename to CallHLS and allow specifying vivado_hls/vitis_hls * [Util] more flexible stream naming in rtlsim_multi_io * Changes for supporting non-equal dilation (#29) * added support for non-equal dilation value along (H, W) dimension * added test cases for non-equal dilation configurations * appending dilation value along dummy dimension correctly (i.e. with a '1') * changed tensor sparsity annotation for consistency * Support infer_datatype for flatten layer (#30) * Support infer_datatype for flatten layer * [InferDT] add more identity op types for datatype inference * [Lint] fix linting issues Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Create python-publish.yml * Add ZCU111 board to part map (#32) * Update AUTHORS.rst Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]> * Update python-publish.yml * Update python-publish.yml * Lint Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]> Co-authored-by: Felix Jentzsch <[email protected]> * fix broken logo in README * Extend Change3DTo4DTensors() to more layers (#31) * Support infer_datatype for flatten layer * 3d_to_4d support for additional nodes * Bugfix * Add ZCU111 to part map * Fix ZCU111 part * Separate PRs * Bugfix, new test case * [Docker] pin pyverilator version * [Util] pyverilate_stitched_ip: disable_common_warnings optional Co-authored-by: Yaman Umuroglu <[email protected]> Co-authored-by: Mirza Mrahorovic <[email protected]> Co-authored-by: jalezeta <[email protected]> * Copied platforms from finn-experimental and made pre-commit conform (#36) * Copied platforms from finn-experimental and made pre-commit conform * Add ZCU102 and ZCU111 platforms * [Util] add comments on platforms.py Co-authored-by: Felix Jentzsch <[email protected]> Co-authored-by: Yaman Umuroglu <[email protected]> * Various fixes from multi-headed net testing (#45) * [QuantAvgPool2d] support broader range of DataTypes for inference * [pyverilator] use case-agnostic r/w fxns in rtlsim_multi_io * [Rename] support renaming multiple i/o + add test * [InferDT] more identity case, support NN upsample/resize * [InferDT] fix typo * [DataLayout] add more layout variants and MakeInputChanLast trafo * [Test] fix test_infer_datatypes * [ModelWrapper] add assertions for multiple ValueInfoProto multiple VIPs cause undefined behavior when changing the graph * [Core] node input ValueInfos can come from graph output too * [Partition] don't duplicate VIP for border nodes * [Test] increase atol for test_batchnorm_to_affine_shufflenet * chore: Add message to AssertionError (#44) * chore: Add message to AssertionError when file passed to ModelWrapper does not exists (or is not a file) * Various fixes from multi-headed net testing (#45) * [QuantAvgPool2d] support broader range of DataTypes for inference * [pyverilator] use case-agnostic r/w fxns in rtlsim_multi_io * [Rename] support renaming multiple i/o + add test * [InferDT] more identity case, support NN upsample/resize * [InferDT] fix typo * [DataLayout] add more layout variants and MakeInputChanLast trafo * [Test] fix test_infer_datatypes * [ModelWrapper] add assertions for multiple ValueInfoProto multiple VIPs cause undefined behavior when changing the graph * [Core] node input ValueInfos can come from graph output too * [Partition] don't duplicate VIP for border nodes * [Test] increase atol for test_batchnorm_to_affine_shufflenet * Run linter Co-authored-by: Yaman Umuroglu <[email protected]> * [Core] hotfix, relax set_tensor_layout checks otherwise running data layout inference on e.g. models with reshape catches on this assertion * [InferShapes] hotfix for defaulting to unknown data layout * [Partition] hotfix: empty protobuf containers with pop() iterating over the container and calling remove(x) can still leave items inside it * [Core] unify exec paths for GenericPartition+StreamingDFPartition * rtlsim improvements (#47) * bugfix in rtlsim_multi_io * allow extra args in pyverilate_stitched_ip * allow multiple i/o in rtlsim_exec through rtlsim_multi_io also add possibility to pass extra args through metadata_prop * support multiple i/o in throughput_test_rtlsim * fast npy->rtlsim packing when innermost dim=1 * [Core] hotfix for 0-len input nodes in find_* fxns * [Core] hotfix: bug in find_consumer implementation * [Core] hotfix: respect model opset version for node-by-node exec * DataType system refactoring and fixed-point types (#50) * [DataType] sketch new classes for FINN DataTypes * [DataType] flesh out refactored DataType for existing types * [Refactor] DataType.X -> DataType["X"] * [DataType] sketch FixedPointType * [DataType] tweaks to FixedPointType + add is_fixed_point * [Test] add test_datatypes_fixedpoint * [Util] data packing for fixed point types * [Test] data un/packing tests for fixed point * [Util] more fixed-point datatype support * [Test] test fixed-point random tensor gen * [DataType] refactor fxn to get accumulator dt cands * [Refactor] get_smallest_possible and get_dt_cands as staticmethod * [Util] print dt name during sanitize_quant_tensors check * [DataType] fix cand types for get_smallest_possible * [Util, Test] fix unpacking for float32 + add test * [DataType, Test] hotfix: override str/repr fxns for DataType * [Refactor] use RandomNormal instead of Const for shape inf. (#51) * [Test] fix and strenghten test_im2col_infer_shapes * [Refactor] reimpl const shape inference in base CustomOp * QONNX ops and new general transformations (#52) * [Op] add skeleton for Quant custom op * [Op] more comments on Quant op * [InferDT] add MaxPool to list of identity ops for datatype inf. * [Transform] temporarily exclude Quant ops from const folding * [Analysis] add skeleton for inference_cost, limited conv support * [Analysis] support inference_cost_matmul * [Deps] add clize * [Util] add inference_cost util with entrypoint * [Docker] add clize * [Infra] add entrypoint for inference_cost * Preliminary implementation of quant operation * Fixed wrong numpy type in quant op * Made Quant pre-commit compatible * Changed docker base image to PyTorch and added Brevitas dockerfile and entrypoint * Implemented preliminary test for Quant operation * Reworked comments for Quant operation * Readability improvements for Quant operation * Created preliminary Quant op documentation * Move Quant op documentation * Made documentation of quant attributes more precise. * Created skeleton for Trunc QNNX operation * [DataType] add experimental SCALED(U)INT DataTypes * [Quant] use new scaled int dtypes for dt inference * [TypInf] use non-integer instead of float check in dt inference * [ConstFold] make op exclusion optional * [Analysis] deal with dyn matmul, add some zero-cost op types * [Analysis] make inference_cost optionally sparse-weights-aware * [Util] pass sparsity option to analysis pass, kwargs only for clize * [Analysis] add more free ops to inference_cost * [Quant] fallback to float32 for failed dtype inference * Inference cost: JSON export, parameter for output onnx * Removed FINN dependency from documentation * Added support for executing qonnx custom_ops. * Add support for Bipolar and Binary FINN datatype for Quant op. (#41) * Add support for Bipolar and Binary FINN datatype for Quant op. * [Quant] custom qnt execution for bipolar until Brevitas bug resolved Co-authored-by: Yaman Umuroglu <[email protected]> * Added support for cost estimation for upsampling (#37) * Update AUTHORS.rst * Create python-publish.yml * Add ZCU111 board to part map (#32) * Update AUTHORS.rst * GHA for PyPI, sdist only (#35) * finn-base v0.0.2 (#34) * Modified set_nodeattr to allow using it on repeated fields (#18) * [base]: changed how the floats, ints, strings, tensors, graphs and sparse_tensors field of AttributeProto is set. * [Core] restrict attributes to tested types Co-authored-by: Yaman Umuroglu <[email protected]> * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst * Support for non-square input images and kernels in LowerConvsToMatMul transformation (#16) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col] function compute_conv_output_dim can now be called in case non-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim. [test_conv_lowering] changed function call to compute_conv_output_dim. [lower_convs_to_matmul] removed old assertion. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [test_conv_lowering]: minor fix for test case depthwise and regular convolutions * Support for non-square input images and kernels for im2col node (#20) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col]: changed how a square kernel is instantiated. [lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated). * [im2col]: minor change in style. * [Im2Col] style fixes and comments Co-authored-by: Yaman Umuroglu <[email protected]> * Update AUTHORS.rst Co-authored-by: Yaman Umuroglu <[email protected]> * Added support for dilation value = 2 for 1D and 2D images/kernels (#17) * [im2col.py]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [im2col]: support for non-square input images and kernels, [test_im2col]: added/modified several test cases for (non-)square images and kernels * [test_general_transformation]: changed kernel_size attribute to list instead of integer as required by im2col node * [base]: changed how the "ints" field of AttributeProto set. [test_general_transformation]: changed the type of the kernel_size attribute to list of integers * removed unused import * [base]: added support for writing repeated fields in AttributeProto * minor style changes * [im2col, test_im2col]: added support for non-equal padding * [lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution. * [test_conv_lowering]: included 1D depthwise and regular convolutions in tests * Revert "[test_conv_lowering]: included 1D depthwise and regular convolutions in tests" This reverts commit 3ff449c42d709e640ca904c41a241bb94fc9e335 * Revert "[lower_convs_to_matmul]: added support for non-square input images and kernels and non-equal padding." This reverts commit 15e34ed8d07d4a55996f162bd3bd1aa24b33c3ac. * Revert "[im2col, test_im2col]: added support for non-equal padding" This reverts commit c524020ee8a7b363eb0c30d70cf21332e9c73678. * [im2col] added support for dilations = 2 for 2D and 1D images. Dilation value must be equal along each axis. [test_im2col] added several test cases in case dilations = 2 (a.o. cases where image and kernel are 2D and 1D, with and without padding, and with stride = 2). [lower_convs_to_matmul] added support for dilation value. Dilation value must be equal along each axis. [test_conv_lowering] added test case for dilations = 2 for 2D and 1D images and kernels, with and without padding, and with stride = 2. * [lower_…

mmrahorovic added 22 commits December 9, 2020 16:53

[im2col.py]: support for non-square input images and kernels, [test_i…

17d72f5

…m2col]: added/modified several test cases for (non-)square images and kernels

[im2col]: support for non-square input images and kernels, [test_im2c…

08da272

…ol]: added/modified several test cases for (non-)square images and kernels

[test_general_transformation]: changed kernel_size attribute to list …

cd3683d

…instead of integer as required by im2col node

[base]: changed how the "ints" field of AttributeProto set.

56a1bb7

[test_general_transformation]: changed the type of the kernel_size attribute to list of integers

removed unused import

477e135

[base]: added support for writing repeated fields in AttributeProto

878622f

minor style changes

8e2279c

[im2col, test_im2col]: added support for non-equal padding

c524020

[lower_convs_to_matmul]: added support for non-square input images an…

15e34ed

…d kernels and non-equal padding. [test_conv_lowering]: added/modified test cases for non-equal padding, depthwise convolution and 'standard' convolution.

[test_conv_lowering]: included 1D depthwise and regular convolutions …

3ff449c

…in tests

Revert "[test_conv_lowering]: included 1D depthwise and regular convo…

3de861a

…lutions in tests" This reverts commit 3ff449c

Revert "[lower_convs_to_matmul]: added support for non-square input i…

ccf07f0

…mages and kernels and non-equal padding." This reverts commit 15e34ed.

Revert "[im2col, test_im2col]: added support for non-equal padding"

220e644

This reverts commit c524020.

[lower_convs_to_matmul] removed old assertion

bce65cf

[test_conv_lowering] added more dilation values to test cases. Dilation values of {1, 2, 3, 4} are tested.

[im2col] function compute_conv_output_dim can now be called in case n…

f70a71c

…on-equal and equal padding is assumed. [test_im2col] changed function call to compute_conv_output_dim [test_conv_lowering] changed function call to compute_conv_output_dim

[im2col]: minor fix with assumption on kernel dimension

e72ee48

[lower_convs_to_matmul]: minor fix with assumption on kernel dimension

[change_3d_tensors_to_4d]: added new transformation that transforms 3…

872cc78

…D tensors to 4D and changes the nodes accordingly [test_4d_conversion]: test function for 3D to 4D tensor transformation

[change_3d_tensors_to_4d]: added new transformation that changes 3D t…

95caa2e

…ensors to 4D. [test_4d_conversion]: added a test case for the 3D to 4D transformation.

[change_3d_tensors_to_4d]: added 3D to 4D transformation (for QuartzN…

38815b6

…et). [test_4d_conversion]: added test case for 3D to 4D transform.

[change_3d_tensors_to_4d]: changed how an invalid graph is handled.

077ea17

[test_4d_conversion]: changed the test case for an invalid graph.

mmrahorovic marked this pull request as ready for review January 16, 2021 13:41

mmrahorovic added 7 commits January 16, 2021 14:56

[im2col]: changed how a square kernel is instantiated.

0d76430

[lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated).

[im2col]: changed how a square kernel is instantiated.

5f60b4d

[lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated).

[im2col]: changed how a square kernel is instantiated.

8563dc5

[lower_convs_to_matmul]: changed how the kernel size attribute is read (based on how a square kernel is instantiated).

[im2col]: minor change in style.

282ae0d

[im2col]: minor style change and changed the way how a square kernel …

8271f3a

…is instantiated.

[test_conv_lowering]: merged tests for depthwise and standard convolu…

1702b93

…tions.

Merge remote-tracking branch 'upstream/dev' into dev

8221d98

mmrahorovic marked this pull request as draft January 20, 2021 15:24

mmrahorovic added 5 commits January 21, 2021 11:47

Merge branch 'feature/non_square_im2col' into 1d_lower_convs_to_matmul

c95584b

[test_conv_lowering]: minor fix for test case depthwise and regular c…

d0b5f1c

…onvolutions

Merge branch '1d_lower_convs_to_matmul' into im2col_dilation

2de123e

[im2col]: minor style adjustment.

74a43cd

[test_conv_lowering]: merged test functions into one test function.

Merge branch 'im2col_dilation' into feature/3d_to_4d_transformation

a9933df

mmrahorovic marked this pull request as ready for review January 21, 2021 17:17

mmrahorovic and others added 5 commits January 27, 2021 17:32

[change_3d_tensors_to_4d]: style fixes and comments.

134b547

[test_4d_converions]: rearranged code.

Merge branch 'dev' into feature/3d_to_4d_transformation

61b1073

Merge branch 'dev' into feature/3d_to_4d_transformation

571705b

Merge branch 'feature/3d_to_4d_transformation' of https://github.com/…

68c169a

…mmrahorovic/finn-base into feature/3d_to_4d_transformation

[Transform] check invalid node list length

b2ed768

maltanar reviewed Feb 6, 2021

View reviewed changes

mmrahorovic added 2 commits February 18, 2021 14:21

[change_3d_tensors_to_4d]: rearranged the code to make it more readable.

bb403f4

Merge remote-tracking branch 'upstream/dev' into feature/3d_to_4d_tra…

825f360

…nsformation

maltanar merged commit 4f4d028 into Xilinx:dev Feb 21, 2021

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Added 3D to 4D (tensor) transformation #19

Added 3D to 4D (tensor) transformation #19

mmrahorovic commented Jan 15, 2021 •

edited

Loading

maltanar Feb 6, 2021

maltanar Feb 6, 2021

maltanar commented Feb 6, 2021

Added 3D to 4D (tensor) transformation #19

Added 3D to 4D (tensor) transformation #19

Conversation

mmrahorovic commented Jan 15, 2021 • edited Loading

maltanar Feb 6, 2021

Choose a reason for hiding this comment

maltanar Feb 6, 2021

Choose a reason for hiding this comment

maltanar commented Feb 6, 2021

mmrahorovic commented Jan 15, 2021 •

edited

Loading