Paddlepaddle avg pool3d

.pre-commit-config.yaml

@@ -7,7 +7,7 @@ repos:
         args:
           - "--preview"
   - repo: https://github.com/PyCQA/autoflake
-    rev: v2.2.0
+    rev: v2.2.1
     hooks:
       - id: autoflake
   - repo: https://github.com/pycqa/flake8

ivy/functional/frontends/paddle/nn/functional/pooling.py

@@ -100,6 +100,46 @@ def avg_pool2d(
     )
 
 
+@to_ivy_arrays_and_back
+@with_unsupported_dtypes({"2.5.1 and below": ("float16", "bfloat16")}, "paddle")
+def avg_pool3d(
+    x,
+    kernel_size,
+    stride=None,
+    padding=0,
+    ceil_mode=False,
+    exclusive=True,
+    divisor_override=None,
+    data_format="NDHWC",
+    name=None,
+):
+    if stride is None:
+        stride = kernel_size
+
+    count_include_pad = not exclusive
+    kernel_size = _broadcast_pooling_helper(kernel_size, "3d", name="kernel_size")
+    padding = _broadcast_pooling_helper(padding, "3d", name="padding")
+
+    # Figure out padding string
+    if all(
+        [pad == ivy.ceil((kernel - 1) / 2) for kernel, pad in zip(kernel_size, padding)]
+    ):
+        padding = "SAME"
+    else:
+        padding = "VALID"
+
+    return ivy.avg_pool3d(
+        x,
+        kernel_size,
+        stride,
+        padding,
+        data_format=data_format,
+        count_include_pad=count_include_pad,
+        ceil_mode=ceil_mode,
+        divisor_override=divisor_override,
+    )
+
+
 @to_ivy_arrays_and_back
 @with_supported_dtypes({"2.5.1 and below": ("float32", "float64")}, "paddle")
 def max_unpool1d(

ivy_tests/test_ivy/test_frontends/test_paddle/test_nn/test_functional/test_pooling.py

@@ -276,6 +276,66 @@ def test_paddle_avg_pool2d(
     )
 
 
+# avg_pool3d
+@handle_frontend_test(
+    fn_tree="paddle.nn.functional.pooling.avg_pool3d",
+    df_x_k_s_p=helpers.arrays_for_pooling(
+        min_dims=4,
+        max_dims=4,
+        min_side=2,
+        max_side=4,
+    ),
+    exclusive=st.booleans(),
+    ceil_mode=st.booleans(),
+    data_format=st.sampled_from(["NDHWC", "NCDHW"]),
+)
+def test_paddle_avg_pool3d(
+    df_x_k_s_p,
+    exclusive,
+    ceil_mode,
+    data_format,
+    *,
+    test_flags,
+    backend_fw,
+    frontend,
+    fn_tree,
+    on_device,
+):
+    input_df, x, kernel, stride, padding = df_x_k_s_p
+
+    if data_format == "NCDHW":
+        x[0] = x[0].reshape(
+            (x[0].shape[0], x[0].shape[4], x[0].shape[1], x[0].shape[2], x[0].shape[3])
+        )
+
+    if len(stride) == 1:
+        stride = (stride[0], stride[0], stride[0])
+
+    if padding == "SAME":
+        padding = test_pooling_functions.calculate_same_padding(
+            kernel, stride, x[0].shape[2:]
+        )
+    else:
+        padding = (0, 0, 0)
+
+    helpers.test_frontend_function(
+        input_dtypes=input_df,
+        test_flags=test_flags,
+        backend_to_test=backend_fw,
+        frontend=frontend,
+        fn_tree=fn_tree,
+        on_device=on_device,
+        x=x[0],
+        kernel_size=kernel,
+        stride=stride,
+        padding=padding,
+        ceil_mode=ceil_mode,
+        exclusive=exclusive,
+        divisor_override=None,
+        data_format=data_format,
+    )
+
+
 # max_unpool1d
 @handle_frontend_test(
     fn_tree="paddle.nn.functional.max_unpool1d",