Removing ignite (#354)

* Remove ignite

* remove neurochem-test

* remove cache-aev

.github/workflows/deploy-docs.yml

@@ -22,7 +22,7 @@ jobs:
     - name: Install dependencies
       run: |
         ci/install_dependencies.sh
-        pip install h5py pytorch-ignite tb-nightly sphinx sphinx_rtd_theme matplotlib pillow sphinx-gallery
+        pip install h5py tb-nightly sphinx sphinx_rtd_theme matplotlib pillow sphinx-gallery
         pip install .
     - name: Download data files
       run: ./download.sh

.github/workflows/docs.yml

@@ -20,7 +20,7 @@ jobs:
     - name: Install dependencies
       run: |
         ci/install_dependencies.sh
-        pip install h5py pytorch-ignite tb-nightly sphinx sphinx_rtd_theme matplotlib pillow sphinx-gallery
+        pip install h5py tb-nightly sphinx sphinx_rtd_theme matplotlib pillow sphinx-gallery
         pip install .
     - name: Download data files
       run: ./download.sh

.github/workflows/tools.yml

@@ -31,9 +31,5 @@ jobs:
       run: pip install h5py
     - name: COMP6 Benchmark
       run: python tools/comp6.py dataset/COMP6/COMP6v1/s66x8
-    - name: Install more dependencies
-      run: pip install pytorch-ignite
     - name: Training Benchmark
       run: python tools/training-benchmark.py dataset/ani1-up_to_gdb4/ani_gdb_s01.h5
-    - name: NeuroChem Test
-      run: python tools/neurochem-test.py dataset/ani1-up_to_gdb4/ani_gdb_s01.h5

.github/workflows/unittest.yml

@@ -12,7 +12,7 @@ jobs:
         python-version: [3.6, 3.7]
         test-filenames: [
           test_aev.py, test_aev_benzene_md.py, test_aev_nist.py, test_aev_tripeptide_md.py,
-          test_data_new.py, test_ignite.py, test_utils.py, test_ase.py, test_energies.py,
+          test_data_new.py, test_utils.py, test_ase.py, test_energies.py,
           test_neurochem.py, test_vibrational.py, test_ensemble.py, test_padding.py,
           test_data.py, test_forces.py, test_structure_optim.py, test_jit_builtin_models.py]
 

docs/api.rst

@@ -77,18 +77,3 @@ TorchANI Optimizater
 
 .. automodule:: torchani.optim
 .. autoclass:: torchani.optim.AdamW
-
-
-Ignite Helpers
-==============
-
-.. automodule:: torchani.ignite
-.. autoclass:: torchani.ignite.Container
-    :members:
-.. autoclass:: torchani.ignite.DictLoss
-.. autoclass:: torchani.ignite.PerAtomDictLoss
-.. autoclass:: torchani.ignite.TransformedLoss
-.. autofunction:: torchani.ignite.MSELoss
-.. autoclass:: torchani.ignite.DictMetric
-.. autofunction:: torchani.ignite.RMSEMetric
-.. autofunction:: torchani.ignite.MaxAEMetric

docs/index.rst

@@ -22,7 +22,6 @@ Welcome to TorchANI's documentation!
     examples/nnp_training
     examples/nnp_training_force
     examples/nnp_training_ignite
-    examples/cache_aev
     examples/neurochem_trainer
 
 .. toctree::

examples/nnp_training.py

@@ -40,8 +40,7 @@
 # .. note::
 #
 #   Besides defining these hyperparameters programmatically,
-#   :mod:`torchani.neurochem` provide tools to read them from file. See also
-#   :ref:`training-example-ignite` for an example of usage.
+#   :mod:`torchani.neurochem` provide tools to read them from file.
 #
 # .. _rHCNO-5.2R_16-3.5A_a4-8.params:
 #   https://github.com/aiqm/torchani/blob/master/torchani/resources/ani-1x_8x/rHCNO-5.2R_16-3.5A_a4-8.params