OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-07 12:21:27 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity
17,646 Commits 8,304 Branches 150 Tags 1.5 GiB
3803d1c901
Commit Graph

5 Commits

Author SHA1 Message Date
Spandan Tiwari
dafee117e8 Removing unused arg f from _model_to_graph(). (#19647) 
Summary:
Input argument `f` in `_model_to_graph()` method in `torch/onnx/utils.py` is unused. This PR removes it. If there's a reason to keep it around, please let me know.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/19647

Reviewed By: dzhulgakov

Differential Revision: D15071720

Pulled By: houseroad

fbshipit-source-id: 59e0dd7a4d5ebd64d0e30f274b3892a4d218c496
 2019-04-26 09:40:52 -07:00
Lara Haidar-Ahmad
9983c24cfc Strip doc_string from exported ONNX models (#18882) 
Summary:
Strip the doc_string by default from the exported ONNX models (this string has the stack trace and information about the local repos and folders, which can be confidential).

The users can still generate the doc_string by specifying add_doc_string=True in torch.onnx.export().
Pull Request resolved: https://github.com/pytorch/pytorch/pull/18882

Differential Revision: D14889684

Pulled By: houseroad

fbshipit-source-id: 26d2c23c8dc3f484544aa854b507ada429adb9b8
 2019-04-18 22:30:00 -07:00
Spandan Tiwari
a64cce326f Add constant folding to ONNX graph during export (Resubmission) (#18698) 
Summary:
Rewritten version of https://github.com/pytorch/pytorch/pull/17771 using graph C++ APIs.

This PR adds the ability to do constant folding on ONNX graphs during PT->ONNX export. This is done mainly to optimize the graph and make it leaner. The two attached snapshots show a multiple-node LSTM model before and after constant folding.
A couple of notes:
1. Constant folding is by default turned off for now. The goal is to turn it on by default once we have validated it through all the tests.
2. Support for folding in nested blocks is not in place, but will be added in the future, if needed.

**Original Model:**
![multiple_lstm_original](https://user-images.githubusercontent.com/23646532/53987630-6ac53980-40d6-11e9-9702-1ccfee124a83.JPG)
**Constant-folded model:**
![multiple_lstm_constant_folded](https://user-images.githubusercontent.com/23646532/53987632-6c8efd00-40d6-11e9-81c5-362c16f68861.JPG)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/18698

Differential Revision: D14889768

Pulled By: houseroad

fbshipit-source-id: b6616b1011de9668f7c4317c880cb8ad4c7b631a
 2019-04-18 00:10:04 -07:00
Lu Fang
feb5d26510 Rename ONNX util test names (#19153) 
Summary:
Rename test cases.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/19153

Reviewed By: zrphercule

Differential Revision: D14890095

Pulled By: houseroad

fbshipit-source-id: 37a787398c88d9cc92b411c2355b43200cf1c4b0
 2019-04-11 11:29:16 -07:00
Lu Fang
ba77eadbca add an utility function to check whether it's in the middle of onnx export or not 
Summary: Pull Request resolved: https://github.com/pytorch/pytorch/pull/19050

Reviewed By: yinghai

Differential Revision: D14849878

Pulled By: houseroad

fbshipit-source-id: a0a4a57f5f9f315ba1334edfccc9284a8099d17f
 2019-04-09 10:07:08 -07:00