Commits · bfc04ac8b8d7ff45695457398d869a6692c4f44d · OpenDAS / torchani

04 Sep, 2018 1 commit
- Support caching AEVs into disk (#88) · bfc04ac8
  Gao, Xiang authored Sep 04, 2018
  
  bfc04ac8
31 Aug, 2018 1 commit
- command line tool to precompute aevs (#83) · 83248cf1
  Gao, Xiang authored Aug 31, 2018
  
  83248cf1
28 Aug, 2018 1 commit
- temporarily disable l2 support for now, planning to reenable it after adaw get... · 0ed9b3f3
  Gao, Xiang authored Aug 28, 2018
```
temporarily disable l2 support for now, planning to reenable it after adaw get merged into pytorch (#80)
```
  0ed9b3f3
27 Aug, 2018 1 commit
- test network architecture and trainer setups (#79) · bfa36346
  Gao, Xiang authored Aug 27, 2018
  
  bfa36346
26 Aug, 2018 1 commit
- NeuroChem trainer loader (#78) · 81c163e7
  Gao, Xiang authored Aug 26, 2018
  
  81c163e7
25 Aug, 2018 1 commit
- Docs improvements (#77) · 9cae6d3f
  Gao, Xiang authored Aug 24, 2018
  
  9cae6d3f
23 Aug, 2018 1 commit
- Improve training related API (#76) · 3cced1e6
  Gao, Xiang authored Aug 23, 2018
  
  3cced1e6
22 Aug, 2018 1 commit
- Some API improvements (#73) · ce21d224
  Gao, Xiang authored Aug 21, 2018
  
  ce21d224
21 Aug, 2018 1 commit
- Refactor codes to put neurochem related codes together (#72) · f146feca
  Gao, Xiang authored Aug 21, 2018
  
  f146feca
17 Aug, 2018 1 commit
- Split batch to avoid performance penalty on padding (#66) · 99c2c3dc
  Gao, Xiang authored Aug 17, 2018
  
  99c2c3dc
13 Aug, 2018 2 commits
- fix padding strip when batching (#65) · 22975fa7
  Gao, Xiang authored Aug 13, 2018
  
  22975fa7
- Modify training related APIs to batch the whole data by padding (#63) · 50fb6cdb
  Gao, Xiang authored Aug 13, 2018
  
  50fb6cdb
11 Aug, 2018 1 commit
- Support batching non isomers in models and aev computer (#62) · b7cab4f1
  Gao, Xiang authored Aug 11, 2018
  
  b7cab4f1
10 Aug, 2018 1 commit
- add TransformedLoss to allow adding non-linearity to computed loss (#61) · b63f6e40
  Gao, Xiang authored Aug 10, 2018
  
  b63f6e40
08 Aug, 2018 1 commit
- fix sort_by_species (#57) · 5bb66915
  Gao, Xiang authored Aug 07, 2018
  
  5bb66915
03 Aug, 2018 2 commits
- improve energy shifter (#52) · 73e447f0
  Gao, Xiang authored Aug 03, 2018
  
  73e447f0
- merge Container and BatchModel to provide richer information in output (#50) · 84439caf
  Gao, Xiang authored Aug 03, 2018
  
  84439caf
02 Aug, 2018 1 commit
- remove explicit device and dtype (#44) · d3ae0788
  Gao, Xiang authored Aug 02, 2018
  
  d3ae0788
01 Aug, 2018 1 commit
- use pytorch's CELU to simplify code (#42) · 0e992fe5
  Gao, Xiang authored Aug 01, 2018
  
  0e992fe5
31 Jul, 2018 7 commits
- completely decouple aev computer and neural networks (#40) · e2e740d0
  Gao, Xiang authored Jul 31, 2018
  
  e2e740d0
- remove force from ANIModel (#39) · ee31b752
  Gao, Xiang authored Jul 31, 2018
  
  ee31b752
- make aev computer accept species as tensor (#38) · 2cb357bd
  Gao, Xiang authored Jul 31, 2018
  
  2cb357bd
- make aev computer take a single tuple as input (#37) · 629bc698
  Gao, Xiang authored Jul 31, 2018
  
  629bc698
- test energy force example (#34) · f6b193bd
  Gao, Xiang authored Jul 30, 2018
  
  f6b193bd
- assert over-fitting in test_ignite (#33) · 45f25ec3
  Gao, Xiang authored Jul 30, 2018
  
  45f25ec3
- working ignite and dataloader with simple examples (#21) · b3744935
  Gao, Xiang authored Jul 30, 2018
  
  b3744935
30 Jul, 2018 3 commits
- add common loss and metrics (#32) · ea718be0
  Gao, Xiang authored Jul 30, 2018
  
  ea718be0
- add DictMetric (#31) · e2bc9f29
  Gao, Xiang authored Jul 30, 2018
  
  e2bc9f29
- allow energy shifter as transformations to dataset (#30) · 10699bf7
  Gao, Xiang authored Jul 30, 2018
  
  10699bf7
29 Jul, 2018 1 commit
- Add ignite API helpers, also temporarily disable JIT and python2 compatibiltiy (#29) · 18e4867d
  Gao, Xiang authored Jul 29, 2018
  
  18e4867d
28 Jul, 2018 1 commit
- helper function to create dataloader, helper module that handle batch (#27) · 965efee2
  Gao, Xiang authored Jul 28, 2018
  
  965efee2
24 Jul, 2018 2 commits
- Delete nn.py and create torchani.models to support rich types of models (#26) · 194e88ff
  Gao, Xiang authored Jul 24, 2018
  
  194e88ff
- rewrite dataloader API to improve performance and reduce code complexity (#18) · 327a9b20
  Gao, Xiang authored Jul 24, 2018
  
  327a9b20
23 Jul, 2018 1 commit
- add unit test for model ensemble (#14) · f6ef4ebb
  Gao, Xiang authored Jul 23, 2018
  
  f6ef4ebb
22 Jul, 2018 2 commits
- add limited python2 compatibility (#13) · ba3036d1
  Gao, Xiang authored Jul 22, 2018
  
  ba3036d1
- add style check to unit test (#12) · 87c4184a
  Gao, Xiang authored Jul 22, 2018
  
  87c4184a
21 Jul, 2018 1 commit
- rewrite unittests to completely remove dependency on NeuroChem (#2) · 83031ce1
  Gao, Xiang authored Jul 21, 2018
```
rewrite unittests to completely remove dependency on NeuroChem
```
  83031ce1