7a92caa560
Add basic image folder style dataset to read directly out of tar files, example in validate.py
5 years ago
d6ac5bbc48
EfficientNet and related cleanup
...
* remove folded_bn support and corresponding untrainable tflite ported weights
* combine bn args into dict
* add inplace support to activations and use where possible for reduced mem on large models
5 years ago
c11973602d
Merge pull request #19 from rwightman/resnet-refactor
...
Resnet refactoring
5 years ago
d97510fd4a
Add new results csv and update README with 3 new ResNet weight results
5 years ago
3d9be78fc6
A bit more ResNet cleanup.
...
* add inplace=True back
* minor comment improvements
* few clarity changes
5 years ago
33436fafad
Add weights for ResNeXt50d model
5 years ago
e78cd79073
Move ResNet additions for Gluon into main ResNet impl. Add ResNet-26 and ResNet-26d models with weights.
5 years ago
c37cb0f2b7
ImageNet-1k vs ImageNet-v2 comparison
5 years ago
6cdf35e670
Add explicit half/fp16 support to loader and validation script
5 years ago
5684c6af32
Update README.md
...
Add blurb about results csv
5 years ago
a6b2f6eca5
Update README, bump version
5 years ago
2a1d09b4c0
Add result csv with summary of all pretrained models
5 years ago
edb425ea48
Add crop_pct arg to validate, extra fields to csv output, 'all' filters pretrained
5 years ago
949b7a81c4
Fix typo in Densenet default resolutions
5 years ago
da52fcf78a
Add NASNet-Large model
5 years ago
f632190fd9
Add .gitattributes
5 years ago
5bc0ed3d7c
Grammar
5 years ago
c89acdfc02
Update EffResNetComparison.ipynb
5 years ago
73642f7462
Adjust notebook path
5 years ago
c30f625215
Update EffResNetComparison
5 years ago
c4916897f8
EfficientNet vs ResNet comparison in Colaboratory
5 years ago
6057496409
Register dpn107
5 years ago
3d1a66b6fc
Version 0.1.6
5 years ago
a6878b5218
Fix DPN config keys that I broke
5 years ago
8a05b8d555
Merge pull request #16 from rwightman/misc-epoch
...
Weights, arguments, epoch counting, and more
5 years ago
9b0070edc9
Add two comments back, fix typo
5 years ago
188aeae8f4
Bump version 0.1.4
5 years ago
c3287aafb3
Slight improvement in EfficientNet-B2 native PyTorch weights
5 years ago
b8762cc67d
Model updates. Add my best ResNet50 weights top-1=78.47. Add some other torchvision weights.
...
* Remove some models that don't exist as pretrained an likely never will (se)resnext152
* Add some torchvision weights as tv_ for models that I have added better weights for
* Add wide resnet recently added to torchvision along with resnext101-32x8d
* Add functionality to model registry to allow filtering on pretrained weight presence
5 years ago
65a634626f
Switch random erasing to doing normal_() on CPU to avoid instability, remove a debug print
5 years ago
c6b32cbe73
A number of tweaks to arguments, epoch handling, config
...
* reorganize train args
* allow resolve_data_config to be used with dict args, not just arparse
* stop incrementing epoch before save, more consistent naming vs csv, etc
* update resume and start epoch handling to match above
* stop auto-incrementing epoch in scheduler
5 years ago
9d653b68a2
Make drop_connect rate scaling match official impl. Fixes #14
5 years ago
13c19e213d
Add native PyTorch EfficientNet B1 and B2 weights. Not quite where I want them, but hitting the brick wall and moving on to other projects...
5 years ago
a0275cfa2f
Fix arg positions in two entrypoint aliases
5 years ago
fe59249701
Bump version to 0.1.2
5 years ago
171c0b88b6
Add model registry and model listing fns, refactor model_factory/create_model fn
5 years ago
8512436436
Add instagram pretrained ResNeXt models from https://pytorch.org/hub/facebookresearch_WSL-Images_resnext/ , update README
5 years ago
1202d053bc
markdown in setup.py
5 years ago
87b92c528e
Some pretrianed URL changes
...
* host some of Cadene's weights on github instead of .fr for speed
* add my old port of ensemble adversarial inception resnet v2
* switch to my TF port of normal inception res v2 and change FC layer back to 'classif' for compat with ens_adv
5 years ago
827a3d6010
Add current checkpoints output back to CheckpointSaver (via logger)
5 years ago
63961b36a2
Missed pnasnet entrypoint
5 years ago
6cc214bd7a
Consistency in model entrypoints
...
* move pretrained entrypoint arg to first pos to be closer to torchvision/hub
* change DPN weight URLS to my github location
5 years ago
b20bb58284
Distributed tweaks
...
* Support PyTorch native DDP as fallback if APEX not present
* Support SyncBN for both APEX and Torch native (if torch >= 1.1)
* EMA model does not appear to need DDP wrapper, no gradients, updated from wrapped model
5 years ago
6fc886acaf
Remove all prints, change most to logging calls, tweak alignment of batch logs, improve setup.py
5 years ago
1d7f2d93a6
Python >= 3.6 requirement
5 years ago
aa4354f466
Big re-org, working towards making pip/module as 'timm'
5 years ago
871f4c1b0c
Change some absolute imports to relative
5 years ago
b9f8d40b10
Fix pretrained override logic for validate, checkpoint always trump pretrained flag during model create
5 years ago
0e1fd11ad8
Merge pull request #12 from rwightman/ema-cleanup
...
Model weights Exponential Moving Average
5 years ago
1019414fd2
Add ported Tensorflow EfficientNet B4/B5 weights
5 years ago
Ross Wightman
Ross Wightman