Migrations
/
pytorch-image-models
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
1291 Commits
41 Branches
41 Tags
28 MiB
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from '4c8bb295ab'
${ noResults }
Commit Graph

585 Commits (4c8bb295abce63d5513d7933f3d65376d9120f7e)

Author SHA1 Message Date
Ross Wightman 0012bf7fb5 Merge branch 'norm_norm_norm' into bits_and_tpu
4 years ago
Ross Wightman 8a93ce6ee3 Fix regnetv/w tests, refactor regnet generator code a bit
4 years ago
Ross Wightman cbc4f33220 Merge branch 'norm_norm_norm' into bits_and_tpu
4 years ago
Ross Wightman 4dec8c8087 Fix skip path regression for updated EfficientNet and RegNet def. Add Pre-Act RegNet support (experimental). Remove BN-TF flag. Add efficientnet_b0_g8_gn model.
4 years ago
Ross Wightman a52a614475 Remove layer experiment which should not have been added
4 years ago
Ross Wightman d829858550 Significant norm update
4 years ago
Ross Wightman ab49d275de Significant norm update
4 years ago
Ross Wightman 57fca2b5b2 Fix c16_evos stem / first conv setup
4 years ago
Ross Wightman 1f54a1fff7 Add C16 and E8 EvoNormS0 configs for RegNetZ BYOB nets
4 years ago
Ross Wightman 4d7a5544f7 Remove inplace sigmoid for consistency with other impl
4 years ago
Ross Wightman 88a5b54802 A few small evonorm tweaks for convergence comparisons
4 years ago
Ross Wightman 66daee4f31 Last change wasn't complete, missed adding full evo_norm changeset
4 years ago
Ross Wightman 7bbbd5ef1b EvoNorm and GroupNormAct options for debugging TPU / XLA concerns
4 years ago
Ross Wightman d04f2f1377 Update drop_path and drop_block (fast impl) to be symbolically traceable, slightly faster
4 years ago
Ross Wightman 834a9ec721 Disable use of timm nn.Linear wrapper since AMP autocast + torchscript use appears fixed
4 years ago
Ross Wightman 78912b6375 Updated EvoNorm implementations with some experimentation. Add FilterResponseNorm. Updated RegnetZ and ResNetV2 model defs for trials.
4 years ago
Ross Wightman 3dc71695bf
Merge pull request #989 from martinsbruveris/feat/resmlp-dino
4 years ago
Ross Wightman 480c676ffa Fix FX breaking assert in evonorm
4 years ago
Martins Bruveris 85c5ff26d7 Added DINO pretrained ResMLP models.
4 years ago
Ross Wightman d633a014e6 Post merge cleanup. Fix potential security issue passing kwargs directly through to serialized web data.
4 years ago
Nathan Raw b18c9e323b
Update helpers.py
4 years ago
Nathan Raw 308d0b9554
Merge branch 'master' into hf-save-and-push
4 years ago
Ross Wightman 1f53db2ece Updated lamhalobotnet weights, 81.5 top-1
4 years ago
Ross Wightman 15ef108eb4 Add better halo2botnet50ts weights, 82 top-1 @ 256
4 years ago
Ross Wightman 734b2244fe Add RegNetZ-D8 (83.5 @ 256, 84 @ 320) and RegNetZ-E8 (84.5 @ 256, 85 @ 320) weights. Update names of existing RegZ models to include group size.
4 years ago
Ross Wightman 93cc08fdc5 Make evonorm variables 1d to match other PyTorch norm layers, will break weight compat for any existing use (likely minimal, easy to fix).
4 years ago
Ross Wightman af607b75cc Prep a set of ResNetV2 models with GroupNorm, EvoNormB0, EvoNormS0 for BN free model experiments on TPU and IPU
4 years ago
Ross Wightman c976a410d9 Add ResNet-50 w/ GN (resnet50_gn) and SEBotNet-33-TS (sebotnet33ts_256) model defs and weights. Update halonet50ts weights w/ slightly better variant in1k val, more robust to test sets.
4 years ago
Ross Wightman f2006b2437 Cleanup qkv_bias cat in beit model so it can be traced
4 years ago
Ross Wightman 1076a65df1 Minor post FX merge cleanup
4 years ago
Ross Wightman 32c9937dec Merge branch 'fx-feature-extract-new' of https://github.com/alexander-soare/pytorch-image-models into alexander-soare-fx-feature-extract-new
4 years ago
Alexander Soare 65d827c7a6 rename notrace registration and standardize trace_utils imports
4 years ago
Ross Wightman 9b2daf2a35 Add ResNeXt-50 weights 81.1 top-1 @ 224, 82 @ 288 with A1 'high aug' recipe
4 years ago
Martins Bruveris 5220711d87 Added B/8 models to ViT.
4 years ago
Alexander Soare 0262a0e8e1 fx ready for review
4 years ago
Alexander Soare d2994016e9 Add try/except guards
4 years ago
Alexander Soare b25ff96768 wip - pre-rebase
4 years ago
Alexander Soare e051dce354 Make all models FX traceable
4 years ago
Alexander Soare cf4561ca72 Add FX based FeatureGraphNet capability
4 years ago
Alexander Soare 0149ec30d7 wip - attempting to rebase
4 years ago
Alexander Soare 02c3a75a45 wip - make it possible to use fx graph in train and eval mode
4 years ago
Alexander Soare bc3d4eb403 wip -rebase
4 years ago
Alexander Soare ab3ac3f25b Add FX based FeatureGraphNet capability
4 years ago
Ross Wightman ddc29da974 Add ResNet101 and ResNet152 weights from higher aug RSB recipes. 81.93 and 82.82 top-1 at 224x224.
4 years ago
Ross Wightman b328e56f49 Update eca_halonext26ts weights to a better set
4 years ago
Ross Wightman 2ddef942b9 Better fix for #954 that doesn't break torchscript, pull torch._assert into timm namespace when it exists
4 years ago
Ross Wightman 4f0f9cb348 Fix #954 by bringing traceable _assert into timm to allow compat w/ PyTorch < 1.8
4 years ago
Ross Wightman ae72d009fa Add weights for lambda_resnet50ts, halo2botnet50ts, lamhalobotnet50ts, updated halonet50ts
4 years ago
Ross Wightman b745d30a3e Fix formatting of last commit
4 years ago
Ross Wightman 3478f1d7f1 Traceability fix for vit models for some experiments
4 years ago