Glenn Jocher
23a7f29e2d
Update hyp.finetune.yaml on 306 VOC generations result
4年前
Glenn Jocher
0a0675da12
Merge remote-tracking branch 'origin/master'
4年前
Glenn Jocher
354a97e4ef
Create sotabench.py
4年前
NanoCode012
d8274d0434
Fix results_file not renaming ( #903 )
4年前
Glenn Jocher
281d78c105
Update train.py ( #902 )
* Update train.py with simplified ckpt names
* Return default hyps to hyp.scratch.yaml
Leave line commented for future use once mystery of best finetuning hyps to apply becomes clearer.
* Force test_batch*_pred.jpg replot on final epoch
This will allow you to see predictions final testing run after training completes in runs/exp0
4年前
Glenn Jocher
d68179953d
Update export.py
4年前
Naman Gupta
6f3db5e662
Remove autoanchor and class checks on resumed training ( #889 )
* Class frequency not calculated on resuming training
Calculation of class frequency is not needed when resuming training.
Anchors can still be recalculated whether resuming or not.
* Check rank for autoanchor
* Update train.py
no autoanchor checks on resume
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
4年前
tkianai
c5969f79f5
add support for DWConv fuse operation ( #898 )
* the value should be a list to unpacking
* filter the None labels
* update readme to remove apex dependency
* add support for DWConv fuse operation
4年前
Glenn Jocher
d3f9bf2bb7
Update datasets.py
4年前
NanoCode012
901243c780
Fix relative path for VOC download ( #880 )
Cleaned up code
4年前
Glenn Jocher
3e04d20c7d
Merge remote-tracking branch 'origin/master'
4年前
Glenn Jocher
f06e2d518c
opt.image_weights bug fix ( #885 )
4年前
NanoCode012
1e15aad6f9
Add InfiniteDataLoader class ( #876 )
* Add InfiniteDataLoader
Only initializes at first epoch. Saves time.
* Moved class to a better location
4年前
Glenn Jocher
69ff781ca5
opt.img_weights bug fix ( #885 )
4年前
Glenn Jocher
987c226849
seaborn plotting correlogram addition
4年前
Glenn Jocher
08e97a2f88
Update hyperparameters to add lrf, anchors
4年前
rafale77
9776e70988
torch.ops.torchvision.nms ( #860 )
Don't load the entire torchvision library just for nms when the function is already in the torch library.
4年前
Glenn Jocher
0c01afc390
Update Dockerfile to upgrade pip
4年前
Glenn Jocher
62e037419f
Update Dockerfile FROM nvcr.io/nvidia/pytorch:20.08-py3
4年前
Glenn Jocher
31d6c389f2
Update general.py plot_evolution()
4年前
Glenn Jocher
1e7674d4dd
Update export.py --img-size comment
4年前
Glenn Jocher
a21bd0687c
Update train.py forward simplification
4年前
Willie Maddox
455f7b8f76
Update general.py ( #823 )
Fixes #822
`init_seeds` from `torch_utils` import is being overwritten by function `init_seeds` in `general.py`
4年前
Glenn Jocher
5a7d79fbe6
Update general.py bbox_iou() with eps improvements ( #736 )
eps reduced to 1e-9 for additional precaution.
4年前
Glenn Jocher
82ed33a052
generalize webcam detection with opt.source.isnumeric() (yolov3 #1462 )
4年前
Glenn Jocher
09402a2174
torch.from_tensor() bug fix
4年前
Glenn Jocher
4fb8cb353f
Update export.py with v3.0 Hardswish() support ( #831 )
4年前
Glenn Jocher
4d7f222f73
Update export.py with v3.0 Hardswish() support
4年前
Glenn Jocher
8666bc507e
Create evolve.sh for infinite hyp evolution
4年前
Glenn Jocher
a8751e50de
model.fuse() fix for export.py ( #827 )
4年前
Glenn Jocher
5f07782f56
Update requirements.txt for coremltools kmeans quantization
4年前
Glenn Jocher
71209a6099
exportable Hardswish() implementation
4年前
Glenn Jocher
fd71fe8451
Update hyp.finetune.yaml
4年前
Glenn Jocher
f7e075fe93
gsutil cp yaml when evolving to GCP bucket
4年前
Glenn Jocher
9dbaa1bbbe
Update plot_results()
4年前
Glenn Jocher
83dc540b1d
remove ema.ema hasattr(ema, 'module') check
4年前
Glenn Jocher
1309400e6b
Update hyp.finetune.yaml on VOC results
4年前
Glenn Jocher
fb05681a04
Update requirements.txt to coremltools==4.0b3
4年前
Glenn Jocher
d0f98c0095
Update yolo.py with model_info(verbose=False)
4年前
Glenn Jocher
b9851222bf
Merge remote-tracking branch 'origin/master'
4年前
Glenn Jocher
c171e45cda
Update test.py with set() ( #806 )
4年前
Glenn Jocher
ea5e97fb9b
FP16 NMS ( #814 )
4年前
Glenn Jocher
b42e8a531b
optimize imports
4年前
Glenn Jocher
4447f4b937
--resume to same runs/exp directory ( #765 )
* initial commit
* add weight backup dir on resume
4年前
Glenn Jocher
5e0b90de8f
CIoU nan bug fix ( #736 )
4年前
NanoCode012
fb4fc8cd02
Fix ema attribute error in DDP mode ( #775 )
* Fix ema error in DDP mode
* Update train.py
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
4年前
Glenn Jocher
da35727c68
--save-txt bug fix ( #763 )
4年前
Glenn Jocher
ebafd1ead5
single command --resume ( #756 )
* single command --resume
* else check files, remove TODO
* argparse.Namespace()
* tensorboard lr
* bug fix in get_latest_run()
4年前
Francisco Ingham
26c3b11fdc
Update torch_utils.py ( #751 )
4年前
Glenn Jocher
2077d78880
update Dockerfile with libgl1-mesa-glx
4年前