4717a3b038
Created using Colaboratory
2021-07-02 12:24:27 +02:00
b6863385b5
Created using Colaboratory
2021-07-02 12:23:09 +02:00
c6c88dc601
Copy-Paste augmentation for YOLOv5 ( #3845 )
...
* Copy-paste augmentation initial commit
* if any segments
* Add obscuration rejection
* Add copy_paste hyperparameter
* Update comments
2021-07-01 00:35:04 +02:00
25d1f2932c
Plot `AutoShape()` detections in ascending order ( #3843 )
2021-06-30 15:10:40 +02:00
7d6af69638
Fix `LoadStreams()` dataloader frame skip issue ( #3833 )
...
* Update datasets.py to read every 4th frame of streams
* Update datasets.py
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
2021-06-30 12:11:29 +02:00
57c5d02bbe
Concise `TransformerBlock()` ( #3821 )
2021-06-29 16:03:10 +02:00
5ea771d93d
Move IoU functions to metrics.py ( #3820 )
2021-06-29 13:18:13 +02:00
3213d8713f
Fix for `dataset_stats()` with updated data.yaml ( #3819 )
...
@KalenMike
2021-06-29 12:44:59 +02:00
02719dde52
Update `feature_visualization()` ( #3807 )
...
* Update `feature_visualization()`
Only plot for data with height, width > 1
* cleanup
* Cleanup
2021-06-28 13:48:14 +02:00
20d45aa4f1
Add feature map visualization ( #3804 )
...
* Add feature map visualization
Add a feature_visualization function to visualize the mid feature map of the model.
* Update yolo.py
* remove boolean from forward and reorder if statement
* remove print from forward
* General cleanup
* Indent
* Update plots.py
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
2021-06-28 13:18:45 +02:00
3974d725b6
Fix warmup `accumulate` ( #3722 )
...
* gradient accumulation during warmup in train.py
Context:
`accumulate` is the number of batches/gradients accumulated before calling the next optimizer.step().
During warmup, it is ramped up from 1 to the final value nbs / batch_size.
Although I have not seen this in other libraries, I like the idea. During warmup, as grads are large, too large steps are more of on issue than gradient noise due to small steps.
The bug:
The condition to perform the opt step is wrong
> if ni % accumulate == 0:
This produces irregular step sizes if `accumulate` is not constant. It becomes relevant when batch_size is small and `accumulate` changes many times during warmup.
This demo also shows the proposed solution, to use a ">=" condition instead:
https://colab.research.google.com/drive/1MA2z2eCXYB_BC5UZqgXueqL_y1Tz_XVq?usp=sharing
Further, I propose not to restrict the number of warmup iterations to >= 1000. If the user changes hyp['warmup_epochs'], this causes unexpected behavior. Also, it makes evolution unstable if this parameter was to be optimized.
* replace last_opt_step tracking by do_step(ni)
* add docstrings
* move down nw
* Update train.py
* revert math import move
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
2021-06-28 12:25:13 +02:00
5e976a274d
Fix competition link ( #3799 )
...
* link to the competition repaired
* Update README.md
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
2021-06-27 15:28:50 +02:00
47543f97b1
FROM nvcr.io/nvidia/pytorch:21.05-py3 ( #3794 )
2021-06-26 17:12:43 +02:00
5e6886c860
Update Dockerfile ( #3792 )
2021-06-26 16:51:17 +02:00
96c87f1711
nvcr.io/nvidia/pytorch:21.06-py3 ( #3791 )
2021-06-26 16:28:06 +02:00
07166ba38c
Update Usage examples ( #3790 )
2021-06-26 16:09:56 +02:00
92d49fde35
Update seeds for single-GPU reproducibility ( #3789 )
...
For seed=0 on single-GPU.
2021-06-26 15:42:40 +02:00
8e7f285051
Fix datasets for aws and get_coco.sh ( #3788 )
...
* merge master
* Update get_coco.sh
2021-06-26 15:29:16 +02:00
bd581b330b
Create hyp.scratch-p6.yaml ( #3787 )
2021-06-26 15:13:06 +02:00
8035b61682
Update objectness IoU sort ( #3786 )
2021-06-26 14:52:18 +02:00
157aa2f886
Objectness IoU Sort ( #3610 )
...
Co-authored-by: U-LAPTOP-5N89P8V7\banhu <ban.huang@foxmail.com>
2021-06-26 14:45:53 +02:00
9dc5d35fce
Update README.md fix banner width ( #3785 )
2021-06-26 13:05:46 +02:00
f89941711c
NGA xView 2018 Dataset Auto-Download ( #3775 )
...
* update clip_coords for numpy
* uncomment
* cleanup
* Add autosplits
* fix
* cleanup
2021-06-26 00:49:05 +02:00
ffb6e11050
W&B: Update Tables API and comply with new dataset_check ( #3772 )
...
* Update tables API and windows path fix
* update dataset check
2021-06-25 20:45:33 +02:00
09246a5a33
fix/incorrect_fitness_import ( #3770 )
2021-06-25 16:16:18 +02:00
f2d97ebb25
Remove DDP MultiHeadAttention fix ( #3768 )
2021-06-25 12:52:05 +02:00
374957317a
Add `xyxy2xywhn()` ( #3765 )
...
* Edit Comments for numpy2torch tensor process
Edit Comments for numpy2torch tensor process
* add xyxy2xywhn
add xyxy2xywhn
* add xyxy2xywhn
* formatting
* pass arguments
pass arguments
* edit comment for xyxy2xywhn()
edit comment for xyxy2xywhn()
* cleanup datasets.py
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
2021-06-25 11:47:46 +02:00
03281f8c76
COCO annotations JSON fix ( #3764 )
2021-06-25 01:55:53 +02:00
f79d7479da
Add optional dataset.yaml `path` attribute ( #3753 )
...
* Add optional dataset.yaml `path` attribute
@KalenMike
* pass locals to python scripts
* handle lists
* update coco128.yaml
* Capitalize first letter
* add test key
* finalize GlobalWheat2020.yaml
* finalize objects365.yaml
* finalize SKU-110K.yaml
* finalize SKU-110K.yaml
* finalize VisDrone.yaml
* NoneType fix
* update download comment
* voc to VOC
* update
* update VOC.yaml
* update VOC.yaml
* remove dashes
* delete get_voc.sh
* force coco and coco128 to ../datasets
* Capitalize Argoverse_HD.yaml
* Capitalize Objects365.yaml
* update Argoverse_HD.yaml
* coco segments fix
* VOC single-thread
* update Argoverse_HD.yaml
* update data_dict in test handling
* create root
2021-06-25 01:25:03 +02:00
417a2f425c
Edit comment ( #3759 )
...
edit comment
2021-06-24 15:57:27 +02:00
ae4261c774
Force non-zero hyp evolution weights `w` ( #3748 )
...
Fix for https://github.com/ultralytics/yolov5/issues/3741
2021-06-23 12:56:22 +02:00
fdc22398fa
Create `data/hyps` directory ( #3747 )
2021-06-23 12:49:38 +02:00
63060910a6
Update `check_datasets()` for dynamic unzip path ( #3732 )
...
@KalenMike
2021-06-22 16:05:38 +02:00
9ac7d388a9
Backwards compatible cache version checks ( #3730 )
2021-06-22 13:50:47 +02:00
0e2d0d54d7
Fix typo ( #3729 )
2021-06-22 13:33:38 +02:00
b83e1a4adc
Fix `img2label_paths()` order ( #3720 )
...
* Fix `img2label_paths()` order
* fix, 1
2021-06-21 22:50:56 +02:00
1f69d12591
Update 4 main ops for paths and .run() ( #3715 )
...
* Add yolov5/ to path
* rename functions to run()
* cleanup
* rename fix
* CI fix
* cleanup find models/export.py
2021-06-21 17:25:04 +02:00
75c0ff43af
[x]W&B: Don't resume transfer learning runs ( #3604 )
...
* Allow config cahnge
* Allow val change in wandb config
* Don't resume transfer learning runs
* Add entity in log dataset
2021-06-21 14:00:25 +02:00
e8810a53e8
Update DDP backend `if dist.is_nccl_available()` ( #3705 )
2021-06-20 17:15:42 +02:00
fbf41e0913
Add `train.run()` method ( #3700 )
...
* Update train.py explicit arguments
* Update train.py
* Add run method
2021-06-20 15:06:58 +02:00
c1af67dcd4
Add torch DP warning ( #3698 )
2021-06-19 19:50:46 +02:00
b3e2f4e08d
Eliminate `total_batch_size` variable ( #3697 )
...
* Eliminate `total_batch_size` variable
* cleanup
* Update train.py
2021-06-19 19:14:59 +02:00
fad27c0046
Update DDP for `torch.distributed.run` with `gloo` backend ( #3680 )
...
* Update DDP for `torch.distributed.run`
* Add LOCAL_RANK
* remove opt.local_rank
* backend="gloo|nccl"
* print
* print
* debug
* debug
* os.getenv
* gloo
* gloo
* gloo
* cleanup
* fix getenv
* cleanup
* cleanup destroy
* try nccl
* return opt
* add --local_rank
* add timeout
* add init_method
* gloo
* move destroy
* move destroy
* move print(opt) under if RANK
* destroy only RANK 0
* move destroy inside train()
* restore destroy outside train()
* update print(opt)
* cleanup
* nccl
* gloo with 60 second timeout
* update namespace printing
2021-06-19 16:30:25 +02:00
5bab9a28e4
Reformat ( #3694 )
2021-06-19 14:21:28 +02:00
bfb2276b1d
Slightly modify CLI execution ( #3687 )
...
* Slightly modify CLI execution
This simple change makes it easier to run the primary functions of this
repo (train/detect/test) from within Python. An object which represents
`opt` can be constructed and fed to the `main` function of each of these
modules, rather than having to call the lower level functions directly,
or run the module as a script.
* Update export.py
Add CLI parsing update for more convenient module usage within Python.
Co-authored-by: Lewis Belcher <lb@desupervised.io>
2021-06-19 12:06:59 +02:00
bf209f6fe9
Skip HSV augmentation when hyperparameters are [0, 0, 0] ( #3686 )
...
* Create shortcircuit in augment_hsv when hyperparameter are zero
* implement faster opt-in
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
2021-06-19 11:51:21 +02:00
814806c61d
Update cache check ( #3691 )
...
Swapped order of operations for faster first per f527704cd3 (r52362419)
2021-06-19 11:22:09 +02:00
463628a4d8
Update README.md ( #3684 )
2021-06-18 17:12:42 +02:00
2729761458
EMA changes for pre-model's batch_size ( #3681 )
...
* EMA changes for pre-model's batch_size
* Update train.py
* Update torch_utils.py
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
2021-06-18 16:26:52 +02:00
f527704cd3
Cache v0.3: improved corrupt image/label reporting ( #3676 )
...
* Cache v0.3: improved corrupt image/label reporting
Fix for https://github.com/ultralytics/yolov5/issues/3656#issuecomment-863660899
* cleanup
2021-06-18 10:21:47 +02:00
Glenn Jocher
Feras Oughali
Zigarss
yellowdolphin
batrlatom
Ayush Chaurasia
Piotr Skalski
Yonghye Kwon
fcakyon
lb-desupervised
Mai Thanh Minh
ZouJiu1