Commit Graph

1312 Commits (8ee9fd15059e807374f52527951399e61d57b1b0)
 

Author SHA1 Message Date
Glenn Jocher 8ee9fd1505
Update README.md (#3996) 2021-07-13 23:07:09 +02:00
Glenn Jocher d204a61834
Alert (no detections) (#3984)
* `Detections()` class `print()` overload

* Update common.py
2021-07-12 19:48:47 +02:00
Glenn Jocher b3dabdcc38
Update `probability` to `p` (#3980) 2021-07-12 15:54:43 +02:00
Glenn Jocher 41fdf9fa53 Created using Colaboratory 2021-07-12 12:43:26 +02:00
KEN 647223a7a8
`Ensemble()` visualize fix (#3973)
* fix visualize error

* Revert "fix visualize error"

* add visualise profile
2021-07-11 19:47:08 +02:00
Glenn Jocher a544d59f52
Update requirements.txt `albumentations>=1.0.2` (#3972) 2021-07-11 19:07:42 +02:00
Glenn Jocher 90e60b403d
Update `np.random.random()` to `random.random()` (#3967) 2021-07-11 16:56:36 +02:00
Glenn Jocher 8298ce5e88
Update `albumentations>=1.0.2` (#3966) 2021-07-11 15:49:30 +02:00
Glenn Jocher 80299a57e2
Numerical stability fix for Albumentations (#3958) 2021-07-10 19:50:53 +02:00
Glenn Jocher 443af8b25a
Cache v0.4 update (#3954) 2021-07-10 14:18:46 +02:00
Glenn Jocher a26e7de2bf
Update augmentations.py (#3948) 2021-07-09 16:45:04 +02:00
Glenn Jocher 248504cf13
Feature visualization improvements 32 (#3947) 2021-07-09 15:23:02 +02:00
Glenn Jocher dabad5793a Created using Colaboratory 2021-07-08 15:45:53 +02:00
Eldar Kurtic e7888af94c
Fix inconsistent NMS IoU value for COCO (#3934)
Evaluation of 'best' and 'last' models will use the same params as the evaluation during the training phase. 
This PR fixes https://github.com/ultralytics/yolov5/issues/3907
2021-07-08 15:29:02 +02:00
Glenn Jocher 8c6f9e15bf
Update `dataset_stats()` for zipped datasets (#3926)
* Update `dataset_stats()` for zipped datasets

@KalenMike

* cleanup
2021-07-08 11:42:30 +02:00
Glenn Jocher 850970e081
`feature_visualization()` CUDA fix (#3925) 2021-07-07 16:23:31 +02:00
jmiranda-laplateforme 588094eb7a
Update `setattr()` default for Hub PIL images (#3923)
Fix inference from PIL source.
2021-07-07 16:13:12 +02:00
Glenn Jocher 411842e058
Fix `torch.hub.list('ultralytics/yolov5')` pathlib bug (#3921) 2021-07-07 16:08:42 +02:00
Glenn Jocher 87b094bcbc
Feature visualization update (#3920)
* Feature visualization update

* Save to jpg (faster)

* Save to png
2021-07-07 15:41:58 +02:00
johnohagan 61047a2b4f
Save PyTorch Hub models to `/root/hub/cache/dir` (#3904)
* Create hubconf.py

* Add save_dir variable

Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
2021-07-07 13:41:46 +02:00
Glenn Jocher 33202b7f0b
YOLOv5 + Albumentations integration (#3882)
* Albumentations integration

* ToGray p=0.01

* print confirmation

* create instance in dataloader init method

* improved version handling

* transform not defined fix

* assert string update

* create check_version()

* add spaces

* update class comment
2021-07-05 18:01:54 +02:00
Glenn Jocher 6a3ee7cf03
Hub models `map_location=device` (#3894)
* Hub models `map_location=device`

* cleanup
2021-07-05 16:20:46 +02:00
Glenn Jocher 8930e22cce
Evolution commented `hyp['anchors']` fix (#3887)
Fix for `KeyError: 'anchors'` error when start hyperparameter evolution:
```bash
python train.py --evolve
```

```bash
Traceback (most recent call last):
  File "E:\yolov5\train.py", line 623, in <module>
    hyp[k] = max(hyp[k], v[1])  # lower limit
KeyError: 'anchors'
```
2021-07-05 12:48:27 +02:00
Glenn Jocher 3c3f8fbd5d
Improved BGR2RGB speeds (#3880)
* Update BGR2RGB ops

* speed improvements

* cleanup
2021-07-04 20:12:32 +02:00
Glenn Jocher 9e8fb9fd0b
Create `utils/augmentations.py` (#3877)
* Create `utils/augmentations.py`

* cleanup
2021-07-04 18:14:04 +02:00
Glenn Jocher 81b31824f5
Models `*.yaml` reformat (#3875) 2021-07-04 16:55:08 +02:00
Glenn Jocher bd88e7f4f2 Created using Colaboratory 2021-07-04 15:55:23 +02:00
ketan-b 9d86b54eb3
Add multi-stream saving feature (#3864)
* Added the recording feature for multiple streams

Thanks for the very cool repo!!
I was trying to record multiple feeds at the same time, but the current version of the detector only had one video writer and one vid_path!
So the streams were not being saved and only were initialized with one frame and this process didn't record the whole thing.

Fix:
I made a list of `vid_writer` and `vid_path` and the `i` from the loop over the `pred` took care of the writer which need to work!

I hope this helps, Thanks!

* Cleanup list lengths

* batch size variable

* Update datasets.py

Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
2021-07-04 12:55:57 +02:00
san-soucie d3e9d69850
`--evolve 300` generations CLI argument (#3863)
* evolve command accepts argument for number of generations

* evolve generations argument used in evolve for loop

* evolve argument boolean fixes

* default to 300 evolve generations

* Update train.py

Co-authored-by: John San Soucie <jsansoucie@whoi.edu>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
2021-07-04 12:14:35 +02:00
Valentin Aliferov 831773f5a2
Add EXIF rotation to YOLOv5 Hub inference (#3852)
* rotating an image according to its exif tag

* Update common.py

* Update datasets.py

* Update datasets.py

faster

* delete extraneous gpg file

* Update common.py

Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
2021-07-02 13:25:54 +02:00
Glenn Jocher 4717a3b038 Created using Colaboratory 2021-07-02 12:24:27 +02:00
Glenn Jocher b6863385b5 Created using Colaboratory 2021-07-02 12:23:09 +02:00
Glenn Jocher 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
Glenn Jocher 25d1f2932c
Plot `AutoShape()` detections in ascending order (#3843) 2021-06-30 15:10:40 +02:00
Feras Oughali 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
Glenn Jocher 57c5d02bbe
Concise `TransformerBlock()` (#3821) 2021-06-29 16:03:10 +02:00
Glenn Jocher 5ea771d93d
Move IoU functions to metrics.py (#3820) 2021-06-29 13:18:13 +02:00
Glenn Jocher 3213d8713f
Fix for `dataset_stats()` with updated data.yaml (#3819)
@KalenMike
2021-06-29 12:44:59 +02:00
Glenn Jocher 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
Zigarss 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
yellowdolphin 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
batrlatom 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
Glenn Jocher 47543f97b1
FROM nvcr.io/nvidia/pytorch:21.05-py3 (#3794) 2021-06-26 17:12:43 +02:00
Glenn Jocher 5e6886c860
Update Dockerfile (#3792) 2021-06-26 16:51:17 +02:00
Glenn Jocher 96c87f1711
nvcr.io/nvidia/pytorch:21.06-py3 (#3791) 2021-06-26 16:28:06 +02:00
Glenn Jocher 07166ba38c
Update Usage examples (#3790) 2021-06-26 16:09:56 +02:00
Glenn Jocher 92d49fde35
Update seeds for single-GPU reproducibility (#3789)
For seed=0 on single-GPU.
2021-06-26 15:42:40 +02:00
Glenn Jocher 8e7f285051
Fix datasets for aws and get_coco.sh (#3788)
* merge master

* Update get_coco.sh
2021-06-26 15:29:16 +02:00
Glenn Jocher bd581b330b
Create hyp.scratch-p6.yaml (#3787) 2021-06-26 15:13:06 +02:00
Glenn Jocher 8035b61682
Update objectness IoU sort (#3786) 2021-06-26 14:52:18 +02:00