mirrors/yolov5
1
0
Fork 0
mirror of https://github.com/ultralytics/yolov5.git synced 2025-06-03 14:49:29 +08:00
Code Issues Projects Releases Wiki Activity
2,019 Commits 10 Branches 10 Tags
Commit Graph

2019 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Glenn Jocher
741cd0eb23
Update tutorial.ipynb (#6771) 2022-02-25 00:12:27 +01:00
Glenn Jocher
c161557563
Default OMP_NUM_THREADS=8 (#6770) 2022-02-25 00:09:14 +01:00
Glenn Jocher
47f265dde2
Update min warmup iterations from 1k to 100 (#6768) 2022-02-24 23:37:25 +01:00
Glenn Jocher
4bab56526b
Update tutorial.ipynb (2 CPUs, 12.7 GB RAM, 42.2/166.8 GB disk) (#6767) 2022-02-24 18:29:59 +01:00
Glenn Jocher
66aaf5184d
Pre-commit table fix (#6744) 2022-02-22 19:35:31 +01:00
Glenn Jocher
3752807c0b
YOLOv5 v6.1 release (#6739) v6.1 2022-02-22 12:35:24 +01:00
Glenn Jocher
2692e67c5f
Updated VOC hyperparameters (#6732) 
* Update hyps

* Update hyp.VOC.yaml

* Update pathlib

* Update hyps

* Update hyps

* Update hyps

* Update hyps
 2022-02-22 12:11:18 +01:00
Glenn Jocher
a936f5f219
Switch default LR scheduler from cos to linear (#6729) 
* Switch default LR scheduler from cos to linear

Based on empirical results of training both ways on all YOLOv5 models.

* linear bug fix
 2022-02-21 14:23:27 +01:00
Jirka Borovec
dbbb57cf0b
GH: add PR template (#6482) 
* GH: add PR template

* Update CONTRIBUTING.md

* Update PULL_REQUEST_TEMPLATE.md

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Update PULL_REQUEST_TEMPLATE.md

* Update PULL_REQUEST_TEMPLATE.md

* Update PULL_REQUEST_TEMPLATE.md

* Update PULL_REQUEST_TEMPLATE.md

Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
 2022-02-21 12:36:43 +01:00
Glenn Jocher
c43f135571
Update nw to max(nd, 1) (#6714) 2022-02-20 21:47:46 +01:00
Glenn Jocher
4de8b24881
Suppress torch AMP-CPU warnings (#6706) 
This is a torch bug, but they seem unable or unwilling to fix it so I'm creating a suppression in YOLOv5. 

Resolves https://github.com/ultralytics/yolov5/issues/6692
 2022-02-19 16:48:33 +01:00
Glenn Jocher
de9c25b35e
Use export_formats() in export.py (#6705) 
* Use `export_formats()` in export.py

* list fix
 2022-02-19 16:08:33 +01:00
Raffaele Galliera
a297efc383
Edge TPU inference fix (#6686) 
* refactor: use edgetpu flag

* fix: remove bitwise and assignation to tflite

* Cleanup and fix tflite

* Cleanup

Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
 2022-02-19 15:10:07 +01:00
Samuel Yvon
0365379016
Fix floating point in number of workers nw (#6701) 
Integer division by a float yields a (rounded) float. This causes
the dataloader to crash when creating a range.
 2022-02-18 21:06:21 +01:00
Glenn Jocher
56697c43ff
Fix CoreML P6 inference (#6700) 
* Fix CoreML P6 inference

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
 2022-02-18 20:38:33 +01:00
Glenn Jocher
3670d91b05
Fix --evolve --bucket gs://... (#6698) 2022-02-18 19:15:37 +01:00
Jiacong Fang
66e5d794c7
Fix TF exports >= 2GB (#6292) 
* Fix exporting saved_model: pb exceeds 2GB

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Replace TF v1.x API with TF v2.x API for saved_model export

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Clean up

* Remove lambda in tf.function()

* Revert "Remove lambda in tf.function()" to be compatible with TF v2.4

This reverts commit 46c7931f11dfdea6ae340c77287c35c30b9e0779.

* Fix for pre-commit.ci

* Cleanup1

* Cleanup2

* Backwards compatibility update

* Update common.py

* Update common.py

* Cleanup3

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
 2022-02-18 16:36:45 +01:00
Glenn Jocher
07221f1591
FROM nvcr.io/nvidia/pytorch:21.10-py3 (#6677) 
Revert to 21.10 on autobuild fail
 2022-02-17 16:05:30 +01:00
Glenn Jocher
18b5081166
FROM nvcr.io/nvidia/pytorch:21.10-py3 (#6673) 
Reordered installation may help reduce resource usage in autobuild
 2022-02-17 15:42:47 +01:00
Glenn Jocher
25456284b4
Update Dockerfile reorder installs (#6672) 
Also `nvidia-tensorboard-plugin-dlprof`, `nvidia-tensorboard` are no longer installed in NVCR base.
 2022-02-17 15:10:19 +01:00
Glenn Jocher
09d263b376
FROM nvcr.io/nvidia/pytorch:21.10-py3 (#6671) 
22.10 returns 'no space left on device' error message.

Seems like a bug at docker. Raised issue in https://github.com/docker/hub-feedback/issues/2209
 2022-02-17 14:51:53 +01:00
Glenn Jocher
bb054d5f55
FROM nvcr.io/nvidia/pytorch:22.01-py3 (#6670) 2022-02-17 14:20:42 +01:00
Glenn Jocher
9a8ebe671c
Update Dockerfile torch==1.10.2+cu113 (#6669) 2022-02-17 13:24:44 +01:00
Glenn Jocher
2e5c67e537
Robust scipy.cluster.vq.kmeans too few points (#6668) 
* Handle `scipy.cluster.vq.kmeans` too few points

Resolves #6664

* Update autoanchor.py

* Cleanup
 2022-02-17 12:55:03 +01:00
Glenn Jocher
7b80545e8e
Add --cache val (#6663) 
New `--cache val` argument will cache validation set only into RAM. Should help multi-GPU training speeds without consuming as much RAM as full `--cache ram`.
 2022-02-17 00:47:42 +01:00
Glenn Jocher
1ff43702a8
Validate with 2x --workers single-GPU/CPU fix (#6659) 
Fix for #6658 for single-GPU and CPU training use cases
 2022-02-16 14:50:37 +01:00
Glenn Jocher
6769021901
Validate with 2x --workers (#6658) 2022-02-16 14:44:44 +01:00
Glenn Jocher
ee6c70ae43
Fix yolov3.yaml remove list (#6655) 
Per https://github.com/ultralytics/yolov3/issues/1887#issuecomment-1041135181
 2022-02-16 10:13:05 +01:00
imyhxy
ca0a00784d
Fixed wandb logger KeyError (#6637) 2022-02-14 17:08:36 +01:00
Glenn Jocher
538930023c
Fix ConfusionMatrix scale vmin=0.0 (#6638) 
Fix attempt for https://github.com/ultralytics/yolov5/issues/6626
 2022-02-14 14:54:55 +01:00
Glenn Jocher
a45e472358
YOLOv5 Export Benchmarks (#6613) 
* Add benchmarks.py

* Update

* Add requirements

* Updates

* Updates

* Updates

* Updates

* Updates

* Updates

* dataset autodownload from root

* Update

* Redirect to /dev/null

* sudo --help

* Cleanup

* Add exports pd df

* Updates

* Updates

* Updates

* Cleanup

* dir handling fix

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Cleanup

* Cleanup2

* Cleanup3

* Cleanup model_type

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
 2022-02-12 16:05:43 +01:00
Ayush Chaurasia
96d8f86085
W&B: don't log media in evolve (#6617) 2022-02-12 13:04:50 +01:00
Ayush Chaurasia
51f7750aad
W&B: Improve resume stability (#6611) 
* log best.pt metrics at train end

* update

* Update __init__.py

* flush callbacks when using evolve

* remember batch size on resuming

* Update train.py

* improve stability of resume

Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
 2022-02-12 13:03:09 +01:00
Glenn Jocher
3eef80457f
Add YOLOv5n to Reproduce section (#6619) 2022-02-11 21:48:01 +01:00
Glenn Jocher
b40bdd6b2d
YouTube dependency fix youtube_dl==2020.12.2 (#6612) 
Per https://github.com/ultralytics/yolov5/issues/5860#issuecomment-1035320018 by @hdnh2006
 2022-02-11 13:46:20 +01:00
or-toledano
cb2ad9f685
Fix ROOT / data when running W&B log_dataset() (#6606) 
* Fix missing data folder when running log_dataset

* Use ROOT/'data'

* PEP8 whitespace
 2022-02-10 16:17:47 +01:00
Glenn Jocher
c21da596f3
Fix hyp_evolve.yaml indexing bug (#6604) 
* Fix `hyp_evolve.yaml` indexing bug

Bug caused hyp_evolve.yaml to display latest generation result rather than best generation result.

* Update plots.py

* Update general.py

* Update general.py

* Update general.py
 2022-02-10 14:51:29 +01:00
Glenn Jocher
a5c9057dcc
Update train.py 2022-02-10 14:15:18 +01:00
Glenn Jocher
364b7c5111
Fix YouTube dislike button bug in pafy package (#6603) 
Per https://github.com/ultralytics/yolov5/issues/6583#issuecomment-1034421945 by @alicera
 2022-02-10 13:13:44 +01:00
Christoph Fürbacher
9cf80b7f60
return opt from train.run() (#6581) 2022-02-09 15:11:52 +01:00
Glenn Jocher
9c513ca629
Add DATASETS_DIR global in general.py (#6578) 2022-02-08 22:20:44 +01:00
Matthias Vadcard
f40854b61b
Allow custom --evolve project names (#6567) 
* Update train.py

As see in #6463, modification on train in evolve process to allow custom save directory.

* fix val

* PEP8

whitespace around operator

* Cleanup

Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
 2022-02-08 16:13:45 +01:00
Glenn Jocher
f627bc53f8
Fix plot_labels() colored histogram bug (#6574) 
* Fix `plot_labels()` colored histogram bug

* Cleanup
 2022-02-08 14:03:50 +01:00
Glenn Jocher
22da63e30f
Fix zero-export handling with if any(f): (#6569) 
* Fix zero-export handling with `if any(f):`

Partial fix for https://github.com/ultralytics/yolov5/issues/6563

* Cleanup
 2022-02-08 12:20:39 +01:00
Glenn Jocher
006eb40ab7
Improved AutoBatch DDP error message (#6568) 
* Improved AutoBatch DDP error message

* Cleanup
 2022-02-08 11:45:30 +01:00
Glenn Jocher
b57abb17f2
Move trainloader functions to class methods (#6559) 
* Move trainloader functions to class methods

* results = ThreadPool(NUM_THREADS).imap(self.load_image, range(n))

* Cleanup
 2022-02-07 18:59:26 +01:00
Glenn Jocher
dc7e093071
Edge TPU TF imports fix (#6542) 
* Edge TPU TF imports fix

Fix for https://github.com/ultralytics/yolov5/issues/6535#issuecomment-1030631526

* Update common.py
 2022-02-05 15:45:44 +01:00
Jirka Borovec
cba4303d32
Fix 6 Flake8 issues (#6541) 
* F541

* F821

* F841

* E741

* E302

* E722

* Apply suggestions from code review

* Update general.py

* Update datasets.py

* Update export.py

* Update plots.py

* Update plots.py

Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
 2022-02-05 15:22:59 +01:00
Jirka Borovec
e1a6a0b100
flake8: code meanings (#6481) 2022-02-04 19:54:34 +01:00
bilzard
aff0281969
Load checkpoint on CPU instead of on GPU (#6516) 
* Load checkpoint on CPU instead of on GPU

* refactor: simplify code

* Cleanup

* Update train.py

Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
 2022-02-04 18:58:18 +01:00