Commit Graph

Select branches

Hide Pull Requests

G0_raw_decode_output

G0_yolov5s6_raw

G10_decoupled_head_light_FGEM_safe

G11_cls_head_enhance_Lightweight

G12_cls_head_enhance_and_ECA

G13_decoupled_head_ECA

G1_backbone

G2_neck

G2_neck_2

G3_head

G4_cls_loss_gain

G5_focal_loss

G6_decoupled_head

G7_RawHead_add_ECA

G8_decoupled_head_light

G9_decoupled_head_light_FGEM

M0_yolov5_raw

master

Mono Color

• 52d3a9aee1 Fix detect.py --update list bug (#8783) Glenn Jocher 2022-07-29 17:07:24 +02:00
• e309a85586 Add val.py no label warning (#8782) Glenn Jocher 2022-07-29 14:45:29 +02:00
• 567397d67a Fix confusion matrix update when no predictions are made (#8748) jbutle55andGlenn Jocher 2022-07-29 06:06:23 -06:00
• 587a3a37c5 Dynamic batch size support for TensorRT (#8526) 4 people 2022-07-29 06:51:16 -05:00
• 3e858633b2 Revert protobuf<=3.20.1 (#8742) Glenn Jocher 2022-07-27 17:27:44 +02:00
• 0b5ac224ae fix: broken `is_docker` check (#8711) 3 people 2022-07-26 18:02:44 +01:00
• c775a296a7 Revert "Fix BGR->RGB Bug in albumentations #8641" (#8727) Glenn Jocher 2022-07-26 19:00:48 +02:00
• d5116bbe9c coremltools>=5.2 for CoreML export (#8725) Glenn Jocher 2022-07-26 17:50:49 +02:00
• 2e1291fdce Fix BGR->RGB Bug in albumentations #8641 (#8695) 3 people 2022-07-26 13:52:56 +02:00
• b367860196 New HUBDatasetStats() class (#8716) Glenn Jocherandpre-commit-ci[bot] 2022-07-25 18:20:01 +02:00
• a6f197ae79 Update dataloaders.py (#8714) Glenn Jocher 2022-07-25 16:52:28 +02:00
• 7215a0fb41 Avoid cv2 window init code on Windows (#8712) Glenn Jocher 2022-07-25 13:57:05 +02:00
• 0ab303b044 Remove else: from load_image() (#8692) Glenn Jocher 2022-07-25 00:02:09 +02:00
• 916bdb1d61 Faster crop saving (#8696) Glenn Jocher 2022-07-23 23:30:30 +02:00
• b510957650 Move .dockerignore to root (#8690) Glenn Jocher 2022-07-23 18:50:19 +02:00
• 7f7bd6fbcd Set torch.cuda.manual_seed_all() for DDP (#8688) Glenn Jocher 2022-07-23 18:24:14 +02:00
• 1c5e92aba1 Add generator and worker seed (#8602) 3 people 2022-07-23 01:25:17 +02:00
• b92430a83b Update hubconf.py to reset logging level to INFO (#8680) Glenn Jocher 2022-07-22 19:01:16 +02:00
• b17629e54f Update warning emojis (#8678) Glenn Jocher 2022-07-22 15:23:22 +02:00
• 38721de7ef Update hubconf.py to reset LOGGER.level after load (#8674) Glenn Jocher 2022-07-22 11:54:31 +02:00
• 4c1784bd15 Use contextlib's suppress method to silence an error (#8668) Glenn Jocher 2022-07-21 23:12:49 +02:00
• 602d7ffb0e Check TensorBoard logger before adding graph (#8664) Rüdiger Busche 2022-07-21 17:40:53 +02:00
• 92e47b85d9 Upgrade onnxsim to v0.4.1 (#8632) daquexianandGlenn Jocher 2022-07-20 01:01:24 +08:00
• fbe67e4653 Fix OMP_NUM_THREADS=1 for macOS (#8624) Glenn Jocher 2022-07-18 17:53:30 +02:00
• 9cf5fd5ac3 assert torch!=1.12.0 for DDP training (#8621) Glenn Jocherandpre-commit-ci[bot] 2022-07-18 15:05:58 +02:00
• 51fb467b63 Refactor optimizer initialization (#8607) Glenn Jocherandpre-commit-ci[bot] 2022-07-17 11:43:52 +02:00
• 24305787ae FROM nvcr.io/nvidia/pytorch:22.06-py3 (#8600) Glenn Jocher 2022-07-17 00:55:55 +02:00
• a34b376d0f Link fuse() to AutoShape() for Hub models (#8599) Glenn Jocher 2022-07-16 23:46:23 +02:00
• 6e86af3de8 Add pip install wheel to avoid legacy setup.py install (#8597) Glenn Jocher 2022-07-16 16:41:43 +02:00
• 5c45a4b13d Update tqdm>=4.64.0 and thop>=0.1.1 (#8596) Glenn Jocher 2022-07-16 16:14:16 +02:00
• cf28dda366 Update scipy>=1.7.3 (#8595) Glenn Jocher 2022-07-16 15:54:34 +02:00
• 7204c1ca25 Explicitly set weight_decay value (#8592) 3 people 2022-07-16 22:51:48 +09:00
• 72a81e7a1c Update requirements.txt comment spacing (#8562) Glenn Jocher 2022-07-15 16:01:29 +02:00
• f4b05680f8 Assert --optimize not used with cuda device (#8569) Colin Wong 2022-07-15 09:01:01 -05:00
• f8722b4429 Raise error on suffix-less model path (#8561) Colin Wong 2022-07-13 04:13:01 -05:00
• 574ceedfc5 Add thop>=0.1.0 (#8558) Glenn Jocher 2022-07-12 14:49:54 +02:00
• fbd3020525 Add TensorRT dependencies (#8553) 曾逸夫（Zeng Yifu） 2022-07-12 19:19:25 +08:00
• 04146371b9 Update metrics.py with IoU protected divides (#8550) Glenn Jocher 2022-07-12 01:58:25 +02:00
• a84cd02387 CIoU protected divides (#8546) Glenn Jocher 2022-07-11 16:07:11 +02:00
• 7dafd1cb29 val.py assert ncm == nc fix (#8545) Glenn Jocher 2022-07-11 15:09:42 +02:00
• 526e650553 Fix LoadImages() with dataset YAML lists (#8517) 3 people 2022-07-08 07:32:40 -05:00
• c215e87393 XML export --half fix (#8522) Glenn Jocher 2022-07-08 13:49:20 +02:00
• 63ba0cb18a Add --half arguments to export.py Usage examples (#8516) Glenn Jocher 2022-07-08 00:46:56 +02:00
• be42a24d23 Properly expose batch_size from OpenVINO similarly to TensorRT (#8514) Colin WongandGlenn Jocher 2022-07-07 16:53:09 -05:00
• f17444abcd Simplify benchmarks.py assertions (#8515) Glenn Jocher 2022-07-07 23:46:55 +02:00
• 3e54651fca Add --hard-fail argument to benchmarks for CI errors (#8513) Glenn Jocherandpre-commit-ci[bot] 2022-07-07 23:41:34 +02:00
• 39d7a93619 Fix AP calculation bug #8464 (#8484) UnglvKitDeandGlenn Jocher 2022-07-07 20:42:09 +02:00
• dd28df98c2 Avoid FP64 ops for MPS support in train.py (#8511) Glenn Jocher 2022-07-07 20:36:23 +02:00
• 9d7bc06ae7 Revert "Expose OpenVINO batch_size similarly to TensorRT" (#8510) Glenn Jocher 2022-07-07 20:13:42 +02:00
• 27d831b6e4 Training reproducibility improvements (#8213) 3 people 2022-07-07 18:09:29 +05:30
• 36f64a981d Update tutorial.ipynb (#8507) Glenn Jocher 2022-07-07 12:34:01 +02:00
• 1ab23fc67f Exclude torch==1.12.0, torchvision==0.13.0 (Fix #8395) (#8497) Junya Morioka 2022-07-07 02:32:58 +09:00
• fdc9d9198e [pre-commit.ci] pre-commit suggestions (#8470) pre-commit-ci[bot] 2022-07-04 22:09:24 +02:00
• c7689198bc Update stale.yml (#8465) Glenn Jocher 2022-07-04 15:01:11 +02:00
• 29d79a6360 Do not prefer Apple MPS (#8446) Glenn Jocher 2022-07-02 18:35:45 +02:00
• da2ee3934e Expose OpenVINO batch_size similarly to TensorRT (#8437) Colin Wong 2022-07-01 15:15:09 -05:00
• d94b4705a6 TRT --half fix autocast images to FP16 (#8435) Glenn Jocher 2022-07-01 15:41:14 +02:00
• 898332433a Update cache comments (#8414) Glenn Jocher 2022-06-30 16:19:22 +02:00
• e50dc38d36 Improve --local_rank arg comment (#8409) 3 people 2022-06-30 17:31:31 +04:30
• 6935a54e60 Implementation of Early Stopping for DDP training (#8345) 3 people 2022-06-29 12:41:46 +02:00
• f76a78e707 Update tutorial.ipynb (#8380) Glenn Jocher 2022-06-28 17:34:24 +02:00
• 0c1324067c Fix ONNX --dynamic export on GPU (#8378) Glenn Jocher 2022-06-28 15:22:15 +02:00
• 50ff6eee31 Update protobuf requirement from <=3.20.1 to <4.21.3 (#8346) dependabot[bot]andGlenn Jocher 2022-06-28 12:04:58 +02:00
• 34df5032a7 Add File Size (MB) column to benchmarks (#8359) Glenn Jocherandpre-commit-ci[bot] 2022-06-27 17:46:49 +02:00
• 8ebf569d14 Fix bias warmup LR init (#8356) Glenn Jocher 2022-06-27 16:11:24 +02:00
• b0814c9521 Revamp Chinese docs (#8350) Zhiqiang Wang 2022-06-27 19:10:30 +08:00
• 0537e8dd13 Allow detect.py to use video size for initial window size (#8330) Nicholas ZoltonandGlenn Jocher 2022-06-26 17:04:11 -05:00
• 5c990e3674 Create README_cn.md (#8344) Glenn Jocherandpre-commit-ci[bot] 2022-06-26 17:46:39 +02:00
• 19f33cbae2 Allow preview resize in detect.py (#8318) Ryan HirasakiandGlenn Jocher 2022-06-24 13:30:44 +00:00
• 724d5b21b7 device_name="MYRIAD" for Intel NCS2 comment (#8327) Glenn Jocher 2022-06-24 14:18:58 +02:00
• a2a1ed201d logging.ERROR on RANK not in (0, 1) (#8284) Glenn Jocher 2022-06-21 17:41:14 +02:00
• 1156a323f8 Created using Colaboratory (#8255) Glenn Jocherandpre-commit-ci[bot] 2022-06-18 14:29:32 +02:00
• 669f707d62 process_batch() as numpy arrays (#8254) Glenn Jocher 2022-06-18 13:54:55 +02:00
• d6051382f1 HUB dataset_stats() error reporting (#8192) Glenn Jocherandpre-commit-ci[bot] 2022-06-17 18:37:08 +02:00
• 303f3785ae Prefer MPS over CPU if available (#8210) Glenn Jocher 2022-06-17 18:28:29 +02:00
• 86e6add07b Remove streaming source sleep period (#8243) Glenn Jocher 2022-06-17 17:59:27 +02:00
• 75bbaa8f44 Reject webcam inference on Colab/Kaggle (#8209) Glenn Jocherandpre-commit-ci[bot] 2022-06-14 14:28:21 +02:00
• 6a67594cbf Bump actions/setup-python from 3 to 4 (#8186) dependabot[bot] 2022-06-13 12:16:55 +02:00
• c23a441c9d Improved dataset_stats() YAML checks (#8125) Glenn Jocherandpre-commit-ci[bot] 2022-06-11 19:30:54 +02:00
• 6adc53ba5f Docker GPU pip install opencv-python<4.6.0.66 (#8164) Glenn Jocher 2022-06-09 22:45:49 +02:00
• 7c2691f9dc FROM nvcr.io/nvidia/pytorch:22.05-py3 (#8162) Glenn Jocher 2022-06-09 22:06:33 +02:00
• 6e4661773e AutoBatch checks against failed solutions (#8159) Glenn Jocher 2022-06-09 17:15:49 +02:00
• cf298fb72f fix(utils): missing edge when resample segments (#8092) HRan2004andGlenn Jocher 2022-06-09 16:52:05 +08:00
• 2dd3db0050 [pre-commit.ci] pre-commit suggestions (#8119) pre-commit-ci[bot]andGlenn Jocher 2022-06-07 00:04:11 +02:00
• 18674e2f03 experimental.py Apple MPS device fix (#8121) Glenn Jocher 2022-06-06 23:58:50 +02:00
• 47233e1698 Improved dataset error introspection (#8091) Glenn Jocher 2022-06-03 12:45:51 +02:00
• 7cef03dddd Fix torch long to float tensor on HUB macOS (#8067) Glenn Jocher 2022-06-01 17:34:46 +02:00
• a80dd66efe Update stale.yml to 300 ops per run (#8061) Glenn Jocher 2022-05-31 20:18:40 +02:00
• 7d87b9eae8 Make Docker actions conditional on ultralytics/yolov5 repo (#8060) Glenn Jocher 2022-05-31 16:36:57 +02:00
• 6dd6aea086 Fix FP32 TensorRT model export (#8046) Sahil ChachraandGlenn Jocher 2022-05-31 16:11:09 +05:30
• 487edd5609 Dataset autodownload fstring updates Glenn Jocher 2022-05-31 12:36:20 +02:00
• 258f4c6438 attempt_load() deserialize fix (#8051) Glenn Jocher 2022-05-31 10:55:14 +02:00
• 50b886ddc6 .detach() on bias init (#8044) tcmyxc 2022-05-31 15:53:36 +08:00
• dcf8073881 Simplify and fix --save-period epoch 0 (#8042) Glenn Jocher 2022-05-30 12:42:12 +02:00
• 632559b1e6 Add Dockerfile descriptions (#8031) Glenn Jocher 2022-05-29 20:43:01 +02:00
• d490bdd7d6 Add best.pt PyTorch Hub inference to CI (#8024) Glenn Jocher 2022-05-28 22:55:03 +02:00
• 7433d3829a Add requirements.txt protobuf<=3.20.1 (#8017) Glenn Jocher 2022-05-28 12:12:27 +02:00
• 5580b27daf check_requirements() "" Windows double quote (#8016) Glenn Jocher 2022-05-28 11:13:44 +02:00
• 07e84370ac Pin CI protobuf==3.20.1 (#8015) Glenn Jocher 2022-05-28 10:53:44 +02:00
• bf1a6dda7b Update CI benchmarks to ONNX autoinstall (#8008) Glenn Jocher 2022-05-27 16:54:18 +02:00