facebookresearch/Detectron issues and pull requests

#1048 - Introduced Safer SMTP Alternative: Added `SMTP_SSL`

Pull Request - State: open - Opened by fazledyn-or 10 months ago - 1 comment
Labels: CLA Signed

#1047 - s,np.bool\b,bool,

Pull Request - State: open - Opened by igorsugak about 1 year ago - 1 comment
Labels: CLA Signed, fb-exported

#1046 - Precision and recall (not AP not AR) values per class

Issue - State: open - Opened by khodabakhshih about 1 year ago

#1045 - Really?

Issue - State: open - Opened by MiaMariPer3 over 1 year ago

#1044 - Update issue_template.md

Pull Request - State: closed - Opened by VarunKeta over 1 year ago - 1 comment

#1043 - why is threshold of detector confidence is 0.05, not 0.5? Helps, bros

Issue - State: open - Opened by meihuanyuyuyu over 1 year ago

#1042 - Suggestion for improvement.

Pull Request - State: open - Opened by ShambhaviVijay over 1 year ago - 2 comments
Labels: CLA Signed

#1039 - problem with adaptive streaming

Issue - State: open - Opened by Rguzman2022 over 1 year ago - 3 comments

#1038 - App etiquette

Issue - State: closed - Opened by 16CentAstrology almost 2 years ago - 2 comments

#1035 - Mask-RCNN model not properly generating segmentation masks for a specific class- custom dataset

Issue - State: closed - Opened by abhinavchadaga almost 2 years ago

#1034 - detectron implantación inparcial,solicitud de inplantacion

Issue - State: open - Opened by bbnuditamesa almost 2 years ago - 2 comments

#1033 - Convert Cityscapes to COCO format: How to convert to other classes (ex: traffic light)

Issue - State: open - Opened by MarceloContreras almost 2 years ago - 1 comment

#1032 - Caffe 2 merged with pytorch new installation instruction?

Issue - State: open - Opened by atulya-deep almost 2 years ago

#1031 - Project dependencies may have API risk issues

Issue - State: open - Opened by PyDeps about 2 years ago

#1030 - Detectron or Detection?

Issue - State: open - Opened by CPones over 2 years ago - 1 comment

#1029 - Is there any script for batch inference?

Issue - State: open - Opened by rogercmq about 3 years ago

#1028 - How to train Faster R-CNN on my own custom dataset?

Issue - State: open - Opened by meshghi about 3 years ago - 4 comments

#1027 - How can I disable the logging system?

Issue - State: closed - Opened by rogercmq about 3 years ago - 1 comment

#1026 - n/a

Issue - State: open - Opened by markbroich about 3 years ago

#1025 - libcaffe2_detectron_ops_gpu.so运行慢

Issue - State: open - Opened by y782348493 over 3 years ago

#1024 - KeyError: u'Key TEST.SCALES was renamed to TEST.SCALE; please update your config. Note: Also convert from a tuple, e.g. (600, ), to a integer, e.g. 600.'

Issue - State: open - Opened by Dou-Yu-xuan over 3 years ago

#1023 - Detectron2 In-Place One-to-Many Augmentations

Issue - State: open - Opened by kmeng01 over 3 years ago - 2 comments

#1022 - 关于bbox损失函数部分，其中mask rcnn的λ权重平衡参数在哪啊配置和代码里没看见啊。。。

Issue - State: closed - Opened by shuxsu over 3 years ago - 1 comment

#1021 - the speed is slow

Issue - State: open - Opened by nwpuwc over 3 years ago

#1019 - Bump pyyaml from 3.12 to 5.4

Pull Request - State: open - Opened by dependabot[bot] over 3 years ago
Labels: CLA Signed, dependencies

#1018 - Multi-machine launch stuck and hangs

Issue - State: closed - Opened by DianCh over 3 years ago - 4 comments

#1017 - What is the IoU threshold for evaluation log on Per-category segmentation AP ?

Issue - State: open - Opened by arnaldododo over 3 years ago

#1016 - What is the IoU threshold for evaluation log on Per-category segmentation AP ?

Issue - State: closed - Opened by arnaldododo over 3 years ago

#1015 - fix yaml version

Pull Request - State: closed - Opened by FastStim over 3 years ago - 1 comment

#1014 - What is the value of boxes in vis_one_image in vis.py ?

Issue - State: open - Opened by sharon-samson over 3 years ago

#1013 - Dockerfile

Issue - State: open - Opened by sh7jacobi over 3 years ago

#1012 - caffe2 converted model performance

Issue - State: open - Opened by juvebogdan over 3 years ago

#1011 - Recall evaluation results lower than expected (instance segmentation)

Issue - State: closed - Opened by jdegeus almost 4 years ago

#1010 - _log_detection_eval_metrics reads precision array from coco_eval wrong

Issue - State: open - Opened by deuscovrigus almost 4 years ago

#1009 - How do you register a cityscape dataset?

Issue - State: open - Opened by slulura almost 4 years ago

#1008 - Not all items detected in picture (just not a big part)

Issue - State: closed - Opened by revaldo666 almost 4 years ago

#1007 - where is focal loss in code

Issue - State: open - Opened by nengwp about 4 years ago - 1 comment

#1005 - Update requirements.txt to install opencv 4.2

Pull Request - State: open - Opened by edupooch about 4 years ago - 3 comments
Labels: CLA Signed

#1004 - Increasing segmentation mask resolution

Issue - State: open - Opened by predrag12 about 4 years ago - 1 comment

#996 - Shape error running inference on exported pretrained COCO Mask-RCNN

Issue - State: closed - Opened by xavierdcruz over 4 years ago - 2 comments

#993 - AssertionError: Override list has odd length

Issue - State: closed - Opened by ChauncyFr over 4 years ago - 3 comments

#991 - Fix backgound sample without replacement

Pull Request - State: closed - Opened by mdv3101 over 4 years ago - 2 comments
Labels: CLA Signed

#990 - Train with Negative Dataset

Issue - State: closed - Opened by mdv3101 over 4 years ago

#988 - yaml.constructor.ConstructorError: could not determine a constructor for the tag ...

Issue - State: closed - Opened by yakupakkaya over 4 years ago - 2 comments

#987 - IndexError:list index out of range

Issue - State: open - Opened by Kingdroper over 4 years ago - 2 comments

#985 - AssertionError: Override list has odd length: ['MODEL.WEIGHTS', 'detectron2://COCO-InstanceSegmentation', 'mask_rcnn_R_50_FPN_3x/137849600/model_final_f10217.pkl']; it must be a list of pairs

Issue - State: open - Opened by Nizar-Mhatli over 4 years ago - 4 comments

#976 - Fix convert_pkl_to_pb import error

Pull Request - State: open - Opened by chenhangcal over 4 years ago - 2 comments
Labels: CLA Signed

#961 - update tempfile objects to accept strings

Pull Request - State: open - Opened by kmboehm almost 5 years ago
Labels: CLA Signed

#958 - Bump pyyaml from 3.12 to 5.1

Pull Request - State: closed - Opened by dependabot[bot] almost 5 years ago - 1 comment
Labels: CLA Signed, dependencies

#954 - Update issue_template.md

Pull Request - State: open - Opened by pragyabansal02 almost 5 years ago - 1 comment

#950 - Added Video Support

Pull Request - State: open - Opened by pratikbhave2 almost 5 years ago - 2 comments
Labels: CLA Signed

#943 - How to output inference result like the annotation format

Issue - State: open - Opened by minan19605 about 5 years ago - 3 comments

#942 - Can we train the Faster RCNN with customer data set?

Pull Request - State: open - Opened by adeebakausar about 5 years ago - 1 comment

#930 - python3 supported

Pull Request - State: open - Opened by zavidovych about 5 years ago - 2 comments
Labels: CLA Signed

#920 - Created the Pull Request Template

Pull Request - State: open - Opened by DeveloperAlfa about 5 years ago - 2 comments
Labels: CLA Signed

#894 - RuntimeError: [enforce fail at conv_op_cudnn.cc:811] status == CUDNN_STATUS_SUCCESS. 8 vs 0. , Error at: /pytorch/caffe2/operators/conv_op_cudnn.cc:811: CUDNN_STATUS_EXECUTION_FAILED

Issue - State: open - Opened by Emma0928 over 5 years ago - 5 comments

#890 - What does KeyError: u'area' means ?

Issue - State: open - Opened by yoshua133 over 5 years ago - 1 comment

#869 - perform training on negative samples (images with no annotation)

Pull Request - State: open - Opened by gcheron over 5 years ago - 4 comments
Labels: CLA Signed

#865 - Rename ndim() -> dim() and size() -> numel()

Pull Request - State: open - Opened by shenyunhang over 5 years ago
Labels: CLA Signed

#863 - Delete issue_template.md

Pull Request - State: open - Opened by sunyspark over 5 years ago - 1 comment

#862 - Rename remaining ndim() -> dim(), fix typos

Pull Request - State: open - Opened by Johnqczhang over 5 years ago
Labels: CLA Signed

#859 - why set freeze_at = 2？

Issue - State: open - Opened by hysshy over 5 years ago - 4 comments

#851 - How can I get the visualization of feature maps after trained with Detectron?

Issue - State: closed - Opened by CPFelix over 5 years ago - 2 comments

#842 - RuntimeError: [enforce fail at context_gpu.cu:234]

Issue - State: open - Opened by JeasonUESTC over 5 years ago - 3 comments

#838 - Request Faster-RCNN-GN pre-trained models

Issue - State: closed - Opened by ltnghia over 5 years ago

#833 - RuntimeError: [enforce fail at net.cc:69] . op BBoxTransform: Source for input rpn_rois is unknown for net detectron, operator input: "rpn_rois" input: "bbox_pred" input: "im_info" output: "pred_bbox" name: "" type: "BBoxTransform" arg { name: "correct_transform_coords" i: 1 } arg { name: "apply_scale" i: 0 } arg { name: "weights" floats: 10 floats: 10 floats: 5 floats: 5 }

Issue - State: open - Opened by pjoshi0621 over 5 years ago - 1 comment

#782 - ImportError: No module named cython_bbox

Issue - State: closed - Opened by dengshanlearn almost 6 years ago - 4 comments

#772 - .ipynb notebook for running Detectron in Google Colab

Pull Request - State: open - Opened by Dene33 almost 6 years ago - 4 comments
Labels: CLA Signed

#738 - Added error msg if freeze_at = 5 but body has 4 blocks

Pull Request - State: open - Opened by Juggernaut93 almost 6 years ago
Labels: CLA Signed

#716 - Find detectron ops from pytorch nightly build

Pull Request - State: open - Opened by darthsuogles almost 6 years ago - 2 comments
Labels: CLA Signed

#713 - Training Fail (train_net.py)

Issue - State: closed - Opened by shadowdk3 almost 6 years ago - 3 comments

#698 - add convert op BatchPermutation to Gather

Pull Request - State: open - Opened by zanadu123 almost 6 years ago
Labels: CLA Signed

#690 - Feature/replaceable detection helper

Pull Request - State: open - Opened by DXist almost 6 years ago - 3 comments
Labels: CLA Signed

#686 - Convert FPN models to .pb for gpu device

Issue - State: closed - Opened by obendidi about 6 years ago - 2 comments

#662 - when I train coco2014 dataset,I get an error,how to solve it?

Issue - State: closed - Opened by gittigxuy about 6 years ago - 1 comment

#659 - Why need the function of 'blob_utils.get_loss_gradients'?

Issue - State: open - Opened by yw155 about 6 years ago - 1 comment

#640 - how to understand 'skeleton' in coco datasets keypoint?

Issue - State: closed - Opened by lilichu about 6 years ago - 3 comments

#635 - Train mask rcnn with COCO

Issue - State: closed - Opened by lan-lw about 6 years ago - 9 comments

#596 - Add vanilla and optimized cpu device to detectron

Pull Request - State: open - Opened by jgong5 about 6 years ago - 8 comments
Labels: CLA Signed

#508 - feat: add network visualization tool

Pull Request - State: open - Opened by WuZhuoran over 6 years ago
Labels: CLA Signed

#504 - evaluation of kps

Issue - State: closed - Opened by jinxuan777 over 6 years ago - 2 comments

#459 - Question about post_nms_topN size in collect_and_distribute_fpn_rpn_proposals.py

Issue - State: closed - Opened by roytseng-tw over 6 years ago - 2 comments

#458 - Support rle annotation in mask r-cnn training

Pull Request - State: closed - Opened by kyamagu over 6 years ago - 3 comments
Labels: CLA Signed

#447 - What is one iteration?

Issue - State: closed - Opened by natasasdj over 6 years ago - 4 comments

#405 - Cannot train with multi-label

Issue - State: open - Opened by kevinlin311tw over 6 years ago - 7 comments

#317 - Error occur when train keypoint rcnn

Issue - State: closed - Opened by Metal-joker over 6 years ago - 7 comments

#303 - will the "Detecting and Recognizing Human-Object Interactions" training code and model release?

Issue - State: open - Opened by EthanGuan over 6 years ago - 17 comments

#284 - Recognize same object in a video

Issue - State: open - Opened by ciolo over 6 years ago - 7 comments
Labels: community help wanted

#257 - Can not find the source code of Retinanet loss function.

Issue - State: closed - Opened by zhouhao94 over 6 years ago - 5 comments

#235 - Add new class

Issue - State: closed - Opened by EgorAntonovich over 6 years ago - 3 comments

#229 - Couldn't download the model through --wts https://s3-us-west-2.amazonaws.com/detectron/35861858/12_2017_baselines/e2e_mask_rcnn_R-101-FPN_2x.yaml.02_32_51.SgT4y1cO/output/train/coco_2014_train:coco_2014_valminusminival/generalized_rcnn/model_final.pkl

Issue - State: closed - Opened by AutomanHan over 6 years ago - 2 comments

GitHub / facebookresearch/Detectron issues and pull requests