roytseng-tw/Detectron.pytorch issues and pull requests

#238 - UnicodeDecodeError: 'ascii' codec can't decode byte 0xf5 in position 1: ordinal not in range(128)

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

#237 - undefined symbol: __cudaRegisterFatBinaryEnd

Issue - State: open - Opened by ZahraAnam over 3 years ago - 1 comment

#236 - driver418，cuda10.1 can run this project?

Issue - State: open - Opened by henbucuoshanghai about 4 years ago

#235 - AssertionError: Range subprocess failed (exit code: 1)

Issue - State: open - Opened by azhuantou about 4 years ago

#234 - AssertionError----assert len(roidb)==len(cached_roidb)

Issue - State: closed - Opened by ZY-Zeng about 4 years ago

#233 - nvcc.exe failed with exit status 1: Problem installing detectron2 on Windows 10

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

#232 - Hello

Issue - State: closed - Opened by XUSean0118 over 4 years ago

#231 - cuda

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

#230 - RuntimeError: PyTorch was compiled without NumPy support

Issue - State: open - Opened by rmxhhh over 4 years ago

#229 - ModuleNotFoundError: No module named 'model.roi_pooling._ext.roi_pooling._roi_pooling

Issue - State: open - Opened by rmxhhh over 4 years ago

#228 - no file name cityscapesscripts.evaluation.instances2dict_with_polygons

Issue - State: open - Opened by baiboat almost 5 years ago

#227 - Conversion to Tensorrt

Issue - State: open - Opened by ma-siddiqui almost 5 years ago

#226 - ImportError using Pytorch 1.1.0

Issue - State: closed - Opened by nQuant about 5 years ago - 3 comments

#225 - _pickle.UnpicklingError: invalid load key, '\x0a'.

Issue - State: open - Opened by guoguotingzhu over 5 years ago - 2 comments

#224 - minor typo fix

Pull Request - State: open - Opened by arvind-iyer over 5 years ago

#223 - Feature Suggestion: Support for Pytorch 1.0

Issue - State: open - Opened by yzhq97 over 5 years ago

#222 - Inference on CPU

Issue - State: open - Opened by ashnair1 over 5 years ago - 1 comment

#221 - UnicodeDecodeError when testing Detectron's Table Detection task

Issue - State: closed - Opened by asediri over 5 years ago - 1 comment

#220 - Accounting for varying object size in images of constant size

Issue - State: open - Opened by ashnair1 over 5 years ago - 1 comment

#219 - update README.md , add compiler version note

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

#218 - Syntax issue on paragraph 42

Pull Request - State: open - Opened by kant over 5 years ago

#217 - How to add new backbone to faster rcnn?

Issue - State: open - Opened by xiaozhuka over 5 years ago

#216 - training process acquired abnormal weights, i.e., nan

Issue - State: open - Opened by 19940312 over 5 years ago - 1 comment

#215 - a question about the time consumed by copying tensor from cpu to gpu

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

#214 - Add custom module to object detection pipeline and train jointly

Issue - State: open - Opened by ashnair1 over 5 years ago

#213 - Question about Faster-RCNN model

Issue - State: open - Opened by zhouhuan2005 over 5 years ago

#212 - No module named utils

Issue - State: open - Opened by MitraTj over 5 years ago

#211 - How to test on COCO test-dev dataset

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

#210 - Nan values for certain classes while testing

Issue - State: closed - Opened by ashnair1 over 5 years ago - 1 comment

#209 - Why does mynn.DataParallel not support multiple gpus?

Issue - State: closed - Opened by ashnair1 over 5 years ago - 2 comments

#208 - Compile the CUDA code

Issue - State: open - Opened by wgs123 over 5 years ago

#207 - Finetune with VOC2007

Issue - State: closed - Opened by Asif6511 over 5 years ago - 2 comments

#206 - wrong in compile

Issue - State: open - Opened by h510166404 over 5 years ago

#205 - error when training with my own coco-type data

Issue - State: closed - Opened by HarrisXia over 5 years ago - 1 comment

#204 - typos fixed, some sentences smoothened

Pull Request - State: closed - Opened by EMCP over 5 years ago

#203 - How to train a custom dataset with custom number of classes

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

#202 - torch.utils.ffi is deprecated with pytorch 0.4.1

Issue - State: closed - Opened by matchumen over 5 years ago - 1 comment

#201 - Index out of range on multi-GPU (8 gpus ) after first epoch

Issue - State: open - Opened by akshitac8 over 5 years ago

#200 - Using ‘Adam’ to train, encountering error.

Issue - State: open - Opened by lilichu over 5 years ago

#199 - Running eval script results in empty output directory

Issue - State: open - Opened by Terry-Kusunoki-Martin over 5 years ago - 2 comments

#198 - stucked when training is done

Issue - State: open - Opened by greathope almost 6 years ago

#197 - New Mask-X-RCNN Structure

Issue - State: open - Opened by timho102003 almost 6 years ago

#196 - Links broken to pre-trained backbones

Issue - State: closed - Opened by jhultman almost 6 years ago

#195 - multi-gpus training is much slower than caffe2 Detectron

Issue - State: closed - Opened by greathope almost 6 years ago

#194 - Support for PyTorch 1.0, TorchVision 0.3.0, PASCAL VOC Dataset and Custom Dataset

Pull Request - State: open - Opened by adityaarun1 almost 6 years ago - 3 comments

#193 - Question about dataaugmentation

Issue - State: open - Opened by yuyijie1995 almost 6 years ago

#192 - How to

Issue - State: closed - Opened by yuyijie1995 almost 6 years ago

#191 - Why set: 'upsample.upconv.weight': None in keypoint net?

Issue - State: closed - Opened by lilichu almost 6 years ago

#190 - Why freeze all bn (affine) layers?

Issue - State: open - Opened by lilichu almost 6 years ago - 1 comment

#189 - why my mask_loss didn't decrease?

Issue - State: open - Opened by lingtengqiu almost 6 years ago

#188 - Loss keep increasing

Issue - State: open - Opened by SingL3 almost 6 years ago - 2 comments

#187 - torch.utils.ffi is deprecated. Please use cpp extensions instead.

Issue - State: open - Opened by engineer1109 almost 6 years ago - 1 comment

#186 - change python3 to python2

Issue - State: closed - Opened by 13070151771 almost 6 years ago

#185 - is RPN_PRE_NMS_TOP_N per image?

Issue - State: open - Opened by wwn1233 almost 6 years ago

#184 - Initial commit for windows

Pull Request - State: closed - Opened by runhang almost 6 years ago

#183 - RetinaNet support

Pull Request - State: open - Opened by miliadis almost 6 years ago

#182 - Can this implementation handle images without objects?

Issue - State: open - Opened by SingL3 almost 6 years ago - 2 comments

#181 - Could I run the code on my macbook pro without GPU??

Issue - State: open - Opened by Hanhanhan11 almost 6 years ago

#180 - Could I modify the max iters ?

Issue - State: open - Opened by CoinCheung almost 6 years ago

#179 - loss_bbox almost equal to 0

Issue - State: open - Opened by liuliu66 about 6 years ago - 3 comments

#178 - Evaluation Precision x Recall

Issue - State: open - Opened by gsaibro about 6 years ago

#177 - adnormal loss for training process

Issue - State: open - Opened by At1a8 about 6 years ago

#176 - why boxes coordinates were not scaled in blobs?

Issue - State: open - Opened by Redaimao about 6 years ago

#174 - AssertionError: Range subprocess failed (exit code: 1)

Issue - State: open - Opened by chelixuan about 6 years ago - 3 comments

#173 - gcc: error trying to exec 'cc1': execvp: No such file or directory

Issue - State: open - Opened by chuxiang93 about 6 years ago

#172 - RuntimeWarning: Negative areas founds: 3 warnings.warn("Negative areas founds: %d" % neg_area_idx.size, RuntimeWarning))

Issue - State: open - Opened by PkuRainBow about 6 years ago - 5 comments

#171 - sh make.sh errors under pytorch version '1.0'.

Issue - State: open - Opened by youngwanLEE about 6 years ago - 7 comments

#170 - Loading detectron weights Error

Issue - State: open - Opened by B2Gdevs about 6 years ago

#168 - cudaCheckError() failed : no kernel image is available for execution on the device

Issue - State: open - Opened by 897051629 about 6 years ago - 4 comments

#167 - Grabbing Features for an SVM

Issue - State: open - Opened by B2Gdevs about 6 years ago - 8 comments

#164 - How can I do inference without mynn.DataParallel?

Issue - State: open - Opened by shiyi001 about 6 years ago - 4 comments

#159 - test error of keypoints

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

#158 - sh make.sh error cython incompatible?

Issue - State: closed - Opened by szrlee about 6 years ago - 10 comments

#154 - What is "detectron_weight_mapping" for?

Issue - State: closed - Opened by bowenc0221 about 6 years ago - 1 comment

#148 - Could not train on v100

Issue - State: open - Opened by qinhaifangpku about 6 years ago - 4 comments

#144 - Memory grows with less TEST.SCALE

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

#142 - Cannot test with the pre-trained Detectron model?

Issue - State: closed - Opened by baiyancheng20 about 6 years ago - 3 comments

#141 - python tools/download_imagenet_weights.py

Issue - State: open - Opened by zq130320339 about 6 years ago - 2 comments

#140 - Training is stuck at some point, I'm not sure if it is a PyTorch problem

Issue - State: closed - Opened by bowenc0221 about 6 years ago - 9 comments

#130 - UnicodeDecodeError: 'ascii' codec can't decode byte 0x95 in position 6: ordinal not in range(128)

Issue - State: closed - Opened by daijucug about 6 years ago - 4 comments

#127 - cannot I finetune from coco-pretrained models

Issue - State: open - Opened by CoinCheung over 6 years ago - 4 comments

#116 - UnicodeDecodeError when using COCO pre-trained Detectron models

Issue - State: closed - Opened by AruniRC over 6 years ago - 1 comment

#115 - Is the NMS soft NMS？

Issue - State: open - Opened by engineer1109 over 6 years ago - 3 comments

#101 - Runtime Error When running inference on Cityscapes dataset

Issue - State: closed - Opened by chengruizhe over 6 years ago - 1 comment

#100 - what is weight momentum for conv layers in pkl file ?

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

#99 - How to modify the mapping_to_detectron?

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

GitHub / roytseng-tw/Detectron.pytorch issues and pull requests