Tencent/PocketFlow issues and pull requests

#95 - inference time of model_transformed.pb is larger than model_original.pb;although model_transformed.pb is samll than model_original.pb

Issue - State: closed - Opened by distrect9 almost 6 years ago - 38 comments
Labels: enhancement

#94 - eval code seems to have bug

Issue - State: open - Opened by distrect9 almost 6 years ago - 5 comments
Labels: bug

#93 - Resnet & Dis-chn-pruned: Waiting for three hours.... No more INFO. Still Keep "Restoring parameters".

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

#92 - Demo for self-defined data sets and models

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

#91 - uniform error

Issue - State: closed - Opened by as754770178 almost 6 years ago - 6 comments

#90 - Setup learning rate in each ModelHelper

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

#89 - How to use distill when compress

Issue - State: closed - Opened by as754770178 almost 6 years ago - 1 comment

#88 - How many epochs to take during a compression process

Issue - State: closed - Opened by as754770178 almost 6 years ago - 12 comments

#87 - FLOPs compute

Issue - State: closed - Opened by leosk-knight almost 6 years ago - 7 comments

#86 - Cannot use multiGPUs

Issue - State: closed - Opened by SunAriesCN almost 6 years ago - 24 comments

#85 - DisChnPrunedLearner with resnet18 on ImageNet can't converge in local mode

Issue - State: closed - Opened by GoldenSpark almost 6 years ago - 7 comments
Labels: bug

#84 - GPU-based Channel Selection

Pull Request - State: closed - Opened by jiaxiang-wu almost 6 years ago - 2 comments

#83 - Pruned model slower inference times

Issue - State: closed - Opened by dhingratul almost 6 years ago - 4 comments
Labels: enhancement

#82 - Can support caffe model or not？？

Issue - State: closed - Opened by pxEkin almost 6 years ago - 1 comment

#81 - fix Shape Error #76

Pull Request - State: closed - Opened by ylfzr almost 6 years ago - 2 comments

#80 - Hi， Can you create a QQ communication group？

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

#79 - About support for 3D convolutions?

Issue - State: closed - Opened by yyjabidintg almost 6 years ago - 1 comment

#78 - apply in 1D convolutions

Issue - State: closed - Opened by 1448643857 almost 6 years ago - 4 comments

#77 - Fix Shape Error in issue #76

Pull Request - State: closed - Opened by ylfzr almost 6 years ago - 1 comment

#76 - Parameter Shape Error for non-uniform Quantization

Issue - State: closed - Opened by ylfzr almost 6 years ago - 1 comment

#75 - Fix for using single GPU in local mode; Change of definition of idle GPU

Pull Request - State: closed - Opened by KranthiGV almost 6 years ago - 3 comments

#74 - numerical stability for 1 bit quantization

Pull Request - State: closed - Opened by ylfzr almost 6 years ago - 1 comment

#73 - Numerical Problem Quantizing Model to 1 Bit

Issue - State: closed - Opened by ylfzr almost 6 years ago - 1 comment

#72 - Train ResNet-20 model for the CIFAR-10 on colab not working

Issue - State: closed - Opened by besherh almost 6 years ago - 13 comments

#71 - mpi usage about multi GPU

Issue - State: closed - Opened by ginger0106 almost 6 years ago - 2 comments

#70 - Update the command for model conversion

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

#69 - How to use Pre trained model for Weight sparsifictaion?

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

#68 - Question on 'uniform_quantization_tf'

Issue - State: closed - Opened by wangxianrui almost 6 years ago - 2 comments

#67 - batch_size related

Issue - State: closed - Opened by backyes almost 6 years ago - 1 comment

#66 - Fix a few documentation issues

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

#65 - [Issue#52]Fix an bug of channel pruning

Pull Request - State: closed - Opened by psyyz10 almost 6 years ago

#64 - Can't use MultiGPU in on channel mode

Issue - State: closed - Opened by xiaomr almost 6 years ago - 18 comments

#63 - Question about transforming the quantized model to '.tflite' format

Issue - State: closed - Opened by Maximilianxu almost 6 years ago - 7 comments
Labels: documentation

#62 - I can't convert model compressed by uniform-tf to pb or tflite？

Issue - State: closed - Opened by sunzhe09 almost 6 years ago - 3 comments

#61 - questions about TensorFlow Quantization Wrapper

Issue - State: closed - Opened by wangxianrui almost 6 years ago - 12 comments
Labels: documentation

#60 - Model Pruning taking too much time to train

Issue - State: closed - Opened by dhingratul almost 6 years ago - 8 comments
Labels: enhancement

#59 - I want to compress my own model such as Gan's or mtcnn ,how can I change the code?can you supply a demo or document about how can we change the file to work on my own model?

Issue - State: closed - Opened by sunzhe09 almost 6 years ago - 5 comments
Labels: documentation

#58 - Does it support Mobilenet-v2-SSD ?

Issue - State: closed - Opened by haikuoyao about 6 years ago - 5 comments

#57 - Question on the uniform quantization

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

#56 - How to use the pretrained models?

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

#55 - distill model in full-prec mode

Issue - State: closed - Opened by as754770178 about 6 years ago - 7 comments

#54 - calculate distillation loss

Issue - State: closed - Opened by as754770178 about 6 years ago - 2 comments

#53 - Can not run training on local mode

Issue - State: closed - Opened by besherh about 6 years ago - 5 comments

#52 - I can't run "Automatic Channel Pruning" on local mode .

Issue - State: closed - Opened by sunzhe09 about 6 years ago - 12 comments

#51 - how to compress in my dataset

Issue - State: closed - Opened by as754770178 about 6 years ago - 6 comments

#50 - "ValueError: unrecognized learner's name: tf-uniform“ . In tutorial, --learner specification is not accordant with code

Issue - State: closed - Opened by LinkenMu about 6 years ago - 3 comments
Labels: documentation

#49 - is this fully_quantized model?

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

#48 - What's the training parameters for MobileNetV2 8 Bit Fixed-point?

Issue - State: closed - Opened by edgeflowdlai about 6 years ago - 12 comments
Labels: documentation

#47 - tensorflow.python.framework.errors_impl.FailedPreconditionError: Bad file descriptor

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

#46 - ./scripts/run_local.sh nets/lenet_at_cifar10_run.py gets a segmentfault ?

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

#45 - how to set my data path

Issue - State: closed - Opened by 1448643857 about 6 years ago - 3 comments

#44 - can pocketflow be used for object detection model

Issue - State: closed - Opened by GallonDeng about 6 years ago - 16 comments
Labels: enhancement

#43 - model path

Issue - State: closed - Opened by as754770178 about 6 years ago - 9 comments

#42 - The description for uql_bucket_size and uql_quantize_all_layers is the same in the uniform quantization doc.

Issue - State: closed - Opened by qingqing01 about 6 years ago - 2 comments
Labels: documentation

#41 - How can I convert sparise weight model to *.pb

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

#40 - Add source files for documentations

Pull Request - State: closed - Opened by jiaxiang-wu about 6 years ago

#39 - Can I use my own dataset to retrain?

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

#38 - Add source files for documentations

Pull Request - State: closed - Opened by jiaxiang-wu about 6 years ago - 1 comment

#37 - why it reports FLAGS.data_dir_local is None?

Issue - State: closed - Opened by zhangqianYY about 6 years ago - 5 comments

#36 - Make channel pruned model be able to convert to tflite

Pull Request - State: closed - Opened by psyyz10 about 6 years ago

#35 - How to compress model in Multi-machine

Issue - State: closed - Opened by as754770178 about 6 years ago - 12 comments

#34 - Can't use GPU in the local mode.

Issue - State: closed - Opened by howtocodewang about 6 years ago - 12 comments
Labels: enhancement

#33 - Add requirement.txt

Pull Request - State: closed - Opened by psyyz10 about 6 years ago - 1 comment

#32 - Documentation Issue

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

#31 - the produced mode can be used on nvidia-tx2(arm) gpu chip, which is jetson not x86?

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

#30 - Resnet20 cifar10 compress experiment cant get a small model

Issue - State: closed - Opened by xiaomr about 6 years ago - 12 comments

#29 - Single GPU card is not available on PC

Issue - State: closed - Opened by Aspirinkb about 6 years ago - 3 comments
Labels: enhancement

#28 - The name of variables in net is prefixed by 'model'

Issue - State: closed - Opened by as754770178 about 6 years ago - 1 comment

#27 - The name of variables in net is prefixed by 'model'

Issue - State: closed - Opened by as754770178 about 6 years ago - 6 comments

#26 - how i use docker mode or seven mode to train using muti-GPUs

Issue - State: closed - Opened by herbiezhao about 6 years ago - 5 comments

#25 - i can't get the horovod images using the cmd 'docker pull uber/horovod' which provided by tutorial,how can i get the docker enviroment?

Issue - State: closed - Opened by MaxwellHan about 6 years ago - 8 comments

#24 - no requirements.txt file to help user install the necessary packages

Issue - State: closed - Opened by EricDoug about 6 years ago - 2 comments
Labels: enhancement

#23 - Why learner has hard-code about dataset, such as, different dataset uses different learning rate control policy.

Issue - State: open - Opened by backyes about 6 years ago - 3 comments
Labels: enhancement

#22 - Update some documentation

Pull Request - State: closed - Opened by psyyz10 about 6 years ago - 1 comment

#21 - Update some documentation

Pull Request - State: closed - Opened by psyyz10 about 6 years ago - 1 comment

#20 - about transformer framework

Issue - State: open - Opened by chenweihua91 about 6 years ago - 4 comments
Labels: enhancement

#19 - The structure of my resnet is different from built in PocketFlow，I how to change the structure?

Issue - State: closed - Opened by as754770178 about 6 years ago - 11 comments

#18 - Segmentation fault

Issue - State: closed - Opened by as754770178 about 6 years ago - 7 comments

#17 - how to use local model

Issue - State: closed - Opened by as754770178 about 6 years ago - 1 comment

#16 - Supported operations of PocketFlow

Issue - State: open - Opened by lanyastar about 6 years ago - 4 comments
Labels: enhancement

#15 - Compressed model file from Channel Pruning

Issue - State: closed - Opened by dhingratul about 6 years ago - 10 comments
Labels: enhancement

#14 - Running Quantized model on a GPU

Issue - State: closed - Opened by dhingratul about 6 years ago - 12 comments

#13 - about fine tune

Issue - State: closed - Opened by yuanry16 about 6 years ago - 1 comment

#12 - Running PocketFlow in the local mode with cifar-10, it appears "./scripts/run_local.sh: line 45: 9232 Segmentation fault"

Issue - State: closed - Opened by violet17 about 6 years ago - 17 comments

#11 - Support for 1D convolutions

Issue - State: closed - Opened by dorian95 about 6 years ago - 3 comments

#10 - A question about DL framework

Issue - State: closed - Opened by JSnobody about 6 years ago - 8 comments

#9 - Which optimization functions need to be retrained?

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

#8 - Channel selection cost too much time

Issue - State: closed - Opened by BowieHsu about 6 years ago - 5 comments
Labels: enhancement

#7 - Pretrained models

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

#6 - K8S Supported?

Issue - State: open - Opened by xieydd about 6 years ago - 2 comments
Labels: enhancement

#5 - Issue: tensorflow.python.framework.errors_impl.NotFoundError: "/home/karen/cifar-10-batches-bin/"; No such file or directory

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

#4 - add english translation of CONTRIBUTING

Pull Request - State: closed - Opened by kxylu about 6 years ago - 1 comment

GitHub / Tencent/PocketFlow issues and pull requests