PocketFlow
PocketFlow copied to clipboard
Tencent
An Automatic Model Compression (AutoMC) framework for developing smaller and faster AI applications.
QQ group
I have sent application to join QQ group (ID: 827277965) in pocketflow github many times,but group owner hasn't agreed till now.
This template is for miscellaneous issues not covered by the other issue categories. For questions on how to work with PocketFlow, or support for problems that are not verified bugs...
First of all, thanks for the open source on AutoMC. I really appreciate it. When I tried to run `ChannelPrunedLearner` on MobileNetV1 using the following command ```sh ./scripts/run_local.sh nets/mobilenet_at_ilsvrc12_run.py --learner...
When I perform channel pruning on the mobilenet at ilsvrc12 dataset,this error occured. But the pruning at cifar10 dataset can be done normally.
在使用auto通道裁剪中，使用error与flops的组合约束方法，在不指定裁剪率的情况下模型无法收敛。总共搜索200次，最佳值出现在50多次，此时模型保留率为0.85左右，后续搜索出来的结果模型保留率趋向于1，也就是模型不裁剪，并且稳定的使模型保留率为1。请问这是什么原因呢？
**Describe the bug** Background: The ssd-vgg-16 model is pruned by learner chn-pruned-rmt method and have been fine trained 120000 iters.It makes a dir named models_cpr_eval to strore the final pruned-trained...
I have trouble training **Faster RCNN** with **Pascal VOC** dataset (Batch size > 1). It seems like inputs have **different shape** so that they cant be gathered into batchs. I...
**Is your feature request related to a problem? Please describe.** The default logging create a gigantic file for tf summary and tensorboard. Sometimes, it may exhaust all disk space. My...
I use DCP to prune mobilenet_v2 with prune ratio 0.5. After pruning model, I use export_chn_pruned_tflite_model to compress model from ckpt to pb. But I find that the size of...
could you share me the details about deep learning frame supported by pocket flow
