as754770178

Results 28 comments of as754770178

import error

my gcc version is 4.9.0 ``` # The result of `python3 setup.py install` running install_lib running build_py running build_ext creating build/bdist.linux-x86_64/egg creating build/bdist.linux-x86_64/egg/warpctc_pytorch copying build/lib.linux-x86_64-3.7/warpctc_pytorch/_warp_ctc.cpython-37m-x86_64-linux-gnu.so -> build/bdist.linux-x86_64/egg/warpctc_pytorch copying build/lib.linux-x86_64-3.7/warpctc_pytorch/__init__.py ->...

import error

The end result of make: ``` Scanning dependencies of target warpctc Linking CXX shared library libwarpctc.so [ 33%] Built target warpctc [ 50%] Scanning dependencies of target test_cpu Building NVCC...

If still exec the compress process when inference?

thanks. 1. Whether the model size will be reduced? Because the model add new variable, such as, kernel_gthr and kernel_shrk, and not delete the old variable. 2. Whether the model...

If still exec the compress process when inference?

This is the pb file generated by `export_pb_tflite_models.py`, that the pb file of compress model is bigger than the pb file of original model. ``` -rw-rw-r--. 1 zgz zgz 95082914...

If still exec the compress process when inference?

I modify code beacuse report error `ValueError: Cannot feed value of shape (1, 224, 224, 3) for Tensor 'import/net_input:0', which has shape '(32, 224, 224, 3)'`. So I change `net['input_data']...

If still exec the compress process when inference?

The first result is to use my custom`resnet_v1_50`. The second is to use `resnet_v1_20` provided by PocketFlow.

If still exec the compress process when inference?

This is the log of first result. Through the log, we can know that the channel is pruned. ``` 2019-01-25 18:30:51.087294: I tensorflow/core/platform/cpu_feature_guard.cc:140] Your CPU supports instructions that this TensorFlow...

If still exec the compress process when inference?

I use the model save in `cp_best_path`.

If still exec the compress process when inference?

1. The conv in residual brach will be compressed? and why the size of pb file become bigger? 2. How to use the "tf.gather + 3 x 3 conv" scheme?

If still exec the compress process when inference?

sorry, my code is not the latest version. I will pull.