dingfengffffff/bazel200_tensorflow220 repository overview

Use tensorflow transform_graph to transform your graphe is ver useful. But use bazel to build it is very slow. So this image may help you. Use the comman as shown below.

docker run --rm -it -v /home/ding/Downloads/te:/tensorflow dingfengffffff/bazel200_tensorflow220 transform_graph --in_graph=/tensorflow/deeplab_256_quant.pb --out_graph=/tensorflow/deeplab_256_quant_2.pb --inputs=MobilenetV2/MobilenetV2/input --outputs=ResizeBilinear_2 --transforms='strip_unused_nodes(type=float, shape="1,256,256,3")'