Convert Caffe models to TensorFlow 2.x.

• Ubuntu 18.04
• Tensorflow 2.3
• Caffe and Pycaffe

Install caffe For Ubuntu

apt install caffe-cpu
apt install python3-caffe-cpu
pip3 install --upgrade scikit-image

Run convert.py to convert an existing Caffe model to TensorFlow.

python3 convert.py <prototxt file> --caffemodel <caffemodel file> --data-output-path <data file in numpy format> --code-output-path <filename to save graph as python class> --phase train --standalone-output-path <path to save savedmodel> --input_node <innode> --output_node <onode1>,<onode2>

Make sure you're using the latest Caffe format (see the notes section for more info).

The output consists of 3 files:

1. A data file (in NumPy's native format) containing the model's learned parameters.
2. A Python class that constructs the model's graph.
3. Tensorflow SavedModel file

• Only the new Caffe model format is supported. If you have an old model, use the upgrade_net_proto_text and upgrade_net_proto_binary tools that ship with Caffe to upgrade them first. Also make sure you're using a fairly recent version of Caffe.

• It appears that Caffe and TensorFlow cannot be concurrently invoked (CUDA conflicts - even with set_mode_cpu). This makes it a two-stage process: first extract the parameters with convert.py, then import it into TensorFlow.

• Pycaffe is strictly required.

• Only a subset of Caffe layers and accompanying parameters are currently supported.

• The border values are handled differently by Caffe and TensorFlow. However, these don't appear to affect things too much.

• Image rescaling can affect the ILSVRC2012 top 5 accuracy listed above slightly. VGG16 expects isotropic rescaling (anisotropic reduces accuracy to 88.45%) whereas BVLC's implementation of GoogLeNet expects anisotropic (isotropic reduces accuracy to 87.7%).

• The support class kaffe.tensorflow.Network has no internal dependencies. It can be safely extracted and deployed without the rest of this library.

1. https://github.com/ethereon/caffe-tensorflow
2. https://github.com/dhaase-de/caffe-tensorflow-python3
3. https://github.com/linkfluence/caffe-tensorflow
4. https://github.com/davidsandberg/caffe-tensorflow