bedapisl

It would be also useful for converting models to TensorRT, which has problems with dynamic batch sizes.

Hello, the model is SSD network from https://github.com/pierluigiferrari, the problematic layer is this one: https://github.com/pierluigiferrari/ssd_keras/blob/master/keras_layers/keras_layer_DecodeDetections.py#L27

Hello, here is a simple example which demonstrates the problem: ``` import keras import sys sys.path.insert(0, "./ssd_keras") #path to https://github.com/pierluigiferrari/ssd_keras from keras_layers.keras_layer_DecodeDetections import DecodeDetections import onnxmltools output_onnx_model = 'model.onnx' keras_model...

Yes. But I don't remember the exact details. I cloned the `yaml-cpp` repo, added the `-fPIC` option somewhere to the `CMakeLists` (I guess line 174) and recompiled and reinstalled `yaml-cpp`...

What is the output of the command `pip install fast-colorthief==0.0.4` for you? And what OS are you using?

The problem is that I did not upload wheels for Python 3.8, with Python 3.7 it should work. I will fix it eventually, but now I have a lot of...

Should work now with Python 3.8, 3.9 and 3.10 on Linux.

Hello, would you install CMake and then try installing with: ``` git clone https://github.com/bedapisl/fast-colorthief cd fast-colorthief git submodule update --init --recursive pip3 install . ```

Hello, using this library you cannot get the percentage of each color. But the colors in the palette should be ordered by their representation in the image.

That is not supported sorry.