shariq-audiofocus

Will sparse models ever result in smaller/compressed *.tflite models? This would be a huge plus for low power use cases as it would reduce I/O. Currently I'm working with a...

@gordinmitya Thanks, I hadn't heard of structural pruning, seems like that could lead to smaller tflite binaries if it eliminates entire filters. Is structural pruning on the model-optimization roadmap? Re:...

@liyunlu0618 - My use case: 1. Online, Streaming, Speech-Enhancement-like Task. Input Audio -> Dense -> LSTM -> Dense -> Output Audio. During training the Dense layers are actually CONV layers...

Also according to @podlipensky `tensorflow_graphics_gpu` is deprecated and should be removed from `README.md`

Also I had trouble with the OPENEXR installation so I had to install it from source and then tell the setup.py script to skip openexr. Here is my script in...

Having this issue too, @mcbaron did you ever find a solution/workaround?

Found a workaround: Comment out Line #89 in config/dependencies.cmake. It appears that libsamplerate doesn't use Carbon anymore so the patch is not necessary. ` PATCH_COMMAND ${GIT_EXECUTABLE} apply ${CMAKE_SOURCE_DIR}/config/fix_carbon.patch` to `#...

@sampathrajapaksha - We've found the best approach is to do knowledge distillation (KD) to shrink your model and therefore improve inference time. This paper has some good ideas: https://arxiv.org/pdf/1910.01108.pdf%3C/p%3E and...