Mason Swofford

After training an efficientdet-lite0 or efficientdet-d0 model with quantization aware training (using `keras.train --hparams="model_optimizations.quantize={}" ...`), both the saved model and exported tflite model achieve 0 mAP when evaluated with keras.eval...

Seems like the dependencies are messed up. I've been trying to install and update them for about a half hour but have given up.

It would be helpful to see a streaming setup with latency benchmarks. Does anyone have TTFB benchmarks under any hardware?

Adding dask to Code Library Search plugin

3

I wanted to let you know that users of our Code Library Search Chatgpt plugin have requested we support dask. Unless we hear otherwise from you, we’ll add support for...

Azure TTS voice issue

3

If I change the voice using `AzureSynthesizer(AzureSynthesizerConfig.from_output_device(speaker_output, voice_name="en-GB-RyanNeural"))`, there seems to be some issue with the playback frequency. This occurs regardless of the voice I change it to. For comparison,...