Skip to content

Instantly share code, notes, and snippets.

@deepak-karkala

deepak-karkala/tf_model_quantisation.py

Created December 20, 2020 19:25
Show Gist options
  • Save deepak-karkala/a125c1a8b21296172cfa5c95f6fbc792 to your computer and use it in GitHub Desktop.
Save deepak-karkala/a125c1a8b21296172cfa5c95f6fbc792 to your computer and use it in GitHub Desktop.
Download ZIP
Model quantisation using TFLite
Raw
tf_model_quantisation.py
# Model quantisation using TFLite
save_model_path = "/tmp/"
# Save original model in tflite format
tflite_models_dir = pathlib.Path(save_model_path)
tflite_models_dir.mkdir(exist_ok=True, parents=True)
converter = tf.lite.TFLiteConverter.from_keras_model(model)
tflite_model = converter.convert()
tflite_model_file = tflite_models_dir/"model.tflite"
print(tflite_model_file.write_bytes(tflite_model))
# Save quantized model in tflite format
converter.optimizations = [tf.lite.Optimize.DEFAULT]
converter.target_spec.supported_types = [tf.float16]
tflite_fp16_model = converter.convert()
tflite_model_fp16_file = tflite_models_dir/"model_quant_f16.tflite"
print(tflite_model_fp16_file.write_bytes(tflite_fp16_model))
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment