Create new layers, loss functions, and develop state-of-the-art models. Write custom building blocks to express new ideas for research. Keras models are made by connecting configurable building blocks together, with few restrictions. It provides clear and actionable feedback for user errors. Keras has a simple, consistent interface optimized for common use cases. It’s used for fast prototyping, advanced research, and production, with three key advantages: QKeras is a quantization extension to Keras that provides drop-in replacement for some of the Keras layers, especially the ones that creates parameters and activation layers, and perform arithmetic operations, so that we can quickly create a deep quantized version of Keras network.Īccording to Tensorflow documentation, Keras is a high-level API to build and train deep learning models. For example, quantized_bits and quantized_relu bits and int_bits from Qtools will match exactly ac_fixed datatypes (if you rely on QKeras alone, the correct datatype should be ac_fixed, where is_negative has to be inferred from the other parameters of the quantizer. Qtools will estimate the sizes and types of operations to perform inference, with its data sizes compatible with high-level synthesis datatypes.Qtools for estimating effort to perform inference Stochastic behavior (including stochastic rouding) is disabled during inference
