Data processing method, device, computer equipment and storage medium

A neural network quantization method applied to a terminal, wherein the terminal is configured to run a neural network to perform a target task, and the method comprises:

• according to a precision requirement of the target task and/or the terminal, determining a plurality of pieces of data to be quantized from target data, wherein each piece of data to be quantized is a subset of the target data, the target data is data to be operated and quantized in a layer of the neural network to be quantized, and the data to be operated and quantized includes at least one of an input neuron, a weight, a bias, and a gradient;
• quantizing each piece of data to be quantized according to a corresponding quantization parameter to obtain a piece of quantized data;
• and
• obtaining a quantization result of the target data according to the piece of quantized data corresponding to each piece of data to be quantized, so that an operation is performed in the layer to be quantized according to the quantization result of the target data.

A neural network quantization device applied to a terminal, wherein the terminal is configured to run a neural network to perform a target task, and the device comprises:

• a data determination module configured to, according to a precision requirement of the target task and/or the terminal, determine a plurality of pieces of data to be quantized from target data, wherein each piece of data to be quantized is a subset of the target data, the target data is data to be operated and quantized in a layer to be quantized of the neural network, and the data to be operated includes at least one of an input neuron, a weight, a bias, and a gradient;
• a data quantization module configured to quantize each piece of data to be quantized according to a corresponding quantization parameter to obtain a piece of quantized data corresponding to each piece of data to be quantized;
• and
• a result determination module configured to obtain a quantization result of the target data according to the piece of quantized data corresponding to each piece of data to be quantized, so that an operation is performed in the layer to be quantized according to the quantization result of the target data.