Quantization method based on neural network model and related equipment thereof

The invention discloses a model quantification method based on a neural network and related equipment thereof, and belongs to the field of artificial intelligence. The method comprises the following steps: acquiring a neural network graph structure, wherein the neural network graph structure comprises a plurality of operation nodes; inserting a plurality of quantization nodes in the neural network graph structure to obtain a quantization model graph structure; the quantitative model graph structure is trained according to sample data to obtain a quantitative model, and the size of the quantitative model is smaller than that of the neural network graph structure; wherein the training comprises the step of quantizing the input data of each quantization node by utilizing each quantization node in the plurality of quantization nodes to obtain the output data of each quantization node. 一种基于神经网络的模型量化方法及其相关设备，用于人工智能领域。该方法包括：获取神经网络图结构，所述神经网络图结构包括多个运算节点；在所述神经网络图结构中插入多个量化节点，以得到量化模型图结构；根据样本数据对所述量化模型图结构进行训练，以得到量化模型，所述量化模