Text classification model training method and device, computer equipment and storage medium

The invention discloses a text classification model training method. The method comprises the following steps: acquiring a text training data set; executing a dynamic learning rate adjustment operation on the initial local model in a training process to obtain a prediction label; according to the sample label corresponding to the same text training data and all the prediction labels, determining a prediction loss value, and screening all the text training data according to the prediction loss value to obtain target text data; training the first local model through the target text data to obtain a second local model; and obtaining local model parameters corresponding to the second local model, so that the server generates a text classification model according to the local model parameters sent by all the clients. According to the method, the initial local model is trained through all the text training data, and the prediction loss values under different learning rates are determined. The second local model is