Model training method and device, electronic equipment and storage medium

Abstract

The invention discloses a model training method and device, electronic equipment and a storage medium, and relates to the technical field of deep learning. The specific scheme comprises the followingsteps of receiving a task category label input by a user; generating at least one training sample corresponding to the task category label based on the task category label; extracting a training sample from all the training samples corresponding to the task category label as a current training sample; in response to the fact that the to-be-trained model does not meet the preset convergence condition, inputting the current training sample into the to-be-trained model, and training the to-be-trained model by using the current training sample; and repeatedly executing the operation of extractingthe current training sample until the model to be trained meets a preset convergence condition. According to the embodiment of the invention, model training can be realized without acquiring the training sample with the label in advance, so that the manpower consumption and the capital cost of manual labeling are greatly reduced.

Description

technical field [0001] The present application relates to the field of artificial intelligence, and further relates to the field of deep learning technology, especially a model training method, device, electronic equipment and storage medium. Background technique [0002] With the advent of the era of big data, data acquisition has become relatively easy, but the data used for training often needs to be manually screened and labeled before training. A large amount of training data means that it takes a lot of manpower, time, and money to label the data, which greatly limits the training speed of the artificial intelligence model, which in turn affects the model iteration speed and model launch time. [0003] In the existing technology, a supervised artificial intelligence model training method is usually used. Taking image classification tasks as an example, an appropriate amount of training data with labels is manually annotated, and training data is extracted based on trad...

AI Technical Summary

Problems solved by technology

This method relies on a large amount of manually labeled data, which requires a lot of manpower, money and time to invest in data labeling work; when the task is urgent, it may not be possible to obtain a large amount of labeled data, which will lead to model failure. If the performance requirements cannot be met in the short term, it cannot be launched; if the training task is to control an extremely important risk, then if any risk leaks, it will cause irreparable consequences to the company

It can be seen that the use of existing model training methods not only has low model training efficiency, but may also lead to immeasurable consequences

Images