Autonomous training method and device of algorithm model, equipment and storage medium

The invention provides an autonomous training method and device of an algorithm model, equipment and a storage medium, and relates to the technical field of data processing. The method comprises the steps of determining whether to start a model training task according to parameters of a to-be-executed training task; if it is determined that the model training task is started, training a model corresponding to the to-be-executed training task based on a training data set to obtain a training index; judging according to the training index and a target parameter of the model, and determining whether an evaluation condition is met or not; and if the evaluation condition is reached, obtaining a trained target model. Compared with the prior art, the method avoids the problem that the algorithm model needs a lot of manual participation and is highly dependent on manual work. 本申请提供一种算法模型的自主训练方法、装置、设备及存储介质，涉及数据处理技术领域。该方法包括：根据待执行训练任务的参数，确定是否启动模型训练任务；若确定启动模型训练任务，则基于训练数据集对所述待执行训练任务对应的模型进行训练，获取训练指标；根据所述训练指标和所述模型的目标参数进行判断，