ngtranai09/CostSensitive

该数据集包含了模型的名称、训练集和测试集的大小，以及训练过程中的参数设置，如自动寻找批处理大小、梯度累积步数、学习率、日志记录步数、学习率调度器类型、训练周期数、输出目录、报告输出目标、保存策略、保存限制、随机种子、预热步数和权重衰减等。此外，还包括了可训练参数的数量、内存分配情况、训练时间以及性能指标，如准确率、F1宏平均值、F1加权平均值、精确度和召回率等。但是具体的数据集内容和用途没有描述。

The dataset includes the model name, the size of the training and test sets, and the parameter settings during training, such as auto-find batch size, gradient accumulation steps, learning rate, logging steps, learning rate scheduler type, number of training epochs, output directory, report target, save strategy, save total limit, seed, warmup steps, and weight decay, etc. It also includes the number of trainable parameters, memory allocation, training time, and performance metrics such as accuracy, F1 macro, F1 weighted, precision, and recall. However, the specific content and purpose of the dataset are not described.