108 Commits

Include renames

Author	SHA1	Message	Date
ViperEkura	0a03a15679	test(trainer): 调整测试参数以提高训练和断言的稳定性	2025-10-02 14:34:02 +08:00
ViperEkura	17f1a12f27	test(trainer): 增强测试用例以支持随机配置和多轮对话训练	2025-09-30 16:33:37 +08:00
ViperEkura	e467420475	fix(khaosz/trainer): 修正训练步数计算逻辑	2025-09-29 19:05:26 +08:00
ViperEkura	0ebf53008e	refactor(test): 更新训练配置参数名称并优化测试逻辑	2025-09-28 22:14:39 +08:00
ViperEkura	0b96b11a6e	test(trainer): 增加训练中断与检查点恢复测试	2025-09-28 14:38:23 +08:00
ViperEkura	c8a38743a4	fix(tests): 更新测试代码以验证优化器状态的保存与加载	2025-09-28 14:00:38 +08:00
ViperEkura	4fcdc87c95	feat(trainer): 重构数据集与策略模块以支持字典形式的数据返回	2025-09-27 14:11:27 +08:00
ViperEkura	a4443765ee	Initial commit	2025-09-27 12:02:22 +08:00