Closed cuppersd closed 1 month ago
No response
- OS: - Python: - Transformers: - PyTorch: - CUDA (`python -c 'import torch; print(torch.version.cuda)'`):
你好,你可以调用 chat 函数时传入 sampling=False, 这会使用默认 num_beams=3的 beam search 解码, 或者你可以每次调用前通过torch.cuda.manual_seed_all 方法设定随机数种子, 也可以复现结果
sampling=False
num_beams=3
torch.cuda.manual_seed_all
是否已有关于该错误的issue或讨论? | Is there an existing issue / discussion for this?
该问题是否在FAQ中有解答? | Is there an existing answer for this in FAQ?
当前行为 | Current Behavior
No response
期望行为 | Expected Behavior
No response
复现方法 | Steps To Reproduce
No response
运行环境 | Environment
备注 | Anything else?
No response