按照README上面的实例运行报错

[sxk000]

感谢大佬分享！

我在按照README上面的实例：

加载模型

from transformers import T5Tokenizer, T5ForConditionalGeneration
tokenizer = T5Tokenizer.from_pretrained("ClueAI/ChatYuan-large-v1")
model = T5ForConditionalGeneration.from_pretrained("ClueAI/ChatYuan-large-v1")

使用

import torch
from transformers import AutoTokenizer

修改colab笔记本设置为gpu，推理更快

device = torch.device('cuda')
model.to(device)
def preprocess(text):
text = text.replace("\n", "\n").replace("\t", "\t")
return text

def postprocess(text):
return text.replace("\n", "\n").replace("\t", "\t")

def answer(text, sample=True, top_p=1, temperature=0.7):
'''sample：是否抽样。生成任务，可以设置为True;
top_p：0-1之间，生成的内容越多样'''
text = preprocess(text)
encoding = tokenizer(text=[text], truncation=True, padding=True, max_length=768, return_tensors="pt").to(device)
if not sample:
out = model.generate(**encoding, return_dict_in_generate=True, output_scores=False, max_new_tokens=512, num_beams=1, length_penalty=0.6)
else:
out = model.generate(**encoding, return_dict_in_generate=True, output_scores=False, max_new_tokens=512, do_sample=True, top_p=top_p, temperature=temperature, no_repeat_ngram_size=3)
out_text = tokenizer.batch_decode(out["sequences"], skip_special_tokens=True)
return postprocess(out_text[0])
print("end...")

input_text0 = "帮我写一个请假条，我因为新冠不舒服，需要请假3天，请领导批准"
input_text1 = "你能干什么"
input_text2 = "写一封英文商务邮件给英国客户，表达因为物流延误，不能如期到达，我们可以赔偿贵公司所有损失"
input_text3 = "写一个文章，题目是未来城市"
input_text4 = "写一个诗歌，关于冬天"
input_text5 = "从南京到上海的路线"
input_text6 = "学前教育专业岗位实习中，在学生方面会存在问题，请提出改进措施。800字"
input_text7 = "根据标题生成文章：标题：屈臣氏里的化妆品到底怎么样？正文：化妆品，要讲究科学运用，合理搭配。屈臣氏起码是正品连锁店。请继续后面的文字。"
input_text8 = "帮我对比几款GPU，列出详细参数对比，并且给出最终结论"
input_list = [input_text0, input_text1, input_text2, input_text3, input_text4, input_text5, input_text6, input_text7, input_text8]
for i, input_text in enumerate(input_list):
input_text = "用户：" + input_text + "\n小元："
print(f"示例{i}".center(50, "="))
output_text = answer(input_text)
print(f"{input_text}{output_text}")

报如下错误：

=======================示例0========================
Traceback (most recent call last):
File "", line 4, in
File "", line 9, in answer
File "/home/anaconda3/envs/s20230220e310fsb/lib/python3.10/site-packages/torch/autograd/grad_mode.py", line 27, in decorate_context
return func(*args, **kwargs)
File "/home/anaconda3/envs/s20230220e310fsb/lib/python3.10/site-packages/transformers/generation/utils.py", line 1252, in generate
model_kwargs = self._prepare_encoder_decoder_kwargs_for_generation(
File "/home/anaconda3/envs/s20230220e310fsb/lib/python3.10/site-packages/transformers/generation/utils.py", line 617, in _prepare_encoder_decoder_kwargs_for_generation
model_kwargs["encoder_outputs"]: ModelOutput = encoder(**encoder_kwargs)
File "/home/anaconda3/envs/s20230220e310fsb/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl
return forward_call(*input, **kwargs)
File "/home/anaconda3/envs/s20230220e310fsb/lib/python3.10/site-packages/transformers/models/t5/modeling_t5.py", line 1055, in forward
layer_outputs = layer_module(
File "/home/anaconda3/envs/s20230220e310fsb/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl
return forward_call(*input, **kwargs)
File "/home/anaconda3/envs/s20230220e310fsb/lib/python3.10/site-packages/transformers/models/t5/modeling_t5.py", line 687, in forward
self_attention_outputs = self.layer[0](
File "/home/anaconda3/envs/s20230220e310fsb/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl
return forward_call(*input, **kwargs)
File "/home/anaconda3/envs/s20230220e310fsb/lib/python3.10/site-packages/transformers/models/t5/modeling_t5.py", line 593, in forward
attention_output = self.SelfAttention(
File "/home/anaconda3/envs/s20230220e310fsb/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl
return forward_call(*input, **kwargs)
File "/home/anaconda3/envs/s20230220e310fsb/lib/python3.10/site-packages/transformers/models/t5/modeling_t5.py", line 512, in forward
query_states = shape(self.q(hidden_states)) # (batch_size, n_heads, seq_length, dim_per_head)
File "/home/anaconda3/envs/s20230220e310fsb/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1194, in _call_impl
return forward_call(*input, **kwargs)
File "/home/anaconda3/envs/s20230220e310fsb/lib/python3.10/site-packages/torch/nn/modules/linear.py", line 114, in forward
return F.linear(input, self.weight, self.bias)
RuntimeError: CUDA error: CUBLAS_STATUS_INVALID_VALUE when calling cublasSgemm( handle, opa, opb, m, n, k, &alpha, a, lda, b, ldb, &beta, c, ldc)

[joytianya]

看着像是环境配置问题，可以先用colab跑下试试

[sxk000]

看着像是环境配置问题，可以先用colab跑下试试

是的，重新安装了pytorch，可以了，谢谢你