一 背景
huggingface相关环境的安装和问题处理本篇暂不涉及,后续补充。这里以一个模型为例,完成从模型介绍到加载、运行的完整过程,作为我们熟悉huggingface的一个示例。
二 模型
这里选择google/pegasus-newsroom模型作为示例。
2.1 介绍
模型介绍参见https://huggingface.co/docs/transformers/main/model_doc/pegasus,模型是在论文《PEGASUS: Pre-training with Extracted Gap-sentences forAbstractive Summarization》中提出的,作者:Jingqing Zhang。基本思想是,PEGASUS在预训练阶段,将输入的文档的重要句子remove/mask,通过其它的句子预测生成,类似于摘要生成的做法。
2.2 使用示例
https://huggingface.co/google/pegasus-newsroom/tree/main
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("google/pegasus-newsroom") model = AutoModelForSeq2SeqLM.from_pretrained("google/pegasus-newsroom")
2.3 遇到问题
按理说应该可以顺利执行,但实际上不出意外地遇到了意外。在执行时还是报错提示无法加载模型,信息如下:
Traceback (most recent call last): File "<stdin>", line 1, in <module> File "/root/onnx/tutorial-env/lib/python3.10/site-packages/transformers/models/auto/tokenization_auto.py", line 709, in from_pretrained return tokenizer_class_fast.from_pretrained(pretrained_model_name_or_path, *inputs, **kwargs) File "/root/onnx/tutorial-env/lib/python3.10/site-packages/transformers/tokenization_utils_base.py", line 1809, in from_pretrained raise EnvironmentError( OSError: Can't load tokenizer for 'google/pegasus-newsroom'. If you were trying to load it from 'https://huggingface.co/models', make sure you don't have a local directory with the same name. Otherwise, make sure 'google/pegasus-newsroom' is the correct path to a directory containing all relevant files for a PegasusTokenizerFast tokenizer.
在最后一行,OSError这段,给出了两种错误可能的提示:
(1)确保本地没有同名目录
这一点显然,从来都没有创建过这个目录;
(2)确认'google/pegasus-newsroom'是一个包含所有相关文件的正确目录
这是从huggingface官网上复制过来的代码,不可能会出错。那问题出在哪里了?
三 问题排查
3.1 SSH拉取模型文件
通过资料搜搜,和huggingface官网的模型页面查看,发现如下:
可以通过git拉取模型文件
不过执行后有如下报错:
所以改为使用SSH方式:
报了权限错误,不过还好,看到publickey的提示,应该是设置一下访问授权就可以了。
git clone git@hf.co:google/pegasus-newsroom 正克隆到 'pegasus-newsroom'... Warning: Permanently added the ECDSA host key for IP address '3.210.66.237' to the list of known hosts. Permission denied (publickey). fatal: Could not read from remote repository. Please make sure you have the correct access rights and the repository exists.
3.2 ssh key生成与添加
在https://huggingface.co/docs/hub/security-git-ssh 中有相关的操作描述(当然在实际操作中发现也有坑。。。),简单整理如下:
1、检查是否存在SSH key,由于是linux系统,所以默认是在~/.ssh目录下。由于我们之前没有生成过,所以没有(有也没关系,直接覆盖生成就好)
- id_rsa.pub
- id_ecdsa.pub
- id_ed25519.pub
2、如果没有,那么先生成,使用ssh-keygen命令,引号内是你注册huggingface时使用的邮箱:
ssh-keygen -t ed25519 -C "your.email@example.co"
3、生成完毕后,使用ssh-add命令加入到你的SSH agent中:
ssh-add ~/.ssh/id_ed25519
在第三步可能会遇到报错,例如我本地执行时错误如下:
Could not open a connection to your authentication agent.
无法正常添加,这种情况需要先执行ssh-agent bash,然后再次执行ssh-add 添加即可。
接下来就可以拉模型文件了:
git clone git@hf.co:google/pegasus-newsroom 正克隆到 'pegasus-newsroom'... remote: Enumerating objects: 33, done. remote: Total 33 (delta 0), reused 0 (delta 0), pack-reused 33 接收对象中: 100% (33/33), 931.72 KiB | 604.00 KiB/s, done. 处理 delta 中: 100% (12/12), done. Downloading pytorch_model.bin (2.3 GB)
下载成功。
不过跟huggingface的描述相比,还有有个地方有些问题。按照huggingface的文档描述,ssh-add 添加id_ed25519成功后,在终端执行ssh -T git@hf.co 命令,应该能看到包含你用户名的提示信息。但如上所述,我已经成功添加,并且可以拉取模型文件了,在终端执行命令后还是只有: “Hi anonymous, welcome to Hugging Face.”,按照文档描述这应该是失败的状态。这里暂时没有解决,留待后续继续排查。
四 继续运行模型
4.1 网络问题
回过头来,我们继续尝试对google/pegasus-newsroom的尝试。依次执行命令如下:
from transformers import AutoTokenizer, PegasusModel tokenizer = AutoTokenizer.from_pretrained("google/pegasus-large") model = PegasusModel.from_pretrained("google/pegasus-large") inputs = tokenizer("Studies have been shown that owning a dog is good for you", return_tensors="pt") decoder_inputs = tokenizer("Studies show that", return_tensors="pt") outputs = model(input_ids=inputs.input_ids, decoder_input_ids=decoder_inputs.input_ids) last_hidden_states = outputs.last_hidden_state list(last_hidden_states.shape)
执行成功。
不过我们重复执行时,发现这里还有个问题,执行:model = PegasusModel.from_pretrained("google/pegasus-large") 时,依然会报连接失败的错误,而且失败的概率还比较大,所以依然需要继续解决。不过这个稍微分析一下,就知道是国内众所周知的“网络环境”问题,如果可以“科学上网”,那么就可以解决。不过相信也有很多小伙伴不具备这样的环境,或者风险较大,所以需要考虑采用其他更合法的方式。
4.2 离线模式
官网和其他可搜到的资料,基本都推荐采用离线模式。也就是把模型通过git或者手工下载再上传到服务器的指定目录,然后修改执行脚本从本地加载的方式。
由于上面我们已经完成了ssh的配置,并且可以git clone拉取模型文件,所以就直接加载已经拉下来的模型,脚本如下:
>>> from transformers import AutoTokenizer, AutoModelForMaskedLM >>> tokenizer = AutoTokenizer.from_pretrained("/root/onnx/model/huggingface/pegasus-newsroom") >>> >>> from transformers import AutoTokenizer, AutoModel >>> model = AutoModel.from_pretrained("/root/onnx/model/huggingface/pegasus-newsroom") Some weights of the model checkpoint at /root/onnx/model/huggingface/pegasus-newsroom were not used when initializing PegasusModel: ['final_logits_bias'] - This IS expected if you are initializing PegasusModel from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model). - This IS NOT expected if you are initializing PegasusModel from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). Some weights of PegasusModel were not initialized from the model checkpoint at /root/onnx/model/huggingface/pegasus-newsroom and are newly initialized: ['model.encoder.embed_positions.weight', 'model.decoder.embed_positions.weight'] You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference. >>> >>> model = model.eval()
到这里,算是跑通了整个运行流程。
五 后续
接下来,将继续验证huggingface转onnx,和加载onnx并对外提供服务。