wangyuxinwhy / uniem

unified embedding model
Apache License 2.0
814 stars 61 forks source link

No sentence-transformers model found #100

Open tfka opened 12 months ago

tfka commented 12 months ago

🐛 bug 说明

No sentence-transformers model found with name ./finetuned-model/model. Creating a new one with MEAN pooling. 基于m3e finetune模型但是得到这样的提示,是我的训练代码设置问题吗?

df = pd.read_json('train_m3e.jsonl', lines=True)
dataset = Dataset.from_pandas(df)
dataset = dataset.train_test_split(test_size=0.1, seed=42)
dataset['validation'] = dataset.pop('test')
# 指定训练的模型为 m3e-small
finetuner = FineTuner.from_pretrained('moka-ai/m3e-large', dataset=dataset)
fintuned_model = finetuner.run(epochs=5)

Python Version

3.10

ChinesePainting commented 11 months ago

一样的问题,解决了吗

wangyuxinwhy commented 11 months ago

升级 uniem 试试,我这边没有办法复现这个 bug

python -m pip install -U uniem