-
请问你这里的bert预训练模型用的哪里的,我用的huggingface的bert-base-chinese,不但训练效果差,而且代码也一直报vocab的索引对不上的问题。
glMa7 updated
8 months ago
-
![image](https://user-images.githubusercontent.com/70731194/177521512-91a75186-415e-4f7f-80c4-af01ff5bd255.png)
预训练模型:https://huggingface.co/uer/chinese_roberta_L-12_H-768
训练命令:
python finetune/…
-
训练集使用https://github.com/zjy-ucas/ChineseNER
**训练命令**
```
bert-base-ner-train \
-data_dir=/home/bert/BERT-BiLSTM-CRF-NER/data/ \
-bert_config_file=/home/bert/chinese_L-12_H-768_A-12/bert_config.…
-
Because the chinese pretrained vocab does not include all the english words, so I split english words into characters. Then how do I represent whitespace between english words?
-
AttributeError: 'NoneType' object has no attribute 'tokenize'
求救
-
python finetune/run_classifier.py --pretrained_model_path models/roberta-base-finetuned-dianping-chinese/pytorch_model.bin \
--vocab_path models/google_zh_vocab.txt…
-
hi大佬
[gpt2-chinese](https://huggingface.co/uer/gpt2-chinese-cluecorpussmall),我看了,但是没有相关微调的代码,[在此](https://huggingface.co/uer/gpt2-chinese-cluecorpussmall)看到了如下代码,但是数据格式没有告知,也不知道是不是纯文本,
```
python…
-
Hi. I have a phoneme-based Zipformer model.
Before this [PR](https://github.com/k2-fsa/sherpa-onnx/pull/828), I was able to apply hotwords encoding for phoneme sequences, e.g. `ɪ z/dʒ ʌ s t/b ɛ s t…
w11wo updated
3 months ago
-
您好,我想试试微调和训练你们的Chinese GPT2 Lyric Model,但是发现没有说明corpora/lyric.txt的数据格式,请问怎么把自己下载的歌词处理成corpora/lyric.txt需要的格式呢?
```
python3 preprocess.py --corpus_path corpora/lyric.txt \
--…
-
联通的兄弟,在ollama的模型仓库上传一下,或者发一下ollama的modelfile