-
I'm trying to fine tune a pre-trained wav2vec 2.0 base model (no finetuning) on ~1h of data, with the following command:
python3 fairseq/train.py --distributed-world-size 1 \
"path/to/data" \
-…
-
Resources are user-contributed, which is great and all, but reading unfamiliar games/warm-ups has left me with many questions about what the game is, especially if it isn't explained well.
Includin…
-
Hey, could you add support for erhua. Combinations such as 事儿 = shìr are not handled. Even in standard Chinese (news etc.), erhua is often heard.
-
With version 1.63, after appending the library path to the system, I could import the `DenseSolver` via
```python
from org.openlca.eigen.solvers import DenseSolver
```
As of version 1.7.0 there is…
-
不好意思，想请教您一下
有一个huggingface的pretrained model, 我在跑您的模型之前，已经用convert_bert_from_huggingface_to_uer的script做了转换，但每次都是报这么个错误，研究很久，实在没明白为什么。。
中文和英文的pretrained model我都试过，都是从google官网上重新下载的，但都是提示相同的错误。
就想…
-
AttributeError: module 'torch.distributed' has no attribute 'init_process_group'
-
使用经过脚本转换后的huggingface上的mengzi-t5-base模型时报错：
```
RuntimeError: Error(s) in loading state_dict for Model:
size mismatch for embedding.word_embedding.weight: copying a param with shape torch.S…
-
## What
As part of [Understand what changes we will make to our processes to ship things we're unsure of ](https://github.com/alphagov/govuk-design-system/issues/4165) we have suggested some ideas fo…
-
### 🐛 bug 说明
你好，没有找到hfl/chinese-roberta-wwm-ext的small版，只找到了base和large，请问train_m3e.py是在这个roberta-wwm基础上训练得到的吗？
### Python Version
None
-
The following seems like a reasonable way to check for uniqueness of email for a given collection:
``` javascript
// 'unique:users:email'
Checkit.Validators.unique = function(val, collection, field) …
jclem updated
9 years ago