-
I am unable to set bos_token_id=0 for a new SentencePiece tokenizer (MBART).
Here is what I'm doing?
```bash
wget https://s3.amazonaws.com/models.huggingface.co/bert/facebook/mbart-large-en-ro/se…
-
# 🚀 Feature request
The Text Regression task aims at delivering scalar output for each input sequence, which is widely used in NLP applications.
In my own use, I'm working on the research to…
-
###Defined a class that inherits from RobertaPretrainedModel
This class is defined to get from_pretrained()
class XLMRobertaPreTrainedModel(RobertaPreTrainedModel):
"""
An …
-
### Problem
I'm getting the following error when I'm trying to apply **static quantization** (ONNX) with the `ORTQuantizer` .
### Tests
This error occurs for:
- my custom script
- the exam…
-
# 🚀 Feature request
Could we maybe have a consistent naming convention for speech models? So far we have:
- [`input_features`](https://huggingface.co/transformers/model_doc/speech_to_text.html#spe…
-
When I try to evaluate during the training the following warning throws:
**NaN or Inf found in input tensor**
I have been using XLM-Roberta model. These are arguments I have used:
```
model_type…
-
**Describe the bug**
I'm trying to limit the number of CPU processes that will be used by a `ClassificationModel` for inference by providing the `process_count` argument (as described [here](https:/…
-
In the wiki page for XLM-R
https://huggingface.co/transformers/model_doc/xlmroberta.html
the model name should be 'xlm-roberta-base' instead of 'roberta-base'
@sgugger
-
## 🐛 Bug
Model I am using (Bert, XLNet....): XLMRoberta
Language I am using the model on (English, Chinese....): multi-language, but mostly english
The problem arise when:
try to token…
-
## Environment info
-`adapter-transformers` version: 2.1.2
- Platform: Linux-5.8.0-1041-azure-x86_64-with-glibc2.29
- Python version: 3.8.10
- PyTorch version (GPU?): 1.8.1+cu102 (True)
…