-
I am using the summarization code provided in example/pytorch/summarization/run_summarization.py
However, I could not add the argument "repetition_penalty" when generating.
After tracing the sou…
-
For the first assignment, find examples of information design (schedules, signage/way finding, campaign/advocacy, product, space/architecture etc.) and learn from them.
Each of you will pick three …
-
## Environment info
- `transformers` version: 4.6.0
- Platform:
- Python version:
- PyTorch version (GPU?): 1.7.1 cuda 9.2
- Tensorflow version (GPU?):
- Using GPU in script?:
- Using dist…
-
## 🐛 Bug
The mask token id of BART is different between fairseq (torch.hub) and huggingface, and this discrepancy leads to different results in mask_filling. So I wonder which token id is actually…
-
# 🚀 Feature request
Add support for exporting summarization models to ONNX.
## Motivation
I want to serve summarization models on edge, through an ONNX runtime. However, I am unable to convert …
-
Why there is no training example for T5 or MT5???
Could you please give me a link to an example? I had a hard time to write a code with various errors:
This is my code:
```
import torch
from dat…
-
# 🐛 Bug
## Information
Model I am using (Bert, XLNet ...): Bart
Language I am using the model on (English, Chinese ...): English
## To reproduce
Steps to reproduce the behavior:
```
…
-
I know that multi-GPU training is handled by the trainer class automatically through `CUDA_VISIBLE_DEVICES=...` flag in transformers. But, I'm having a weird problem. Like, after setting `CUDA_VISIBLE…
-
### Who can help
Models:
- bart, t5: @patrickvonplaten, @patil-suraj
## Information
I am using t5-large and t5-base to train my customer model with my customer csv dataset through runn…
-
## Environment info
- `transformers` version: 4.9.0.dev0
- Platform: Linux-5.4.0-1043-gcp-x86_64-with-glibc2.29
- Python version: 3.8.10
- Flax version (CPU?/GPU?/TPU?): 0.3.4 (tpu)
- Jax version…