-
I run the source code both on GPU (A100) and CPU, but I got the result like below.
It's even slower with the proposed algorithms.
My device info:
CPU
processor : 124
vendor_id : GenuineIntel
c…
-
When I try to reproduce the demo I run into problems with java, I tried jre-17 and jre-19 but it doesn't seem to be a problem with the java version.
Is there any good solution for this please?
`…
-
我看到代码里有
```
def form_ngram(self, input_tensor, n=2)
def torch_bleu(self, ref_tensor, sys_tensor, pad_id, n_gram=2):
```
这俩个函数来计算bleu score.请问为什么不使用nltk这种可以计算bleu的三方库而选择自己实现呢?请问下主要的考量是什么?
-
Hi,
Whether for T5 or any seq2seq NMT model, does quantization work the same way for the encoder ?
Thanks.
-
Create table for words and nouns. Query it when generating usernames
Reference: https://stackoverflow.com/questions/8674718/best-way-to-select-random-rows-postgresql
-
Soit le code de base :
```latex
\documentclass[11pt]{article}
\usepackage{Scrabble}
\begin{document}
\PlateauScrabble[Echelle=0.55,Cadre=false]
\end{document}
```
Ceci produit la grille:
…
-
Le calcul se fait aujourd'hui seulement sur les 4 mairies adjacentes, pour ne pas faire exploser la combinatoire du calcul. Il faudrait l'augmenter pour 6 au moins je pense, pour éviter des trous, par…
laem updated
8 months ago
-
I have a question about the metrics that are defined on HELM. Currently, there are around 59 metrics and those are:
All the metrics currently provided by HELM
**Accuracy**
- none
- Quasi-ex…
-
### System Info
- `transformers` version: 4.23.1
- Platform: Windows-10-10.0.19044-SP0
- Python version: 3.8.10
- Huggingface_hub version: 0.10.1
- PyTorch version (GPU?): 1.10.1+cu111 (True)
- …
-
**Problem Description:**
I believe that users are looking for basic metrics related to annotators.
**Solution Proposal:**
*Metrics for Annotator Suggestions vs. Responses*
We should be able to obt…