-
This option would enable to re-rank nbest hypotheses generated with nucleus sampling.
-
I am using ollama with llama3.1 model and using a tool to draft email responses. The input passed to the tool is has a string with \n character which is a valid string in python. but I am getting a cu…
-
After running the code given following the steps mentioned, getting this error in the test.out.nbest.logs
```
[2023-02-11 11:35:41] Vocabulary file model.restricted/vocab.spm must not contain empt…
-
## ❓ Questions and Help
### Before asking:
1. search the issues.
2. search the docs.
#### What is your question?
I want to generate topK translated sentences. I found in the documents tha…
-
本人使用 SoftCorrect 对 wenet 的 CTC 多候选序列进行纠错时遇到以下问题:
```
test checkpoints/detector_finetune/checkpoint16.pt
infile_list: ['test']
Setting word embedding dropout of encoder to 0.0
Decoder not exit! Us…
-
[Guys, I have gym now so I'll submit this and write the rest of this later today. ]
I am creating an issue to describe a plan for multi-pass n-best-list rescoring. This will also require
new cod…
-
I would like to sample from my decoder.
The docs say I can:
But I get an error:
```bash
usage: translate.py [-h] [--config CONFIG] [--input INPUT] [--input-factors INPUT_FACTORS [INPUT_FACTO…
-
Hi,
Thanks for the work. I am trying to use the pre-trained model, but I don't know how to get the decoding score for the corresponding decoding results.
```
nbests = speech2text(speech)
tex…
-
- INFO - metrics - Writing predictions to: ./predictions_test_./du2021/384_bert/.json
Traceback (most recent call last):
File "main_bert.py", line 1036, in
main()
File "main_bert.py", l…
-
It's slowly crunching through forward translation on nanna
```
for i in /mnt/nanna0/nbogoych/data/data/fr-en/fr-en-prod/translated/corpus/file.*.ref; do f=$(dirname $i)/$(basename $i .ref).nbest; if…