-
Hi , I trained sbert with triplet loss with default distance_metric --> Euclidian distance got descent embeddings and cosine score between relevant/irrelevant sentences. But when i tried with TripletD…
-
Hello,
I would like to test this promising framework on a similarity classification task. So basically, I have got a dataset with 3 columns: (sentence1,sentence2,label). From what I understand, curre…
-
i tend to work more by sentence than paragraph, and sentence highlighting is really useful to ensure you don't ever miss a sentence as your focus switches back and forth between source and target.
i …
-
Great work!
However, I can't see anything about the datasets. Could you please either leave them here or leave links of how to get them? Also, what would be the licenses of these datasets?
-
https://simonwillison.net/2023/Oct/23/embeddings/
-
The #447 experimentations provide me with an idea.
Like me, you certainly often see that NMT are loosing some words in the translations.
Here is an example of the involved internal mechanism:
`…
-
There is no doubt that this work is very powerful and great. And I also successfully implemented the Chinese to English transfer operation. My question is that the text content in the titles is too sm…
-
After running IBM1 Translation model in the given code, I've summed the translation probabilites (the probability for a target word given source word), and it doesn't sum to 1:
```python
from nltk…
-
Feature description
---------
Thanks for the hard work! The newest version actually solve this problem by showing the key-value pairs in the right side. Thus we could put some context in the right s…
-
Hi, I don't know if this was intended (I doubt it), but apparently this package uses the same json to store both user credentials (email and password) and preconfigured words/answers pairs. This means…