-
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…
-
For the range of LIME method parameters defined in issue dianna-ai/dianna-exploration#173
Use 2 dianna [dashboard](https://github.com/dianna-ai/dianna/tree/main/dashboard) instances to compare pai…
-
Hi!
Would it be possible to get the checkpoints of the pre-trained RetroMAE models? In particular, the model after this pre-training stage described here https://github.com/FlagOpen/FlagEmbedding/b…
-
can't run this example on jax or pytorch backend
it just works on tensorflow backend
https://keras.io/examples/nlp/neural_machine_translation_with_keras_nlp/
also inferencing is significantly s…
-
Hello! I am a fresh man in the field of LLMs. I am reading your code and I have a question about the evaluation of CrowS-Pairs. In https://github.com/RUCAIBox/LLMSurvey/blob/4c324d19683901f0fc2c5eb464…
-
## 0. Paper
[Machine Translation Evaluation with BERT Regressor](https://arxiv.org/abs/1907.12679)
## 1. What is it?
They proposed a method to evaluate (output, reference) sentences with BERT.
…
a1da4 updated
4 years ago
-
I went through the loss code, and couldn't quite figure out how the number of negatives are selected.
I am assuming all other sentences (other than the paired one) is selected as a negative label?
…
-
Hi!
Thank you very much for this package. Very helpful!
I have a question. Do you think it is possible to implement this in a way that distinguishes between 'positive' and 'negative' sentiment a…
-
thanks for sharing:
in the readme you say "Data: 200m chinese internet question answering pairs. Vocab: 52777, jieba CWS enhanced with forward maximum matching."
so you train process is just pair of…
-
I have a list of text pairs with a label of 0/1 indicating whether two sentences are paraphrases.
There are two evaluation metrics I find suitable for this task:
1) BinaryClassificationEvaluator, …