-
I have a question about 'cls',
bert has NSP loss, so 'cls' output can express the sentence,
In tiny bert, 'cls' can express the sentence?
-
_franzm created the following on Feb 15:_
## Starting points
### mnlp-exp ColBERT repository in https://github.ibm.com/mnlp-exp/ColBERT
**Supports**
- additional representations (e.g. XLMR, …
-
_franzm created the following on Feb 15:_
## Starting points
### mnlp-exp ColBERT repository in https://github.ibm.com/mnlp-exp/ColBERT
**Supports**
- additional representations (e.g. XLMR, …
-
I have used BERT NextSentencePredictor to find similar sentences or similar news, However, It's super slow. Even on Tesla V100 which is the fastest GPU till now. It takes around 10secs for a query tit…
-
# New Features
## Prompt API
* Template String 新增支持关键字 `prefix`和`options`,新增 `position`, `token_type`, `length`, `encoder`, `hidden_size` 等7个属性 #3724
* 新增支持 PrefixTemplate
* 解除 `InputExample` …
-
Hi @prajjwal1,
Thank you very much for this great repo and work.
I would be interested in working with your bert-tiny model.
For this work, it would be of importance to know what data exactly …
-
The original papers mentioned: `` Specifically, let T denote a set of teacher layers that we use to distill knowledge to the student model.''
And the code in trainer provides ``[2, 5, 8, 11]'' only, …
-
# New Features
## Pipelines
+ Pipelines融入RocketQA轻量化小模型。https://github.com/PaddlePaddle/PaddleNLP/pull/3078
+ 新增基于ERNIE3.0的RocketQA召回模型。 https://github.com/PaddlePaddle/PaddleNLP/pull/3033
+ 新…
-
TypeError: forward() got an unexpected keyword argument 'passage_start_t'
a bunch of the parameters in "batch" also seem to throw the same error upon further investigation, such as "start_of_word"
I…
-
Hello, I would like to create a custom tokenizer, save it and be able to load it in some way, which would also work for pretrained tokenizers from Transformers library.
Right now I have this script:
…