dqnguyen commited on
Commit
6747716
1 Parent(s): dc5f5fd

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +0 -54
README.md CHANGED
@@ -27,57 +27,3 @@ For further information or requests, please go to [BERTweet's homepage](https://
27
  <img width="275" alt="irony" src="https://user-images.githubusercontent.com/2412555/135724595-15f4f2c8-bbb6-4ee6-82a0-034769dec183.png" />
28
  </p>
29
 
30
- ### <a name="models2"></a> Pre-trained models
31
-
32
-
33
- Model | #params | Arch. | Pre-training data
34
- ---|---|---|---
35
- `vinai/bertweet-base` | 135M | base | 850M English Tweets (cased)
36
- `vinai/bertweet-covid19-base-cased` | 135M | base | 23M COVID-19 English Tweets (cased)
37
- `vinai/bertweet-covid19-base-uncased` | 135M | base | 23M COVID-19 English Tweets (uncased)
38
- `vinai/bertweet-large` | 355M | large | 873M English Tweets (cased)
39
-
40
-
41
- ### <a name="usage2"></a> Example usage
42
-
43
-
44
- ```python
45
- import torch
46
- from transformers import AutoModel, AutoTokenizer
47
-
48
- bertweet = AutoModel.from_pretrained("vinai/bertweet-large")
49
- tokenizer = AutoTokenizer.from_pretrained("vinai/bertweet-large")
50
-
51
- # INPUT TWEET IS ALREADY NORMALIZED!
52
- line = "SC has first two presumptive cases of coronavirus , DHEC confirms HTTPURL via @USER :cry:"
53
-
54
- input_ids = torch.tensor([tokenizer.encode(line)])
55
-
56
- with torch.no_grad():
57
- features = bertweet(input_ids) # Models outputs are now tuples
58
-
59
- ## With TensorFlow 2.0+:
60
- # from transformers import TFAutoModel
61
- # bertweet = TFAutoModel.from_pretrained("vinai/bertweet-large")
62
- ```
63
-
64
- ### <a name="preprocess"></a> Normalize raw input Tweets
65
-
66
- Before applying BPE to the pre-training corpus of English Tweets, we tokenized these Tweets using `TweetTokenizer` from the NLTK toolkit and used the `emoji` package to translate emotion icons into text strings (here, each icon is referred to as a word token). We also normalized the Tweets by converting user mentions and web/url links into special tokens `@USER` and `HTTPURL`, respectively. Thus it is recommended to also apply the same pre-processing step for BERTweet-based downstream applications w.r.t. the raw input Tweets.
67
-
68
- For `vinai/bertweet-large`, given the raw input Tweets, to obtain the same pre-processing output, users could employ our [TweetNormalizer](https://github.com/VinAIResearch/BERTweet/blob/master/TweetNormalizer.py) module.
69
-
70
- - Installation: `pip3 install nltk emoji==0.6.0`
71
- - The `emoji` version must be either 0.5.4 or 0.6.0. Newer `emoji` versions have been updated to newer versions of the Emoji Charts, thus not consistent with the one used for pre-processing our pre-training Tweet corpus.
72
-
73
- ```python
74
- import torch
75
- from transformers import AutoTokenizer
76
- from TweetNormalizer import normalizeTweet
77
-
78
- tokenizer = AutoTokenizer.from_pretrained("vinai/bertweet-large")
79
-
80
- line = normalizeTweet("DHEC confirms https://postandcourier.com/health/covid19/sc-has-first-two-presumptive-cases-of-coronavirus-dhec-confirms/article_bddfe4ae-5fd3-11ea-9ce4-5f495366cee6.html?utm_medium=social&utm_source=twitter&utm_campaign=user-share… via @postandcourier 😢")
81
-
82
- input_ids = torch.tensor([tokenizer.encode(line)])
83
- ```
 
27
  <img width="275" alt="irony" src="https://user-images.githubusercontent.com/2412555/135724595-15f4f2c8-bbb6-4ee6-82a0-034769dec183.png" />
28
  </p>
29