Migrate model card from transformers-repo
Browse filesRead announcement at https://discuss.huggingface.co/t/announcement-all-model-cards-will-be-migrated-to-hf-co-model-repos/2755
Original file history: https://github.com/huggingface/transformers/commits/master/model_cards/ganeshkharad/gk-hinglish-sentiment/README.md
README.md
ADDED
@@ -0,0 +1,72 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- hi-en
|
4 |
+
|
5 |
+
tags:
|
6 |
+
- sentiment
|
7 |
+
- multilingual
|
8 |
+
- hindi codemix
|
9 |
+
- hinglish
|
10 |
+
license: apache-2.0
|
11 |
+
datasets:
|
12 |
+
- sail
|
13 |
+
---
|
14 |
+
|
15 |
+
# Sentiment Classification for hinglish text: `gk-hinglish-sentiment`
|
16 |
+
|
17 |
+
## Model description
|
18 |
+
|
19 |
+
Trained small amount of reviews dataset
|
20 |
+
|
21 |
+
## Intended uses & limitations
|
22 |
+
|
23 |
+
I wanted something to work well with hinglish data as it is being used in India mostly.
|
24 |
+
The training data was not much as expected
|
25 |
+
|
26 |
+
#### How to use
|
27 |
+
|
28 |
+
```python
|
29 |
+
#sample code
|
30 |
+
from transformers import BertTokenizer, BertForSequenceClassification
|
31 |
+
tokenizerg = BertTokenizer.from_pretrained("/content/model")
|
32 |
+
modelg = BertForSequenceClassification.from_pretrained("/content/model")
|
33 |
+
|
34 |
+
text = "kuch bhi type karo hinglish mai"
|
35 |
+
encoded_input = tokenizerg(text, return_tensors='pt')
|
36 |
+
output = modelg(**encoded_input)
|
37 |
+
print(output)
|
38 |
+
#output contains 3 lables LABEL_0 = Negative ,LABEL_1 = Nuetral ,LABEL_2 = Positive
|
39 |
+
```
|
40 |
+
|
41 |
+
#### Limitations and bias
|
42 |
+
|
43 |
+
The data contains only hinglish codemixed text it and was very much limited may be I will Update this model if I can get good amount of data
|
44 |
+
|
45 |
+
## Training data
|
46 |
+
|
47 |
+
Training data contains labeled data for 3 labels
|
48 |
+
|
49 |
+
link to the pre-trained model card with description of the pre-training data.
|
50 |
+
I have Tuned below model
|
51 |
+
|
52 |
+
https://huggingface.co/rohanrajpal/bert-base-multilingual-codemixed-cased-sentiment
|
53 |
+
|
54 |
+
|
55 |
+
### BibTeX entry and citation info
|
56 |
+
|
57 |
+
```@inproceedings{khanuja-etal-2020-gluecos,
|
58 |
+
title = "{GLUEC}o{S}: An Evaluation Benchmark for Code-Switched {NLP}",
|
59 |
+
author = "Khanuja, Simran and
|
60 |
+
Dandapat, Sandipan and
|
61 |
+
Srinivasan, Anirudh and
|
62 |
+
Sitaram, Sunayana and
|
63 |
+
Choudhury, Monojit",
|
64 |
+
booktitle = "Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics",
|
65 |
+
month = jul,
|
66 |
+
year = "2020",
|
67 |
+
address = "Online",
|
68 |
+
publisher = "Association for Computational Linguistics",
|
69 |
+
url = "https://www.aclweb.org/anthology/2020.acl-main.329",
|
70 |
+
pages = "3575--3585"
|
71 |
+
}
|
72 |
+
```
|