StephanAkkerman commited on
Commit
9872c2a
1 Parent(s): 8746881

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +71 -3
README.md CHANGED
@@ -9,13 +9,81 @@ metrics:
9
  - f1
10
  pipeline_tag: text-classification
11
  tags:
12
- - finance
13
  - sentiment
 
14
  - sentiment-analysis
15
- - tweets
16
  - twitter
 
 
 
 
 
17
  base_model: StephanAkkerman/FinTwitBERT
18
  ---
19
 
20
  # FinTwitBERT-sentiment
21
- FinTwitBERT-sentiment is a finetuned model for classifying the sentiment of financial tweets.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9
  - f1
10
  pipeline_tag: text-classification
11
  tags:
 
12
  - sentiment
13
+ - finance
14
  - sentiment-analysis
15
+ - financial-sentiment-analysis
16
  - twitter
17
+ - tweets
18
+ - stocks
19
+ - stock-market
20
+ - crypto
21
+ - cryptocurrency
22
  base_model: StephanAkkerman/FinTwitBERT
23
  ---
24
 
25
  # FinTwitBERT-sentiment
26
+
27
+ FinTwitBERT-sentiment is a finetuned model for classifying the sentiment of financial tweets. It uses [FinTwitBERT](https://huggingface.co/StephanAkkerman/FinTwitBERT) as a base model, which has been pre-trained on 1 million financial tweets.
28
+ This approach ensures that the FinTwitBERT-sentiment has seen enough financial tweets, which have an informal nature, compared to other financial texts, such as news headlines.
29
+ Therefore this model performs great on informal financial texts, seen on social media.
30
+
31
+ ## Intended Uses
32
+
33
+ FinTwitBERT-sentiment is intended for classifying financial tweets or other financial social media texts.
34
+
35
+ ## More Information
36
+
37
+ For a comprehensive overview, including the training setup and analysis of the model, visit the [FinTwitBERT GitHub repository](https://github.com/TimKoornstra/FinTwitBERT).
38
+
39
+ ## Usage
40
+
41
+ Using [HuggingFace's transformers library](https://huggingface.co/docs/transformers/index) the model and tokenizers can be converted into a pipeline for text classification.
42
+
43
+ ```python
44
+ from transformers import BertForSequenceClassification, AutoTokenizer, pipeline
45
+
46
+ model = BertForSequenceClassification.from_pretrained(
47
+ "StephanAkkerman/FinTwitBERT-sentiment",
48
+ num_labels=3,
49
+ id2label={0: "NEUTRAL", 1: "BULLISH", 2: "BEARISH"},
50
+ label2id={"NEUTRAL": 0, "BULLISH": 1, "BEARISH": 2},
51
+ )
52
+ model.config.problem_type = "single_label_classification"
53
+ tokenizer = AutoTokenizer.from_pretrained(
54
+ "StephanAkkerman/FinTwitBERT-sentiment"
55
+ )
56
+ model.eval()
57
+ pipeline = pipeline(
58
+ "text-classification", model=model, tokenizer=tokenizer
59
+ )
60
+
61
+ # Sentences we want the sentiment for
62
+ sentence = ["I love you"]
63
+
64
+ # Get the predicted sentiment
65
+ print(pipeline(sentence))
66
+ ```
67
+
68
+ ## Training
69
+
70
+ The model was trained with the following parameters:
71
+
72
+ ## Citing & Authors
73
+
74
+ If you use FinTwitBERT or FinTwitBERT-sentiment in your research, please cite us as follows, noting that both authors contributed equally to this work:
75
+
76
+ ```bibtex
77
+ @misc{FinTwitBERT,
78
+ author = {Stephan Akkerman, Tim Koornstra},
79
+ title = {FinTwitBERT: A Specialized Language Model for Financial Tweets},
80
+ year = {2023},
81
+ publisher = {GitHub},
82
+ journal = {GitHub repository},
83
+ howpublished = {\url{https://github.com/TimKoornstra/FinTwitBERT}}
84
+ }
85
+ ```
86
+
87
+ ## License
88
+
89
+ This project is licensed under the MIT License. See the [LICENSE](https://choosealicense.com/licenses/mit/) file for details.