Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,26 @@
|
|
1 |
-
|
2 |
-
|
3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# FinBERT-FOMC
|
2 |
+
FinBERT-FOMC is a FinBERT model fine-tuned on the data used FOMC minutes 2006.1 to 2023.2 with 3535 relabel complex sentences.
|
3 |
+
**Input:**
|
4 |
+
A financial text.
|
5 |
+
|
6 |
+
**Output:**
|
7 |
+
Positive, Negative, Neutral
|
8 |
+
|
9 |
+
# How to use
|
10 |
+
You can use this model with Transformers pipeline for FinBERT-FOMC.
|
11 |
+
|
12 |
+
```bash
|
13 |
+
from transformers import BertTokenizer, BertForSequenceClassification, pipeline
|
14 |
+
|
15 |
+
finbert = BertForSequenceClassification.from_pretrained('ZiweiChen/FinBERT-FOMC',num_labels=3)
|
16 |
+
tokenizer = BertTokenizer.from_pretrained('ZiweiChen/FinBERT-FOMC')
|
17 |
+
finbert_fomc = pipeline("text-classification", model=finbert, tokenizer=tokenizer)
|
18 |
+
|
19 |
+
sentences = ["Spending on cars and light trucks increased somewhat in July after a lackluster pace in the second quarter but apparently weakened in August"]
|
20 |
+
results = finbert_fomc(sentences)
|
21 |
+
print(results)
|
22 |
+
# [{'label': 'Negative', 'score': 0.994509756565094}]
|
23 |
+
|
24 |
+
```
|
25 |
+
Visit https://github.com/Incredible88/FinBERT-FOMC for more details
|
26 |
+
|