Update README.md
Browse files
README.md
CHANGED
@@ -1 +1,26 @@
|
|
1 |
-
EconBERTa - RoBERTa further trained on 4GB of uncompressed text sourced from economics books
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
EconBERTa - RoBERTa further trained on 4GB of uncompressed text sourced from economics books.
|
2 |
+
|
3 |
+
Example usage for MLM:
|
4 |
+
|
5 |
+
from transformers import RobertaTokenizer, RobertaForMaskedLM
|
6 |
+
from transformers import pipeline
|
7 |
+
|
8 |
+
tokenizer = RobertaTokenizer.from_pretrained('roberta-base')
|
9 |
+
model = RobertaForMaskedLM.from_pretrained('models').cpu()
|
10 |
+
model.eval()
|
11 |
+
mlm = pipeline('fill-mask', model = model, tokenizer = tokenizer)
|
12 |
+
test = "ECB - euro, FED - <mask>, BoJ - yen"
|
13 |
+
print(mlm(test)[:2])
|
14 |
+
|
15 |
+
Will give:
|
16 |
+
|
17 |
+
[{'sequence': 'ECB - euro, FED - dollar, BoJ - yen',
|
18 |
+
'score': 0.7342271208763123,
|
19 |
+
'token': 1404,
|
20 |
+
'token_str': ' dollar'},
|
21 |
+
{'sequence': 'ECB - euro, FED - dollars, BoJ - yen',
|
22 |
+
'score': 0.10828445851802826,
|
23 |
+
'token': 1932,
|
24 |
+
'token_str': ' dollars'}]
|
25 |
+
|
26 |
+
|