figurative-nlp
commited on
Commit
•
57892b4
1
Parent(s):
94b95cc
Update README.md
Browse files
README.md
CHANGED
@@ -1,4 +1,20 @@
|
|
1 |
-
This model can convert the literal expression to figurative/metaphorical expression.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
|
3 |
For example (the <m> and </m> is the mark that inform the model which literal expression we want to convert it as figurative expression):
|
4 |
|
|
|
1 |
+
This model can convert the literal expression to figurative/metaphorical expression. Below is the usage of our model:
|
2 |
+
|
3 |
+
|
4 |
+
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
|
5 |
+
|
6 |
+
tokenizer = AutoTokenizer.from_pretrained("figurative-nlp/t5-figurative-generation")
|
7 |
+
model = AutoModelForSeq2SeqLM.from_pretrained("figurative-nlp/t5-figurative-generation")
|
8 |
+
|
9 |
+
|
10 |
+
input_ids = tokenizer(
|
11 |
+
"research is <m> very difficult </m> for me.", return_tensors="pt"
|
12 |
+
).input_ids # Batch size 1
|
13 |
+
outputs = model.generate(input_ids,beam_search = 5)
|
14 |
+
result = tokenizer.decode(outputs[0], skip_special_tokens=True)
|
15 |
+
#result : research is a tough nut to crack for me.
|
16 |
+
|
17 |
+
|
18 |
|
19 |
For example (the <m> and </m> is the mark that inform the model which literal expression we want to convert it as figurative expression):
|
20 |
|