Chi Honolulu
commited on
Commit
·
ce082fb
1
Parent(s):
dbe41d3
Upload README.md
Browse files
README.md
CHANGED
@@ -33,14 +33,18 @@ Here is how to use this model to classify a context-window of a dialogue:
|
|
33 |
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
|
34 |
import torch
|
35 |
|
|
|
36 |
test_texts = ['Utterance2']
|
|
|
37 |
test_text_pairs = ['Utterance1;Utterance2;Utterance3']
|
38 |
|
|
|
39 |
checkpoint_path = "chi2024/mt5-base-binary-cs-iiia"
|
40 |
model = AutoModelForSeq2SeqLM.from_pretrained(checkpoint_path)\
|
41 |
.to("cuda" if torch.cuda.is_available() else "cpu")
|
42 |
tokenizer = AutoTokenizer.from_pretrained(checkpoint_path)
|
43 |
|
|
|
44 |
def verbalize_input(text: str, text_pair: str) -> str:
|
45 |
return "Utterance: %s\nContext: %s" % (text, text_pair)
|
46 |
|
@@ -53,6 +57,7 @@ def predict_one(text, pair):
|
|
53 |
tokenizer.batch_decode(outputs, skip_special_tokens=True)]
|
54 |
return decoded
|
55 |
|
|
|
56 |
preds_txt = [predict_one(t,p) for t,p in zip(test_texts, test_text_pairs)]
|
57 |
preds_lbl = [1 if x == 'positive' else 0 for x in preds_txt]
|
58 |
print(preds_lbl)
|
|
|
33 |
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
|
34 |
import torch
|
35 |
|
36 |
+
# Target utterance
|
37 |
test_texts = ['Utterance2']
|
38 |
+
# Bi-directional context of the target utterance
|
39 |
test_text_pairs = ['Utterance1;Utterance2;Utterance3']
|
40 |
|
41 |
+
# Load the model and tokenizer
|
42 |
checkpoint_path = "chi2024/mt5-base-binary-cs-iiia"
|
43 |
model = AutoModelForSeq2SeqLM.from_pretrained(checkpoint_path)\
|
44 |
.to("cuda" if torch.cuda.is_available() else "cpu")
|
45 |
tokenizer = AutoTokenizer.from_pretrained(checkpoint_path)
|
46 |
|
47 |
+
# Define helper functions
|
48 |
def verbalize_input(text: str, text_pair: str) -> str:
|
49 |
return "Utterance: %s\nContext: %s" % (text, text_pair)
|
50 |
|
|
|
57 |
tokenizer.batch_decode(outputs, skip_special_tokens=True)]
|
58 |
return decoded
|
59 |
|
60 |
+
# Run the prediction
|
61 |
preds_txt = [predict_one(t,p) for t,p in zip(test_texts, test_text_pairs)]
|
62 |
preds_lbl = [1 if x == 'positive' else 0 for x in preds_txt]
|
63 |
print(preds_lbl)
|