Commit
•
5db2699
1
Parent(s):
6d7b7a7
Update README.md
Browse files
README.md
CHANGED
@@ -1,6 +1,9 @@
|
|
1 |
---
|
2 |
-
|
3 |
-
|
|
|
|
|
|
|
4 |
---
|
5 |
|
6 |
# Model Card for Pegasus for Claim Summarization
|
@@ -30,6 +33,11 @@ This was the best performing model at the time of experimentation.
|
|
30 |
- **Repository:** https://github.com/varadhbhatnagar/FC-Claim-Det
|
31 |
- **Paper:** https://aclanthology.org/2022.coling-1.259/
|
32 |
|
|
|
|
|
|
|
|
|
|
|
33 |
# Uses
|
34 |
|
35 |
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
@@ -60,6 +68,7 @@ those in the corresponding paper as those experiments were conducted in the mont
|
|
60 |
<!-- This should link to a Data Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
61 |
|
62 |
[Data](https://github.com/varadhbhatnagar/FC-Claim-Det/blob/main/public_data/released_data.csv)
|
|
|
63 |
## Training Procedure
|
64 |
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
65 |
Finetuning the pretrained Distilled PEGASUS model on the 567 pairs released in our paper.
|
@@ -124,7 +133,7 @@ tokeizer = PegasusTokenizerFast.from_pretrained('varadhbhatnagar/fc-claim-det-DP
|
|
124 |
model = PegasusForConditionalGeneration.from_pretrained('varadhbhatnagar/fc-claim-det-DPEGASUS')
|
125 |
|
126 |
text ='world health organisation has taken a complete u turn and said that corona patients neither need isolate nor quarantine nor social distance and it can not even transmit from one patient to another'
|
127 |
-
tokenized_text = tokeizer.encode(text, return_tensors="pt")
|
128 |
|
129 |
summary_ids = model.generate(tokenized_text,
|
130 |
num_beams=6,
|
|
|
1 |
---
|
2 |
+
'[object Object]': null
|
3 |
+
license: apache-2.0
|
4 |
+
language:
|
5 |
+
- en
|
6 |
+
pipeline_tag: summarization
|
7 |
---
|
8 |
|
9 |
# Model Card for Pegasus for Claim Summarization
|
|
|
33 |
- **Repository:** https://github.com/varadhbhatnagar/FC-Claim-Det
|
34 |
- **Paper:** https://aclanthology.org/2022.coling-1.259/
|
35 |
|
36 |
+
## Tokenizer
|
37 |
+
|
38 |
+
Same as https://huggingface.co/sshleifer/distill-pegasus-cnn-16-4
|
39 |
+
|
40 |
+
|
41 |
# Uses
|
42 |
|
43 |
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
|
|
68 |
<!-- This should link to a Data Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
|
69 |
|
70 |
[Data](https://github.com/varadhbhatnagar/FC-Claim-Det/blob/main/public_data/released_data.csv)
|
71 |
+
|
72 |
## Training Procedure
|
73 |
<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
|
74 |
Finetuning the pretrained Distilled PEGASUS model on the 567 pairs released in our paper.
|
|
|
133 |
model = PegasusForConditionalGeneration.from_pretrained('varadhbhatnagar/fc-claim-det-DPEGASUS')
|
134 |
|
135 |
text ='world health organisation has taken a complete u turn and said that corona patients neither need isolate nor quarantine nor social distance and it can not even transmit from one patient to another'
|
136 |
+
tokenized_text = tokeizer.encode(text, return_tensors="pt")
|
137 |
|
138 |
summary_ids = model.generate(tokenized_text,
|
139 |
num_beams=6,
|