lidiya
/
bart-large-xsum-samsum
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
lidiya/bart-large-xsum-samsum is a forked repo from huggingface. License: apache-2-0
5 Commits 1 Branch 0 Tags 3 GiB
Text 100%
Go to file
Lidiya Murakhovs'ka c8b48dc25a Adding `safetensors` variant of this model (#2) 
- Adding `safetensors` variant of this model (c7620b40fb565d18472fa03923a4caa4c8c654cf)


Co-authored-by: Safetensors convertbot <SFconvertbot@users.noreply.huggingface.co>
 		2023-03-16 22:44:01 +00:00
.gitattributes Adding `safetensors` variant of this model (#2) 2023-03-16 22:44:01 +00:00
README.md Fix typo in ROUGE metrics (#1) 2022-07-20 14:55:59 +00:00
config.json Initial commit 2021-05-26 21:21:46 +00:00
merges.txt Initial commit 2021-05-26 21:21:46 +00:00
model.safetensors Adding `safetensors` variant of this model (#2) 2023-03-16 22:44:01 +00:00
pytorch_model.bin Initial commit 2021-05-26 21:21:46 +00:00
special_tokens_map.json Initial commit 2021-05-26 21:21:46 +00:00
tokenizer.json Initial commit 2021-05-26 21:21:46 +00:00
tokenizer_config.json Initial commit 2021-05-26 21:21:46 +00:00
vocab.json Initial commit 2021-05-26 21:21:46 +00:00

README.md

language tags license datasets widget model-index
en
bart
seq2seq
summarization
apache-2.0
samsum
text
Hannah: Hey, do you have Betty's number? Amanda: Lemme check Amanda: Sorry, can't find it. Amanda: Ask Larry Amanda: He called her last time we were at the park together Hannah: I don't know him well Amanda: Don't be shy, he's very nice Hannah: If you say so.. Hannah: I'd rather you texted him Amanda: Just text him 🙂 Hannah: Urgh.. Alright Hannah: Bye Amanda: Bye bye
name results
bart-large-xsum-samsum
task dataset metrics
name type
Abstractive Text Summarization abstractive-text-summarization
name type
SAMSum Corpus: A Human-annotated Dialogue Dataset for Abstractive Summarization samsum
name type value
Validation ROUGE-1 rouge-1 54.3921
name type value
Validation ROUGE-2 rouge-2 29.8078
name type value
Validation ROUGE-L rouge-l 45.1543
name type value
Test ROUGE-1 rouge-1 53.3059
name type value
Test ROUGE-2 rouge-2 28.355
name type value
Test ROUGE-L rouge-l 44.0953

bart-large-xsum-samsum

This model was obtained by fine-tuning facebook/bart-large-xsum on Samsum dataset.

Usage

from transformers import pipeline

summarizer = pipeline("summarization", model="lidiya/bart-large-xsum-samsum")
conversation = '''Hannah: Hey, do you have Betty's number?
Amanda: Lemme check
Amanda: Sorry, can't find it.
Amanda: Ask Larry
Amanda: He called her last time we were at the park together
Hannah: I don't know him well
Amanda: Don't be shy, he's very nice
Hannah: If you say so..
Hannah: I'd rather you texted him
Amanda: Just text him 🙂
Hannah: Urgh.. Alright
Hannah: Bye
Amanda: Bye bye                                       
'''
summarizer(conversation)

Training procedure

Results

key value
eval_rouge1 54.3921
eval_rouge2 29.8078
eval_rougeL 45.1543
eval_rougeLsum 49.942
test_rouge1 53.3059
test_rouge2 28.355
test_rougeL 44.0953
test_rougeLsum 48.9246