GitHub - Mo-Abdelhameed/Text-Summarizer-NLP-Project: Text summarizer using AraBART fine tuned architecture.

Mo-Abdelhameed / Text-Summarizer-NLP-Project Public

forked from marwankhalid9876/NLP-Project

Notifications You must be signed in to change notification settings
Fork 0
Star 0

Text summarizer using AraBART fine tuned architecture.

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 20 Commits
.DS_Store		.DS_Store
.gitignore		.gitignore
Comparing_the_Effectiveness_of_Stemming_and_Lemmatization_in_Arabic_Text_Summarization_with_XLSum (4).pdf		Comparing_the_Effectiveness_of_Stemming_and_Lemmatization_in_Arabic_Text_Summarization_with_XLSum (4).pdf
First Presentation.pptx		First Presentation.pptx
L_outputs_text.joblib		L_outputs_text.joblib
MS1 Preprocessing.ipynb		MS1 Preprocessing.ipynb
Report_2.pdf		Report_2.pdf
S_outputs_text.joblib		S_outputs_text.joblib
Training.ipynb		Training.ipynb
evaluation.ipynb		evaluation.ipynb
outputs_tokenized.joblib		outputs_tokenized.joblib
readme.txt		readme.txt

Repository files navigation

To download the preprocessed data : https://drive.google.com/file/d/1IVIH23SIiRGmndN9wX9G5F9pJPV7dp7Y/view?usp=share_link
After downloading preprocessing data, you have to put it in this path "../" which means the parent directory where our project directory exists.

Packages that have to be installed : pandas, numpy, datasets, transformers, joblib, rouge

To train the model you have to run Training.ipynb cells in their order
To do the evaluation you have to run evaluation.ipynb cells