File size: 866 Bytes
6aafbf1
 
 
680040f
 
 
 
 
 
 
 
 
4173f0f
680040f
 
 
4173f0f
680040f
 
 
31abf58
680040f
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
---
license: apache-2.0
---
# Text-Summarizer
## About
An Abstractive text summarizer trained using lstm based sequence to sequence model with attention mechanisim. The attention model is used for generating each word of the summary conditioned on the input sentence.

Used CNN_DailyMail dataset.

### Training Model Overview
loss graph

![train_log.jpg](https://s3.amazonaws.com/moonup/production/uploads/1665205596889-634100e4b8b51e0098db811f.jpeg)

encoder-decoder overview

![model_plot.jpg](https://s3.amazonaws.com/moonup/production/uploads/1665205551954-634100e4b8b51e0098db811f.jpeg)

#### Conclusion
🫶 The machine learning model to convert a text document to abstract is done successfully.

🫶 Created a Flask app using an api call from this repository & deployed the app in heroku app.

##### Deployment:
🫶 https://text-summariser-v1.herokuapp.com/