Comments (3)
I believe 20k words is a relatively big dictionary. Usually, 8k vocabularies are good enough for a English learner to understand a news.
The model is limited to generate a short summary (about 10 words) from a few sentences (about 30 words in total). If the input is too long, I don't know what the model can generate. Also please normalize the input(lowercase and segmentation). Hello word!
->hello world !
I admit that this system is far from acceptable for a practical application.
from tensorflow-summarization.
Thanks for your good explanation~
I still have few questions:
-
In order to adjust to a paragraph with ~ 100 words, any parameters in neural network I need to adjust also?
-
I notice in R Nallapati's paper (https://arxiv.org/abs/1602.06023) they implement a Switching Generator-Pointer to handle OOV, is it something that would improve the situation?
from tensorflow-summarization.
-
No. But the training data is around 30 words. So it shouldn't work on longer text unless you train with other data.
-
Yes. I would recommend PointerNet using similar idea but different model. Also, this model is trained on longer documents.
from tensorflow-summarization.
Related Issues (20)
- doc_dict and sum_dict? HOT 1
- Evaluation metric
- Tensorboard Visulasation HOT 1
- csv instead of txt file HOT 1
- AttributeError HOT 9
- No valid.article.filter and valid.title.filter files
- Can work on CUDA 9.1?
- InvalidArgumentError HOT 1
- Beam Search Algorithm
- Purpose of doc_dict.txt and sum_dict.txt HOT 2
- How to Deploy it in Django Project
- Summary Length based on data size
- How to train the modal with my own data set
- Unable to resolve below error Error while test pretrained model. Geting error in create model (in bigrumodel.py [line number 90] while setting up attention variable) before model.saver.restore is called . My tensorflow version is 1.5 and this code is in version 1.1 ERROR: TypeError: __new__() missing 3 required positional arguments: 'time', 'alignments', and 'alignment_history'
- How to do Rouge Evaluation?
- how to increase the number of word output in the final summary?
- Pretrained model is trained on? HOT 4
- Benchmarks?
- list index out of range
- How can i train further using the pretrained checkpoint files? HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from tensorflow-summarization.