https://github.com/fr0st1n/seq2seq-chatbot

Seq2Seq Chatbot Using Tensorflow.
https://github.com/fr0st1n/seq2seq-chatbot

artificial-intelligence chatbot neural-network telegram tensorflow

Last synced: 22 days ago
JSON representation

Seq2Seq Chatbot Using Tensorflow.

• Host: GitHub
• URL: https://github.com/fr0st1n/seq2seq-chatbot
• Owner: FR0ST1N
• License: mit
• Created: 2017-04-03T15:29:39.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2020-01-21T09:46:02.000Z (almost 5 years ago)
• Last Synced: 2024-10-05T17:06:47.962Z (about 1 month ago)
• Topics: artificial-intelligence, chatbot, neural-network, telegram, tensorflow
• Language: Python
• Size: 166 KB
• Stars: 4
• Watchers: 3
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Seq2Seq-Chatbot

Seq2Seq Chatbot Using Tensorflow.

## Dependencies

* Tensorflow

* Numpy

* Six

## Usage

To preapre the enc and dec file for your own data use the [pre_processing.py][1] inside the data dir.

1. Start training the model by changing the `mode = 'train'` in [bot.py][2].

2. Model will be saved in the checkpoint dir every few steps based on the value assigned to `steps_per_checkpoint`.

3. After training is finished set `mode = 'test'` and execute the script.

### Triggers

If you are training a closed domain chatbot then you can use the [trigger.py][3] to make the chatbot perform an action based on the decoded output.

* Place the triggers in the [dec][4] file and make sure the trigger symbol is in [vocab20000.dec][5].

* Update the [trigger.json][6].

* Update the conditional statement and the action to perform in [trigger.py][3].

* Train and run your model.

### Other Files

* [telegram.py][7] - Just assign the API key to the `bot` var to interact with your bot on telegram.

* [debug.py][8] - Prints everything.

## References

1. [A Neural Conversational Model][9]

2. [Tensorflow Sequence-to-Sequence Models][10]

[1]: https://github.com/FR0ST1N/Seq2Seq-Chatbot/blob/master/data/pre_processing.py

[2]: https://github.com/FR0ST1N/Seq2Seq-Chatbot/blob/master/bot.py

[3]: https://github.com/FR0ST1N/Seq2Seq-Chatbot/blob/master/trigger.py

[4]: https://github.com/FR0ST1N/Seq2Seq-Chatbot/blob/master/data/train.dec

[5]: https://github.com/FR0ST1N/Seq2Seq-Chatbot/blob/master/checkpoint/vocab20000.dec

[6]: https://github.com/FR0ST1N/Seq2Seq-Chatbot/blob/master/trigger.json

[7]: https://github.com/FR0ST1N/Seq2Seq-Chatbot/blob/master/telegram.py

[8]: https://github.com/FR0ST1N/Seq2Seq-Chatbot/blob/master/debug.py

[9]: https://arxiv.org/abs/1506.05869

[10]: https://www.tensorflow.org/tutorials/seq2seq

## Licenses

* [Seq2Seq-Chatbot](https://github.com/FR0ST1N/Seq2Seq-Chatbot/blob/master/LICENSE)

* [seq2seq](https://github.com/google/seq2seq/blob/master/LICENSE)