https://github.com/wb14123/seq2seq-couplet

Play couplet with seq2seq model. 用深度学习对对联。
https://github.com/wb14123/seq2seq-couplet

deep-learning machine-learning seq2seq

Last synced: about 2 months ago
JSON representation

Play couplet with seq2seq model. 用深度学习对对联。

• Host: GitHub
• URL: https://github.com/wb14123/seq2seq-couplet
• Owner: wb14123
• License: agpl-3.0
• Created: 2018-02-24T02:44:52.000Z (over 6 years ago)
• Default Branch: master
• Last Pushed: 2024-07-25T02:24:32.000Z (about 2 months ago)
• Last Synced: 2024-07-25T03:32:36.605Z (about 2 months ago)
• Topics: deep-learning, machine-learning, seq2seq
• Language: Python
• Homepage: https://ai.binwang.me/couplet
• Size: 40 KB
• Stars: 5,488
• Watchers: 126
• Forks: 1,070
• Open Issues: 11
• Metadata Files:
  • Readme: README.markdown
  • License: LICENSE

Awesome Lists containing this project

• awesome - seq2seq-couplet - Play couplet with seq2seq model. 用深度学习对对联。 (Python)
• awesome-list - seq2seq-couplet - 用深度学习对对联。 (Natural Language Processing / Conversation & Translation)

README

        

This is a project use seq2seq model to play couplets (对对联)。 This project is written with Tensorflow. You can try the demo at [https://ai.binwang.me/couplet](https://ai.binwang.me/couplet).

Pre-requirements

--------------

* Tensorflow

* Python 3.6

* Dataset

Dataset

-----------

You will need some data to run this program, the dataset can be downloaded from [this project](https://github.com/wb14123/couplet-dataset).

** Note: If you are using your own dataset, you need to add `` and `<\s>` as the first two line into the vocabs file. **

Usage

------------

### Train

Open `couplet.py` and config the file locations and hyperparams. Then run `python couplet.py` to train the model. You can see the training loss and bleu score at Tensorbloard. You may need to re-config `learning_rate` when you find the loss stops descresing. Here is an example of the loss graph:

![loss graph](https://user-images.githubusercontent.com/1906051/36624881-50586e54-1950-11e8-8383-232763831cbc.png)

If you stoped the training and want to continue to train it. You can set `restore_model` to `True` and use `m.train(, start=)`, which `start` is the steps you've already run.

I've trained the model on a Nvidia GTX-1080 GPU for about 4 days.

### Run the trained model

Open `server.py` and config the `vocab_file` and `model_dir` params. Then run `python server.py` will start a web service that can play couplet.

Or build the Docker image with Dockerfile and run it with Docker. Remember to mount the correct model file paths into the Docker container.

Examples

-------------

Here are some examples generated by this model:

| 上联                        | 下联                |

|-----------------------------|--------------------|

| 殷勤怕负三春意                | 潇洒难书一字愁        |

| 如此清秋何吝酒                | 这般明月不须钱        |

| 天朗气清风和畅                | 云蒸霞蔚日光辉        |

| 梦里不知身是客                | 醉时已觉酒为朋        |

| 千秋月色君长看                | 一夜风声我自怜        |