https://github.com/jorgechato/sauce

Super smart, super custom autocomplete code. AI to build code with you. You can implement it in a plugin for example:
https://github.com/jorgechato/sauce

keras public python tensorflow

Last synced: 27 days ago
JSON representation

Super smart, super custom autocomplete code. AI to build code with you. You can implement it in a plugin for example:

• Host: GitHub
• URL: https://github.com/jorgechato/sauce
• Owner: jorgechato
• Created: 2017-12-29T21:56:36.000Z (about 7 years ago)
• Default Branch: master
• Last Pushed: 2018-04-18T08:22:19.000Z (almost 7 years ago)
• Last Synced: 2024-11-07T10:45:51.522Z (3 months ago)
• Topics: keras, public, python, tensorflow
• Language: Python
• Homepage: https://github.com/jorgechato/vim-sauce
• Size: 18.6 KB
• Stars: 0
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Sauce

A recurrent neural network & LSTM AI to autocomplete code.

### Install

#### Local

```zsh

$ pip install -r requirements.txt

```

#### Docker

```zsh

$ docker build -t sauce .

$ docker run -d \

-v ${PWD}/data:/code/data \

-v ${PWD}/out:/code/out \

-v ${PWD}/weights:/code/weights \

--name sauce-train sauce

```

### Get data

Concatenate all files you have written splitted by language in a single file. You

can use the following command:

```zsh

$ cat /**/*.py > ./data/python.txt

```

### Train

```zsh

$ python train.py --data 

# or to show the full options

$ python train.py -h

```

### Folder structure

```zsh

.

├── Dockerfile

├── README.md

├── requirements.txt

├── sauce

│   ├── data.py

│   ├── __init__.py

│   ├── LSTM

│   │   ├── bidirectional.py

│   │   ├── __init__.py

│   │   └── text_generation.py

│   └── utils.py

├── train.py

├── predict.py

├── api.py

├── data

├── out

└── weights

```