Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/RemiRigal/DatasetExplorer
A web tool for local dataset browsing and processing developped using the Flask + Angular stack.
https://github.com/RemiRigal/DatasetExplorer
ai angular data-processing data-science data-visualization dataset dataset-analysis docker docker-compose flask web-application
Last synced: 17 days ago
JSON representation
A web tool for local dataset browsing and processing developped using the Flask + Angular stack.
- Host: GitHub
- URL: https://github.com/RemiRigal/DatasetExplorer
- Owner: RemiRigal
- Created: 2020-04-07T14:29:16.000Z (about 4 years ago)
- Default Branch: master
- Last Pushed: 2022-09-08T14:02:50.000Z (almost 2 years ago)
- Last Synced: 2024-05-01T19:26:00.353Z (2 months ago)
- Topics: ai, angular, data-processing, data-science, data-visualization, dataset, dataset-analysis, docker, docker-compose, flask, web-application
- Language: TypeScript
- Homepage: https://remirigal.github.io/DatasetExplorer
- Size: 5.53 MB
- Stars: 5
- Watchers: 1
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- Contributing: docs/contributing.md
Lists
- jimsghstars - RemiRigal/DatasetExplorer - A web tool for local dataset browsing and processing developped using the Flask + Angular stack. (TypeScript)
README
# Dataset Explorer
A web tool for local dataset browsing and processing developped using the Flask + Angular stack.
## Features
Dataset Explorer provides the following features:
- Web-based (local server)
- Light weight and powerful (handle 100k+ files like a breeze)
- Easy visualization of data
- Plugin-based tool system for easy custom data processing
## Documentation
Documentation is hosted on [GitHub Pages](https://remirigal.github.io/DatasetExplorer/).
## Getting Started
### Using Docker
A docker compose file is provided, you can start the tool with:
```shell script
docker-compose up -d
```
However, you must mount your dataset directory in the `/data` directory of the backend container. You can do so by updating the following line of the `docker-compose.yml` file:
```yaml
volumes:
- ./data:/data # Replace '.data/' by the root path to your dataset
```
### From source
#### Backend (Flask)
The backend requires `Python 3.8`.
```shell script
cd backend
pip install -r requirements.txt
export DATASET_EXPLORER_ROOT=/path/to/dataset/root
python app.py
```
#### Frontend (Angular)
The frontend requires `Angular CLI v9.1.0` and `NodeJS v12.16.1`, make sure that they are installed first.
```shell script
cd frontend
npm install
ng serve --host 0.0.0.0
```
### Test tool
The app is available at http://127.0.0.1:4200.
## Write custom plugins/tools
Dataset Explorer allows you to write custom Python tools so that you can instantly test your own processing pipeline on your data. To do so you will need to create a class that inherits the `BasePlugin` class, the simplest plugin looks like this:
```python
# my_custom_plugin.py
import cv2
from dataset_explorer.io import FileType
from dataset_explorer.plugins import BasePlugin
class MyCustomPlugin(BasePlugin):
"""
MyCustomPlugin inherits from the BasePlugin class
"""
def __init__(self):
"""
The child class must provide a name as well as the input/output types of the plugin
It's required that its constructor takes no argument
"""
super(MyCustomPlugin, self).__init__("MyCustomPlugin", FileType.IMAGE, FileType.IMAGE)
def process(self, inFilename, outFilename):
"""
This method is called automatically, the inFilename argument is the path to the file to process
The outFilename is the path to the result file that must be created
For example, this function is converting the input image to black and white
"""
image = cv2.imread(inFilename)
image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
cv2.imwrite(outFilename, image)
```
This file must be placed either at `~/.DatasetExplorer/plugins`, or at any directory listed in the environment variable `DATASET_EXPLORER_PLUGINS` (separated by `:`).