Skip to main content

A No-code solution to create the images transformation pipeline.

Project description

AugBuilder

PyPI version

A No-code solution to create the images transformation pipelin.

  • Installed as a pip package.
  • Runs in a browser.
  • Uses Albumentations library to apply transformations.
  • Benchmarks the pipeline.
  • Generates a YAML config and a tiny chunk of python code to integrate with PyTorch code.

Interface image

Powered by Albumentations and Streamlit.

WIP

Work still in progress. We'll appreciate any feedback from the community: bug-report, feature-request, pull-request.

You can leave anonimous feedback here.

Installation

You need python 3.6+ and pip to install the app.

pip install augbuilder

Usage

Run the app

Run augbuilder from the terminal.

After a few seconds the browser will open the page localhost:8501.

To stop the application press the ctrl+c combination in the terminal.

Step by step guide

Watch this demo video of usage.

youtube video

  1. Drop an image to the upload area.
  2. Use dropdown on left side to select transformations.
  3. Configure transformations below the list of dropdowns.
  4. Random results are shown in the main area.
  5. To regenerate results click "Refresh images" button.
  6. Click "Download config" to get the yaml config and a python integration script.

How to use ONE-OF

Select oneof in list if you want to add this into you transformation list. Then you can add different transformations in it. To close oneof select StopOneOf. Please, don't select THE SAME transformation, it can caused some errors which will be removed later.

Pipeline config example

RandomResizedCrop:
    height: 299
    width: 299
    scale: (0.24, 1.0)
    ratio: (0.75, 1.3333333333333333)
    interpolation: 0
Flip:
Transpose:
OneOf:
    MotionBlur: {'blur_limit': (3, 53)}
    Blur: {'blur_limit': (3, 22)}
ShiftScaleRotate:
    shift_limit: (-0.06, 0.06)
    scale_limit: (-0.1, 0.1)
    rotate_limit: (-90, 90)
    interpolation: 0
    border_mode: 3
    value: [0, 0, 0]
HueSaturationValue:
    hue_shift_limit: (-20, 20)
    sat_shift_limit: (-30, 30)
    val_shift_limit: (-20, 20)

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

augbuilder-0.0.3.tar.gz (12.3 kB view details)

Uploaded Source

Built Distribution

augbuilder-0.0.3-py3-none-any.whl (15.0 kB view details)

Uploaded Python 3

File details

Details for the file augbuilder-0.0.3.tar.gz.

File metadata

  • Download URL: augbuilder-0.0.3.tar.gz
  • Upload date:
  • Size: 12.3 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.2.0 pkginfo/1.5.0.1 requests/2.24.0 setuptools/47.1.0 requests-toolbelt/0.9.1 tqdm/4.48.2 CPython/3.8.5

File hashes

Hashes for augbuilder-0.0.3.tar.gz
Algorithm Hash digest
SHA256 098aceabc85311944a22a570e28437457ce9fb1865fed736e9f357347da02192
MD5 a7fae2a63cf7def667a088dd783b6c98
BLAKE2b-256 8a9eed2d6dc64e7859cdb75ed9dd6fadede627c53fb9c2a40da77ceec97876e8

See more details on using hashes here.

File details

Details for the file augbuilder-0.0.3-py3-none-any.whl.

File metadata

  • Download URL: augbuilder-0.0.3-py3-none-any.whl
  • Upload date:
  • Size: 15.0 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.2.0 pkginfo/1.5.0.1 requests/2.24.0 setuptools/47.1.0 requests-toolbelt/0.9.1 tqdm/4.48.2 CPython/3.8.5

File hashes

Hashes for augbuilder-0.0.3-py3-none-any.whl
Algorithm Hash digest
SHA256 fe31be4ea67c67e730edac681acd31afd04a4fe37b7929cc3968dae0bc609c3b
MD5 8318f1ac4759001ffa97016d10397367
BLAKE2b-256 df3ddf7469330a3aeee9c31c24584648d01093ddb818d1eccd9a7228c63099f4

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page