User friendly image bootstraping framework.
Project description
Label wrapper
User friendly image bootstraping framework.
Label bootstrapping flow
Label wrapper enables label bootstrapping process:
- Load first data batch
- Manually label first batch
- Train first segmentation model
- Load second data batch
- Use first trained segmentation model to predict labels
- Review labels and merge first and second labelled data
- train the second segmentation model
- Repeat steps 4.-7. until out of raw data or review of labels is no longer required.
Technical implementation example
- Load data into dataset
- Export html
- Label
- Export to json
- Import json and convert json to tfrecords
- Train on tfrecords
- Introduce new data
- Predict with trained model to tf records
- Import stored tfrecords and convert to html with labels
- Review stored labels and export to json
- Join reviewed json and manual json (from step 4)
- Repeat 5 - 11 for n times
- Run out of data to label
- Measure performance
TODO
- Finnish dual data dataset with gtiff (add test)
- mask to shapefile (geocoded)
- shapefile exporter
- shapefile imporoter?
- example inference step with a pretrained segmentation cnn
- (maybe) constructor should take json and load it in postinit
- (maybe) Add via html tests with js (selenium?)
Thanks
Label editor used is VIA 2.0.6.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
label-wrapper-0.1.2.tar.gz
(79.8 kB
view hashes)
Built Distribution
Close
Hashes for label_wrapper-0.1.2-py2.py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | d61894fadc6b48003fffff0f1aef29a10652e4489cc6759dbaf950adc70caf05 |
|
MD5 | cf265d82d696d129c05996aa430c8eb7 |
|
BLAKE2b-256 | ed7f6ce5071dfbd587d55e213f7207702a90078147357fdcd5c16d3c038a2c2d |