CNN Research Abstraction Library
The CNN Research Abstraction Library or CRAL in short is a deep learning computer vision library for data scientists, researchers, and developers. With a primary focus on applied deep learning, the CRAL library encourages rapid development and comes with ready-to-use state-of-the-art networks and other pragmatic tools for a variety of applications in the computer vision space.
Our aim is also to make it easier to reproduce and extend the results of various Deep Learning-powered Computer Vision (DLCV) algorithms developed in academia and industrial labs.
List of Algorithms
Object detection
- RetinaNet
- yolov3
- SSD
- FasterRCNN
Instance Segmentation
- MaskRCNN
Semantic Segmentation
- UNet
- UNet ++
- Deeplabv3+
- FpnNet
- PspNet
- SegNet
- LinkNet
Guiding Principles
Simple: To make it easy for deep learning engineers & students alike to use neural networks to build computer vision applications of their choice, using low code approach.
Fast: To accelerate going from experimentation to a working model.
Reproducible: To offer implementations that can easily be trained and reproduced on your own data.
Components
CRAL has a modular design to enable you to use each of its components independently, Alternatively, you can use the pipeline to get started quickly with multiple networks out-of-the-box.
| Components | Description |
|---|---|
| CNN models | Ready to use implementations of State-of-the-art (SOTA) algorithms. |
| Pipeline tools | Load and validate your data before you start training. |
| Optimization and debugging | Integration with Experiment Tracking, HP Optimization and other toolsets to help faster and build transparent models |
Detailed documentation: Link
1 maintainer
Metadata
1 maintainer
Release files for cral 0.4.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| cral-0.4.0-cp36-cp36m-manylinux1_x86_64.whl | CPython 3.6 | CPython 3.6 pymalloc | Linux glibc 2.5+ x86-64 | Details for cral-0.4.0-cp36-cp36m-manylinux1_x86_64.whl |
Release files / cral-0.4.0-cp36-cp36m-manylinux1_x86_64.whl
| Download URL | cral-0.4.0-cp36-cp36m-manylinux1_x86_64.whl |
|---|---|
| Size | 382.6 kB |
| Tags | CPython 3.6 CPython 3.6 pymalloc Linux glibc 2.5+ x86-64 |
|
SHA-256 checksum How to use checksums |
2b173dc21153380400032c695cf544f52661ec09a4556d95530175affbd46ae6
|
|
BLAKE2b-256 checksum How to use checksums |
914351b86691823a4c74d634b16b95ec9ad62ad9bd84eef8e72ac60c32f6a690
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/3.2.0 pkginfo/1.6.1 requests/2.25.0 setuptools/50.3.2 requests-toolbelt/0.9.1 tqdm/4.53.0 CPython/3.6.9
|
Security for cral 0.4.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.
