No project description provided
Project description
# GangliPy
GangliPy is a collection of neural networks evaluated to be the best for certain types of tasks.
# In Progress
## Auto OCR
The first project here is an attempt to recognize all 60,000 unicode characters using a single hidden layer.
Fully connected input and output layers have proven sufficient for recognizing 16x16 unicode images, and can even predict never before seen characters.
![OCR example](https://i.imgur.com/2mwf7XQ.jpg)
- However, ‘predicting never before seen characters’ fades to noise when enough characters are seen, so another layer is
needed to make this ‘variational’ and keep realistic guesses when failing to perfectly predict.
## Unicode Regognition Tests
The unicode recognition tests I’m using for Auto OCR will need to be pulled out and placed into a seperate testing and evaluation repository.
# Installation
There is currently no pip repository available. Once Auto OCR is finished, it will be pulled out into its own repo and will be added as a requirement to this one.
#License
GangliPy is distributed under the terms of [GNU AGPL V3.0](https://choosealicense.com/licenses/agpl-3.0)
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
File details
Details for the file GangliPy-0.0.1.tar.gz
.
File metadata
- Download URL: GangliPy-0.0.1.tar.gz
- Upload date:
- Size: 20.9 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/1.13.0 pkginfo/1.5.0.1 requests/2.21.0 setuptools/41.0.0 requests-toolbelt/0.9.1 tqdm/4.31.1 CPython/3.5.6
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | a43d6d508c0df6ab60a62467def187b7e59465d515b5c3905b5a3010d7a176c5 |
|
MD5 | 35b59203838b2a60ae5ef176188a01f6 |
|
BLAKE2b-256 | e70c2cf8335c9bd9e8f19811efc1cb61f863ec67f6fc921add8bb794470a280e |