Skip to content

Commit 2013093

Browse files
committed
2 parents 7d50615 + eeb1cb2 commit 2013093

File tree

1 file changed

+10
-2
lines changed

1 file changed

+10
-2
lines changed

README.md

Lines changed: 10 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -3,10 +3,18 @@
33
## Description
44
Hand OCR Paint is meant to be useful for testing an OCR Neural Network for handwritten characters.
55

6+
![](https://github.com/SimoSbara/HandOCRPaint/blob/main/example_ocr.gif)
7+
68
## Testing
7-
[Sample Networks](https://github.com/SimoSbara/HandOCRPaint/tree/main/nets) that I trained (Alex Net and Mobile Net) for immediate testing.
9+
[Sample Networks](https://github.com/SimoSbara/HandOCRPaint/tree/main/nets) that I trained:
10+
* [Alex Net](https://medium.com/@siddheshb008/alexnet-architecture-explained-b6240c528bd5);
11+
* [Mobile Net V3](https://towardsdatascience.com/everything-you-need-to-know-about-mobilenetv3-and-its-comparison-with-previous-versions-a5d5e5a6eeaa);
12+
* [Custom OCR KAN Net](https://github.com/Blealtan/efficient-kan) that is equivalent to MLP (trained on MNIST).
813

9-
![](https://github.com/SimoSbara/HandOCRPaint/blob/main/example_ocr.gif)
14+
## Parameters
15+
Use [-1, 1] normal range and image inversion for KAN Network, otherwise [0, 1] and no inversion.
16+
17+
Change parameters according to your training data.
1018

1119
## Libraries used:
1220
* [CxImage](https://www.codeproject.com/Articles/1300/CxImage) for image encoding/decoding;

0 commit comments

Comments
 (0)