Browser-based digit OCR demo using canvas input and a Python neural network backend.
-
Updated
Jul 12, 2025 - Python
#
sigmoidactivation
Here are 4 public repositories matching this topic...
Trains a simple neural network to recognize binary letter patterns and evaluates its performance under random noise.
python machine-learning neural-network numpy jupyter-notebook classification pattern-recognition backpropagation sigmoidactivation handwritten-letters
-
Updated
May 22, 2025 - Jupyter Notebook
This repository implements handwritten digit recognition using a deep neural network trained on the MNIST dataset. It covers data preprocessing, model building, training, and evaluation for high accuracy classification. Additionally, it includes a system for predicting digits from new images. The project demonstrates effective use of deep learning.
python deep-neural-networks deep-learning tensorflow numpy keras python3 seaborn deep-learning-algorithms mnist-classification mnist-dataset image-classification matplotlib cv2 mnist-image-dataset adam-optimizer mnist-handwriting-recognition relu sigmoidactivation relu-function
-
Updated
Feb 3, 2026 - Jupyter Notebook
This project implements a simple feed-forward neural network for recognizing digits (0-9) from image data using backpropagation. The neural network is trained to identify digits based on pre-processed image data, which is filtered and resized before being used to train the network. The images are in grayscale and are converted to binary values
python machine-learning neural-network nn artificial-intelligence datascience computervision imageclassification artificialintelligence digitrecognition sigmoidactivation
-
Updated
Dec 26, 2024 - Python
Improve this page
Add a description, image, and links to the sigmoidactivation topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the sigmoidactivation topic, visit your repo's landing page and select "manage topics."