https://github.com/abhiramdodda/mnist_nn
Neural Network for MNIST. Built from scratch using Numpy and Pandas.
https://github.com/abhiramdodda/mnist_nn
neural-networks numpy pandas
Last synced: 2 months ago
JSON representation
Neural Network for MNIST. Built from scratch using Numpy and Pandas.
- Host: GitHub
- URL: https://github.com/abhiramdodda/mnist_nn
- Owner: AbhiramDodda
- Created: 2024-03-09T17:08:31.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2024-03-09T17:27:25.000Z (over 2 years ago)
- Last Synced: 2025-03-01T11:44:48.202Z (over 1 year ago)
- Topics: neural-networks, numpy, pandas
- Language: Python
- Homepage:
- Size: 3.91 KB
- Stars: 2
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# MNIST_NN
Neural Network for MNIST
### Link for Dataset: https://www.kaggle.com/competitions/digit-recognizer/data
### The Neural Network is built from scratch using Numpy, data loading is done using Pandas.
## Performance: 92% accuracy
## Some details
NN will have a simple two-layer architecture. Input layer a[0] will have 784 units corresponding to the 784 pixels in each 28x28 input image. A hidden layer a[1] will have 10 units with ReLU activation, and finally our output layer a[2] will have 10 units corresponding to the ten digit classes with softmax activation.