metadata
title: Sketch Nn
emoji: π»
colorFrom: pink
colorTo: blue
sdk: gradio
sdk_version: 4.38.1
app_file: app.py
pinned: false
license: mit
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
Sketch NN: Neural Network Designer
Design Neural Networks with a Simple Sketch
Sketch NN is a powerful tool that allows you to design and generate PyTorch neural network code from simple flowchart sketches. With Sketch NN, you can quickly prototype complex neural architectures without writing a single line of code!
π Features
- πΈ Upload or capture flowchart images
- π§ Supports a wide range of neural network layers
- π§ Generates ready-to-use PyTorch code
- π₯οΈ User-friendly Gradio web interface
- π FastAPI backend for scalable deployment
π οΈ Supported Layers
- Convolutional (Conv2D)
- Pooling (MaxPool2D, AvgPool2D)
- Fully Connected (Linear)
- Batch Normalization
- Dropout
- Activation Functions (ReLU, LeakyReLU, Sigmoid, Tanh)
- Recurrent (LSTM, GRU)
- Transformer
- Multi-head Attention
πΌοΈ How It Works
[Insert a diagram or flowchart here showing the process from sketch to code]
- Sketch your neural network architecture
- Upload or capture the image
- Sketch NN processes the image and extracts layer information
- PyTorch code is generated based on the extracted information
- Download and use the generated code in your project
π Getting Started
Prerequisites
- Python 3.7+
- PyTorch
- OpenCV
- Tesseract OCR
Installation
pip install sketch-nn