Agent skill · research science · aj-geddes
Neural Network Design
Design and architect neural networks with various architectures including CNNs, RNNs, Transformers, and attention mechanisms using PyTorch and TensorFlow
Why this skill is useful
Adds executable scripts for various neural network architectures using PyTorch and TensorFlow that enhance the AI's capability to design and implement complex models.
What it needs
Requires tensorflow, torch installed locally. About 4k tokens when loaded. Last updated 2026-03-04. 307 stars on the source repository.
What this skill does
Neural Network Design Overview This skill covers designing and implementing neural network architectures including CNNs, RNNs, Transformers, and ResNets using PyTorch and TensorFlow, with focus on architecture selection, layer composition, and optimization techniques. When to Use Designing custom neural network architectures for computer vision tasks like image classification or object detection Building sequence models for time series forecasting, natural language processing, or video analysis Implementing transformer-based models for language understanding or generation tasks Creating hybrid architectures that combine CNNs, RNNs, and attention mechanisms Optimizing network depth, width, and skip connections for better training and performance Selecting appropriate activation functions, normalization layers, and regularization techniques Core Architecture Types Feedforward Networks (MLPs): Fully connected layers Convolutional Networks (CNNs): Image processing Recurrent Networks (RNNs, LSTMs, GRUs): Sequence processing Transformers: Self-attention based architecture Hybrid Models: Combining multiple architecture types Network Design Principles Depth vs Width: Trade-offs between layers and units Skip Connections: Residual networks for deeper training Normalization: Batch norm, layer norm for stability Regularization: Dropout, L1/L2 preventing overfitting Activation Functions: ReLU, GELU, Swish for non-linearity PyTorch and TensorFlow Implementation Architecture Selection Guide MLP: Tabular data, simple classification CNN: Image classification, object detection LSTM/GRU: Time series, sequential data Transformer: NLP, long-range dependencies ResNet: Very deep networks, image tasks Key Design Considerations Input/output shape compatibility Receptive field size for CNNs Sequence length for RNNs Attention head count for Transformers Skip connection placement for ResNets Deliverables Network architecture definition Parameter count analysis Layer-by-layer description Data f …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills aj-geddes/neural-network-design