Melancholy LSTM:melancholy_LSTM

“This project explores how a small LSTM model learns the rhythm of a melancholy inner monologue.”

🍄 Mushroom Species Classification (using RBFN)

​This project aims to classify mushroom species using a Radial Basis Function Network (RBFN) implemented with PyTorch. ​💾 Dataset ​The dataset used in this project contains features related to mushroom species found in Bolu, Turkey, sourced from Kaggle. ​Dataset Name: Mushroom Species Found in Bolu ​Source: Kaggle - Eydanur Aydın ​File Name: mantar_veriseti.csv ​⚠️ Note: This dataset is the cleaned and preprocessed version of the original Kaggle data, tailored for this specific classification project. ​⚙️ Model and Methodology ​The classification is performed using a Radial Basis Function Network (RBFN) architecture, which differs from traditional Artificial Neural Networks (ANNs). ​Architecture: ​Input Layer: Feature Count (input_dim) ​Hidden Layer (RBF Kernel): 10 Centers (num_centers=10) ​Output Layer: 3 Classes (output_dim=3) ​Kernel Function: Gaussian Kernel (e^{-\beta ||\mathbf{x} - \mathbf{c}||^2}) ​Training: ​Loss Function: nn.CrossEntropyLoss ​Optimization: optim.Adam(lr=0.01) ​The RBF centers (\mathbf{C}) and the \beta parameter are defined as learnable parameters (nn.Parameter), optimized along with the weights and biases of the linear output layer. ​💻 Required Libraries ​The fundamental libraries required to run this project are: ​torch (PyTorch) ​pandas ​scikit-learn