Implemented a transformer network for predicting the remaining useful life of lithium-ion batteries. Conducted data preprocessing and feature extraction from NASA battery datasets to enhance prediction accuracy. Developed an autoencoder with positional encoding for sequence prediction for better performance on training/test datasets. Utilized grid search to fine-tune hyperparameters, achieving a significant reduction in prediction error and improving model reliability