I need an end-to-end workflow that starts with sourcing a suitable time-series dataset of battery State of Charge (SoC) and State of Health (SoH), moves through cleaning and feature engineering, and finishes with a well-tuned LSTM model that reports its accuracy clearly. Everything must run inside a single Google Colab notebook. The notebook should: • import or scrape the raw SoC/SoH data you have located, • document each preprocessing step, • build, train and validate the LSTM network using Keras/TensorFlow, • report common regression metrics (RMSE and MAE at minimum—feel free to add R² or others), • visualise learning curves and prediction vs. ground-truth plots, and • include explanatory comments so anyone can rerun or extend the work. Alongside the notebook, I also need a polished 25-slide PowerPoint deck. Use the rough draft I’m providing as a loose outline.U can refer from Page 9 to 13 of ppt to know about features of dataset,how I want my algorithm to be trained etc. Deliverables 1. Shareable Google Colab notebook with all code, outputs, and a link or citation to the acquired dataset 2. 25-page PowerPoint (.pptx) distilling the project for an academic audience Timeline is tight: everything must be ready within 3 days (4 at the absolute latest).