10 Academy profile site

The project focuses on Large Language Models (LLMs), which are pivotal in advancing natural language processing capabilities. For analysis purposes, two distinct types of datasets were utilized, allowing for a comprehensive examination of the models' performance and effectiveness. To facilitate development, several tools were employed, including NLTK for natural language processing tasks, and a machine learning framework specifically designed for prompt engineering. Python served as the primary programming language, while Streamlit was utilized to create an interactive dashboard that showcases various features and insights derived from the analysis. In addition, various feature extraction tools were integrated to enhance the data processing capabilities. Ultimately, the report provides valuable insights into fundamental concepts related to LLMs, offering a clearer understanding of their functionalities, applications, and the implications of their use in real-world scenarios. 

The project involved a six-week collaboration with Rossmann Pharmaceuticals, focusing on sales predictions to enhance operational efficiency and inventory management. To achieve accurate forecasting, the project utilized four distinct store datasets, each providing valuable insights into sales trends and customer behavior. The analytical approach incorporated various tools and methodologies, including statistical modeling techniques to identify patterns within the data. Core data science libraries in Python, such as Pandas, NumPy, and Scikit-learn, were employed to facilitate data manipulation and analysis. Additionally, machine learning algorithms were implemented to develop predictive models, allowing for more precise sales forecasts. To streamline the development and deployment processes, MLOps practices were integrated, utilizing tools such as DVC (Data Version Control) for managing datasets and experiments, CML (Continuous Machine Learning) for automating the training and evaluation of models, and MLflow for tracking experiments and managing the machine learning lifecycle. This comprehensive approach not only aimed to improve sales predictions but also fostered a robust framework for ongoing model refinement and performance monitoring.