Model Deployment / production

Multiple forecasting models trained and tested. Algorithm development is under preparation.

We implemented prophet, sarimax, xgboost, lstm, etc. algorithms.

• Operating System: Microsoft Windows 10 (Version 10.0.19045, 64-bit) • Processor: AMD Ryzen 7 PRO 3700 • Anaconda 3: specific environment for the ITEA project • JupyterLab Version 4.0.11 • Python Version: 3.10.11 • Libraries and Versions: – pandas (v2.2.2) for data manipulation and analysis – numpy (v1.26.4) for numerical computations and array handling – matplotlib (v3.7.0) for data visualization – statsmodels (v0.14.0) for time series modeling (ARIMA/SARIMA) – pmdarima (v2.0.4) for automated ARIMA order selection – scipy (v1.10.0) for scientific computing and statistical functions – prophet (v1.1.5) for forecasting with holiday and seasonality effects – scikit-learn (v1.5.0) for machine learning and evaluation metrics – xgboost (v2.0.3) for gradient-boosting decision trees – tensorflow (v2.17.0) for constructing and training neural networks, particularly LSTM models We integrated the results with our existing ecosystem for data analysis and dashboards in Power BI.

Pharmaceutical companies, Pharmaceutical manufacturing, Government agencies

Confidential