I will do machine learning and deep learning python projects

I can work with suitable historical time-series data such as sales, demand, orders, product usage, revenue volume, website traffic, operational metrics and similar measurable variables.

It depends on the data frequency and seasonality. More historical observations generally help reveal recurring patterns. Send a sample before ordering if you are unsure.

Depending on your data, I may evaluate ARIMA/SARIMA, Prophet, XGBoost, LightGBM, regression-based models, LSTM/GRU or other appropriate forecasting approaches.

No. Deep learning is not always the best forecasting method. I compare approaches based on your data and validation results rather than choosing a model only because it is complex.

Yes. Standard and Premium packages support multiple agreed series. Large catalogs or hundreds of SKUs should be discussed first for a custom scope.

Depending on the use case, I can use metrics such as MAE, RMSE, MAPE, WAPE or other suitable measures together with backtesting.

No forecasting system can guarantee exact future values. Accuracy depends on historical signal, data quality, external events and how stable the underlying demand pattern is.