Prepare a prototype of a machine learning model for "Digits". The company is developing solutions for efficient operation of industrial enterprises.
The model should predict the recovery rate of gold from gold-bearing ore. Use data on mining and processing parameters.
The model will help optimize production to avoid launching a facility with unprofitable characteristics.
we need to:
- Prepare the data;
- Conduct exploratory data analysis;
- Build and train the model.
During the project, the following tasks were performed:
Data preparation. Exploratory data analysis. Model building and training. The final model was built using the DecisionTreeRegressor algorithm with hyperparameters max_depth=5. The effectiveness of the enrichment was calculated correctly. The model showed the best sMAPE value among the other models and predicted the target feature rougher.output.recovery and final.output.recovery with sMAPE value of 9.58 on the test dataset.
Overall, the project was successful in achieving the goal of building an accurate model to predict the target feature and providing useful insights into the data.