• The given task in this project is to achieve and carry out data storage and retrieval, exploratory data analysis and model selection, data collection, data cleansing, data visualization, data mapping and deviation calculation, unit testing, and so on.
• To complete this list of tasks, three datasets are provided: train, ideal, and test sets. Correlating the given task to a real world scenario in which a company manufactures electronic devices As a data scientist, my job is to analyze the performance of various components used in their devices and track that data against the optimal reading to predict if a piece of equipment is about to fail. In regards to that, we have the data sets from various experiments and tests conducted on these components, resulting in four training datasets (A) and one test dataset (B). Additionally, we have access to datasets for 50 ideal functions (C), which represent the expected behaviour of the components.
• In this context of the study, we are correlating the given task, and the four training datasets can represent different scenarios or conditions under which the components were tested. Each training dataset, for example, could correspond to different operating parameters, environmental conditions, or variations in the manufacturing process. The test dataset is a separate set of data that is used to assess how well the chosen ideal functions generalise to new data.
• The test dataset can be made up of new measurements or observations that represent a different set of conditions not covered by the training datasets. The datasets for 50 ideal functions represent the expected behaviour or theoretical models of the components. Overall, the goal is to select the best-fit ideal functions that minimise deviations by comparing the training data to the ideal functions, and then use those functions to map and evaluate the test data.
The task is to create a Python program that analyses the performance of electronic components by Using training data, select the four best-fitting functions from the fifty available. Furthermore, the program must use the provided test data to decide whether or not each x-y pair of values can be assigned to one of the four ideal functions. If this is the case, the program must additionally execute the mapping and save it alongside the deviation at hand.
The primary goal of this research is to create a reliable and effective Python program. Based on the training data, the program should select the four best-fit ideal functions from a set of 50 functions and then map the test data to these chosen functions while taking a deviation criterion into account.