Abstract: An approach for intelligent optimization of machine learning models for a target environment may be provided herein. The approach may include extracting metadata from a training pipeline for a first machine learning model that has been configured to operate within a first computing environment. The approach may also include mapping the extracted metadata to one or more constraints associated with a second machine learning model that has been configured to operate within a second computing environment. The approach may also include training the second machine learning model configured to the second computing environment, with the dataset that was used to train the first machine learning model, based on the mapped constraints. The approach may also include comparing performance metrics of the first machine learning model to corresponding metrics of the now trained second machine learning model.

Abstract: A method for encoding 2D numerical data comprises determining encoding parameters for a received set of 2D numerical data and generating a set of encoded data from the set of 2D numerical data according to the encoding parameters. The encoding parameters indicate a transitional relationship among a plurality of consecutive data points and a unitization interval for sampling the first set of 2D numerical data. When generating the set of encoded data, the encoding method sets a starting point, samples the set of 2D numerical data according to the unitization interval, and determines a string as a value of each data point of the set of the encoded data. The string indicates a position of a present encoded data point in the set of the encoded data, the transitional relationship, and a difference of magnitude between a present encoded data point and an immediately preceding one.

Abstract: An approach for intelligent optimization of machine learning models for a target environment may be provided herein. The approach may include extracting metadata from a training pipeline for a first machine learning model that has been configured to operate within a first computing environment. The approach may also include mapping the extracted metadata to one or more constraints associated with a second machine learning model that has been configured to operate within a second computing environment. The approach may also include training the second machine learning model configured to the second computing environment, with the dataset that was used to train the first machine learning model, based on the mapped constraints. The approach may also include comparing performance metrics of the first machine learning model to corresponding metrics of the now trained second machine learning model.