Cold spray is a high-speed solid state deposition technique which allows for fabricating coatings and free standings structures by careful tuning of process parameters. However due to the complex dependencies of the process parameters a machine learning approach is utilized in this work to predict the coating properties. A machine learning (ML)-based data driven platform for determining the deposition porosity for cold sprayed deposition with the goal of reducing product cost and time has been developed. In this work, five ML models - Linear Regression, Decision Tree Regression and Random Forest Regression, XGBRegressor and LGBMRegressor are compared for the prediction of coating porosity from spray and material parameters namely Heat Capacity Ratio of carrier gas, Processing Gas Temperature, Processing gas pressure, Standoff Distance, Average Powder Diameter, Powder Material Density, and the Substrate density. A total number of 227 data sets were extracted from an extensive literature survey on cold spray deposition of metal/alloy powders which were used to train the ML models. The data analysis showed strong and weak correlations of several processing parameters with the coating porosity. The processing gas temperature and pressure have a negative and average powder diameter has a positive correlation with the deposition porosity. The coating porosities of 10 unknown sets (which were not included in the training or the validation data sets) of processing parameters predicted by the trained algorithms were compared with each other. Decision Tree regression algorithm showed the most appropriate predictions with R2 Fit of 0.75 and MAE of 2.93, while the Linear Regression model did the worst predictions with R2 Fit of 0.27 and MAE of 5.064.