Machine learning based identification of superconductors

The following are contained:

• Python code to generate features to input into machine learning models for superconducting critical temperatures, as well as the code to implement the machine learning models. 
• Chemical compositions, critical temperatues, and pressures at which the critcial temperatures were measured ("0" indicates ambient pressure, "1" indicates applied pressure) of materials in our cleaned SuperCon data set. 
• Critical temeprature predictions, weight coefficients, and feature-weight products for SuperCon materials at implicit pressure and ambient pressure (made only for those samples with pressures of "0")
• Chemical compositions, identifiers, energies above convex hulls, band gaps, and machine learning features for samples in Materials Project. 
• Critical temperature predictions, weight coefficients, and feature-weight products for samples in Materials Project at implicit pressure and ambient pressure.