Closed computron closed 6 years ago
We are mostly using matminer example data sets except: 1) 1 simple plot for each example to make sure data format is clear (e.g. pf.xy(([1, 2], [2, 4])) ) 2) that one example where I used boston because I couldn't make heatmap_df look nice with our datasets I can try harder to make a nice looking example for heatmap_df using our dbs instead of boston. For now, please feel free to take it out if that's the only example making matminer_examples dependent on sklearn.
@albalu I didn't look carefully, I guess the first example I clicked was the Boston one and I assumed they were all like that.
I'll leave this issue open, I do think you should
delete the Boston function - the simple data functions are good enough and the Boston one doesn't really help you understand more.
add one based on a materials data set. You might have to have the "z" value be an average. For example, "x" might be spacegroup number or lattice type, "y" might be nsites, and "z" might be the average of all elastic tensors for materials with that spacegroup/nsites combination.
@ardunn @albalu
Should the PlotlyFig examples use the matminer example data sets?