The method of using python in the browser definitely was not fast for me but when it eventually finished processing, I liked it.
Would maybe be interesting to see some more stats though, like maybe most skipped songs/artists. Top platforms, ip addresses would maybe be interesting too?
It's not very fast but make the whole development process a lot easier because I can use pandas to query the data.
When I played around with the data, I also tried most skipped artists, but it was hard to interpret, because my most skipped artists were also my most listened to artists, because I get more chances to skip song from them if I listen more to them. If you download the dataset as CSV it's also quite easy to do your own queries with pandas or other software.
Yes, I think sharing and comparing it with friends is the most fun part. There is a download plot button at the bottom to download the plots and tables as images. That's the only way I could think of to make it easily shareable.
Comments
The method of using python in the browser definitely was not fast for me but when it eventually finished processing, I liked it.
Would maybe be interesting to see some more stats though, like maybe most skipped songs/artists. Top platforms, ip addresses would maybe be interesting too?
I do kind of wish it was shareable
It's not very fast but make the whole development process a lot easier because I can use pandas to query the data.
When I played around with the data, I also tried most skipped artists, but it was hard to interpret, because my most skipped artists were also my most listened to artists, because I get more chances to skip song from them if I listen more to them. If you download the dataset as CSV it's also quite easy to do your own queries with pandas or other software.
Yes, I think sharing and comparing it with friends is the most fun part. There is a download plot button at the bottom to download the plots and tables as images. That's the only way I could think of to make it easily shareable.