darkreactions / ESCALATE_report

Transform experimental data into ML ready datasets!
http://escalation.sd2e.org/
MIT License
2 stars 1 forks source link

Include ML model of origin in the perovskitedata.csv #27

Closed cinaljess closed 5 years ago

cinaljess commented 5 years ago

It would be a useful feature to know the ML model of origin from which each experiment was suggested from. This would enable learning on just a single model if desired, not incorporating training data suggested by other models.

Add a column to the perovskitedata.csv file that contains the name of the model of origin, eg from the 'Name' column on escalation.

ipendlet commented 5 years ago

Column added _raw_modelname to the dataset which reports this information where available. Initial copy of this dataframe will be upload for crank 0026 on versioned data repo