Descriptor calculation problem: unable to apply for new dataset

[M-asaki-K]

Hi, Following the tutorial (compositional descriptors calculation for sample dataset, which normally worked), I tried to apply the calculation for another dataset I collected from Materials Project. Unfortunately, the results are all "NaN"s. I'd be appreciated to know if there're some tips to do same calculation for new dataset. sample data

[TsumiNa]

@M-asaki-K Hi! Could you please show me your codes, which can let us reproduce your error.

[M-asaki-K]

Thank you very much for your reply! Here's the sample codes I used. sample codes

[TsumiNa]

@M-asaki-K The date type of composition is incorrect. The acceptable data types is dict or pymatgen.core.composition.PMGComp. Because the CSV format is designed to store table-based information, Any information such as dict will be transformed into str.

[M-asaki-K]

Thank you very much for your kind information!

[TsumiNa]

@M-asaki-K I'd like to close this issue. If you still have questions, feel free to reopen this.