TypeError: in method 'GpuIndexIVFFlat_train', argument 3 of type 'float const *' about ada-nets

[Thomas-wyh]

Hi, thank you for your interest in Ada-NETS !

You can try this.
Modify Line72-73 in tool/faiss_search.py

    query_arr = np.load(queryfile).astype('float32')
    doc_arr = np.load(docfile).astype('float32')

You need to make sure that the type of feature matrix of your custom dataset is float32.
Let me know if it makes sense.

[Nigar21]

Thank you!