How to prepare data

[muammar]

Hi, thanks for putting this together. In your notebook with an example, I see that you are using the dataset you already processed. I wondered if you could list what the procedure to follow is. I would like to use your implementation.

Best,

[aksub99]

Hi @muammar thanks for your question! Preprocessing was done using the preprocess.py script from https://github.com/maxhodak/keras-molecules. Let me know if you have any other questions!

[muammar]

Hi @muammar thanks for your question! Preprocessing was done using the preprocess.py script from https://github.com/maxhodak/keras-molecules. Let me know if you have any other questions!

Thanks for your reply @aksub99 ! I tried using the repo you posted above and I get this error:

Traceback (most recent call last):
  File "preprocess.py", line 86, in <module>
    main()
  File "preprocess.py", line 59, in main
    h5f.create_dataset('charset', data = charset)
  File "/home/elkhatim/miniconda3/envs/py38/lib/python3.8/site-packages/h5py/_hl/group.py", line 136, in create_dataset
    dsid = dataset.make_new_dset(self, shape, dtype, data, **kwds)
  File "/home/elkhatim/miniconda3/envs/py38/lib/python3.8/site-packages/h5py/_hl/dataset.py", line 118, in make_new_dset
    tid = h5t.py_create(dtype, logical=1)
  File "h5py/h5t.pyx", line 1634, in h5py.h5t.py_create
  File "h5py/h5t.pyx", line 1656, in h5py.h5t.py_create
  File "h5py/h5t.pyx", line 1717, in h5py.h5t.py_create
TypeError: No conversion path for dtype: dtype('<U1')

I am asking you because it seems to me that keras-molecules is kind of abandoned... I wondered if you had any idea about this TypeError?

Best,

[aksub99]

I actually remember getting the same error when I had used the code. I can't recall exactly how I fixed it but I vaguely remember that in lines 35 and 38 of preprocess.py, I had made some simple typecasting fixes. So, I'd direct you towards those lines. Hope this helps!