DeepRank
data_set definition in "Tutorial: Deep Learning" is obsolete
Describe the problem
The data_set definition in "Tutorial: Deep Learning" is obsolete, not up-to-date with the data_set definition present in test/test_learn.py.
Environment:
- OS system: Linux
To Reproduce
Steps/commands to reproduce the behaviour:
Run:
data_set = DataSet(database, grid_info={ 'number_of_points': (10, 10, 10), 'resolution': (3, 3, 3)}, select_feature={ 'AtomicDensities' : 'all', 'Features' : ['coulomb','vdwaals','charge','pssm']}, pair_chain_feature=np.add, select_target='IRMSD', dict_filter={'IRMSD':'<4. or >10.'})
Actual Results or Error Info If the data_set is defined as shown in the Tutorial, it raises the following error:
Traceback (most recent call last): File "dl_try.py", line 72, in <module> dict_filter={'IRMSD':'<4. or >10.'}) File "/home/dariomarzella/deeprank/deeprank/learn/DataSet.py", line 202, in __init__ self.process_dataset() File "/home/dariomarzella/deeprank/deeprank/learn/DataSet.py", line 280, in process_dataset self.get_input_shape() File "/home/dariomarzella/deeprank/deeprank/learn/DataSet.py", line 762, in get_input_shape feature, _ = self.map_one_molecule(fname) File "/home/dariomarzella/deeprank/deeprank/learn/DataSet.py", line 1163, in map_one_molecule feat_names, mol_data, grid, npts, angle, axis) File "/home/dariomarzella/deeprank/deeprank/learn/DataSet.py", line 1378, in map_feature data = np.array(mol_data['features/' + name][()]) File "h5py/_objects.pyx", line 54, in h5py._objects.with_phil.wrapper File "h5py/_objects.pyx", line 55, in h5py._objects.with_phil.wrapper File "/home/dariomarzella/anaconda3/lib/python3.7/site-packages/h5py/_hl/group.py", line 262, in __getitem__ oid = h5o.open(self.id, self._e(name), lapl=self._lapl) File "h5py/_objects.pyx", line 54, in h5py._objects.with_phil.wrapper File "h5py/_objects.pyx", line 55, in h5py._objects.with_phil.wrapper File "h5py/h5o.pyx", line 190, in h5py.h5o.open KeyError: "Unable to open object (object 'pssm' doesn't exist)"
