Default Script Fails when Run with Testing/Training Data

[cai-michael]

When running the default test script python scripts/netTCR.py -infile test_data/data.txt -outfile NetTCR_predictions.txt using either Python 2.7.17 or Python 3.6.9 I get the following output:

Traceback (most recent call last): File "scripts/netTCR.py", line 170, in <module> best_params = np.load(param_dir + "params.t." + str(t) + ".v." + str(v) + ".s." + str(s) + ".npz")['arr_0'] File "/home/local/ASUAD/mrcai/.local/lib/python2.7/site-packages/numpy/lib/npyio.py", line 262, in __getitem__ pickle_kwargs=self.pickle_kwargs) File "/home/local/ASUAD/mrcai/.local/lib/python2.7/site-packages/numpy/lib/format.py", line 696, in read_array raise ValueError("Object arrays cannot be loaded when " ValueError: Object arrays cannot be loaded when allow_pickle=False

As a workaround at the moment you can downgrade numpy to 1.16.1 to be able to run the script successfully.

I think the default value in numpy has changed and allow_pickle needs to be directly assigned as True.

[cai-michael]

netTCR/parameters/iedb_mira_pos_uniq.txt

Line 3092 in 070fda1

GILGFVFTL ASSTRS1DTQY HLA-A02:01 1

netTCR/parameters/iedb_mira_pos_uniq.txt

Line 3012 in 070fda1

GILGFVFTL ASSTG7SWEOY HLA-A02:01 1

netTCR/parameters/iedb_mira_pos_uniq.txt

Line 1594 in 070fda1

GILGFVFTL ASSIRSS__QYF HLA-A02:01 1

There's also some training data that seems to have numbers that don't belong when I try to run the data on the training set even.