add a command to docker-compose

docker-compose-preprocess.yml

@@ -9,7 +9,8 @@ services:
       - "/home/cbaakman/deeprank-datasets:/datasets"
       - "/home/cbaakman/pdb:/pdb"
       - "/home/cbaakman/pssm-gen:/pssm-gen"
-    working_dir: /runs/run-mp
+    working_dir: /runs/run2
+    command: mpiexec -n 10 --allow-run-as-root /usr/src/app/scripts/preprocess_bioprodict.py /datasets/54e500aa7123e7f2f6ca4062db0e4fa4_efro_t96_25052021.parq /datasets/GPCR_variantsv2_increased_coverage.hdf5 /pdb /pssm-gen variants.hdf5
     environment:
         - MASTER_ADDR=0.0.0.0
         - MASTER_PORT=1088

scripts/preprocess_bioprodict.py

@@ -38,7 +38,7 @@
 arg_parser.add_argument("pdb_root", help="the path to the pdb root directory")
 arg_parser.add_argument("pssm_root", help="the path to the pssm root directory, containing files generated with PSSMgen")
 arg_parser.add_argument("out_path", help="the path to the output hdf5 file")
-arg_parser.add_argument("-A", "--data-augmentation", help="the number of data augmentations", type=int, default=0)
+arg_parser.add_argument("-A", "--data-augmentation", help="the number of data augmentations", type=int, default=5)
 arg_parser.add_argument("-p", "--grid-points", help="the number of points per edge of the 3d grid", type=int, default=20)
 arg_parser.add_argument("-S", "--grid-size", help="the length in Angstroms of each edge of the 3d grid", type=float, default=20)