bugfixed: update slurm_train.sh

tools/scripts/slurm_train.sh

@@ -21,4 +21,4 @@ srun -p ${PARTITION} \
     --cpus-per-task=${CPUS_PER_TASK} \
     --kill-on-bad-exit=1 \
     ${SRUN_ARGS} \
-    python -u train.py --launcher slurm ${PY_ARGS} --tcp_port $PORT
+    python -u train.py --launcher slurm --tcp_port $PORT ${PY_ARGS}

tools/train.py

@@ -24,7 +24,7 @@ def parse_config():
 
     parser.add_argument('--batch_size', type=int, default=None, required=False, help='batch size for training')
     parser.add_argument('--epochs', type=int, default=None, required=False, help='number of epochs to train for')
-    parser.add_argument('--workers', type=int, default=4, help='number of workers for dataloader')
+    parser.add_argument('--workers', type=int, default=8, help='number of workers for dataloader')
     parser.add_argument('--extra_tag', type=str, default='default', help='extra tag for this experiment')
     parser.add_argument('--ckpt', type=str, default=None, help='checkpoint to start from')
     parser.add_argument('--pretrained_model', type=str, default=None, help='pretrained_model')