|
|
|
|
|
import argparse |
|
|
|
parser = argparse.ArgumentParser() |
|
|
|
parser.add_argument("--pretrain", default=False, action="store_true", help='use vqa2.0 or not') |
|
parser.add_argument("--gpt3", default=False, action="store_true", help='use gpt3 to train on okvqa') |
|
parser.add_argument("--visualBERT", default=False, action="store_true", help='use visualBERT, if false use LXMERT') |
|
|
|
parser.add_argument('--batch_size', type=int, default=128, |
|
help='minibatch size') |
|
parser.add_argument('--seed', type=int, default=4, |
|
help='random seed!') |
|
parser.add_argument('--num_wiki', type=int, default=25, |
|
help='the number of wiki passages') |
|
parser.add_argument('--num_epochs', type=int, default=40, |
|
help='number of epochs') |
|
parser.add_argument('--learning_rate', type=float, default=0.0001, |
|
help='LR') |
|
parser.add_argument('--learning_rate_LXM', type=float, default=0.00001, |
|
help='LR_LXM') |
|
parser.add_argument('--model_dir', type=str, default='xxx/', |
|
help='model file path') |
|
parser.add_argument('--input_type', type=int, default=1, |
|
help='input types: 1==Q-OFA-C-L-O; 2==Q-C-L-O; 3==Q-OFA-L-O; 4==Q-OFA-C-O; 5==Q-OFA-C-L') |
|
parser.add_argument('--describe', type=str, default='', |
|
help='the model description used as the saved-model name') |
|
parser.add_argument("--load_pthpath", default="", |
|
help="To continue training, path to .pth file of saved checkpoint.") |
|
parser.add_argument("--validate", default='True', action="store_true", help="Whether to validate on val split after every epoch.") |
|
parser.add_argument("--dataset", default="okvqa", help="dataset that model training on") |
|
parser.add_argument("--ofa", default="normal", help=" normal or finetune --- load the knowledge from Normal OFA or vqav2-Finetuned OFA") |
|
parser.add_argument('--local_rank', default=-1, type=int, |
|
help='node rank for distributed training') |
|
args = parser.parse_args() |
|
print(args) |
|
|
