args.py

import argparse
import os

PRESAVE_DIR = ""
MODEL_DIR = ""
DATA_DIR = ""
SSD_DIR = ""
name2folder = {
    "webvid": "WebVid",
    "lsmdc": "LSMDC",
    "ivqa": "iVQA",
    "msrvtt": "MSRVTT-QA",
    "msvd": "MSVD-QA",
    "activitynet": "ActivityNet-QA",
    "tgif": "TGIF-QA",
    "how2qa": "How2QA",
    "tvqa": "TVQA",
    "vqa": "VQA",
}


def get_args_parser():
    parser = argparse.ArgumentParser("Set FrozenBiLM", add_help=False)

    # Dataset specific
    parser.add_argument(
        "--combine_datasets",
        nargs="+",
        help="list of datasets to combine for training",
        required=True,
    )
    parser.add_argument(
        "--combine_datasets_val",
        nargs="+",
        help="list of datasets to combine for eval",
        required=True,
    )
    parser.add_argument(
        "--webvid_features_path",
        default=os.path.join(SSD_DIR, "webvid_clipvitl14_features"),
    )
    parser.add_argument(
        "--webvid_train_csv_path",
        default=os.path.join(DATA_DIR, name2folder["webvid"], "train_captions.csv"),
    )
    parser.add_argument(
        "--webvid_val_csv_path",
        default=os.path.join(DATA_DIR, name2folder["webvid"], "val_captions.csv"),
    )
    parser.add_argument(
        "--lsmdc_features_path",
        default=os.path.join(DATA_DIR, name2folder["lsmdc"], "clipvitl14.pth"),
    )
    parser.add_argument(
        "--lsmdc_train_csv_path",
        default=os.path.join(DATA_DIR, name2folder["lsmdc"], "training.csv"),
    )
    parser.add_argument(
        "--lsmdc_val_csv_path",
        default=os.path.join(DATA_DIR, name2folder["lsmdc"], "val.csv"),
    )
    parser.add_argument(
        "--lsmdc_test_csv_path",
        default=os.path.join(DATA_DIR, name2folder["lsmdc"], "test.csv"),
    )
    parser.add_argument(
        "--lsmdc_vocab_path",
        default=os.path.join(DATA_DIR, name2folder["lsmdc"], "vocab.json"),
    )
    parser.add_argument(
        "--lsmdc_subtitles_path",
        default=os.path.join(DATA_DIR, name2folder["lsmdc"], "subtitles.pkl"),
    )
    parser.add_argument(
        "--ivqa_features_path",
        default=os.path.join(DATA_DIR, name2folder["ivqa"], "clipvitl14.pth"),
    )
    parser.add_argument(
        "--ivqa_train_csv_path",
        default=os.path.join(DATA_DIR, name2folder["ivqa"], "train.csv"),
    )
    parser.add_argument(
        "--ivqa_val_csv_path",
        default=os.path.join(DATA_DIR, name2folder["ivqa"], "val.csv"),
    )
    parser.add_argument(
        "--ivqa_test_csv_path",
        default=os.path.join(DATA_DIR, name2folder["ivqa"], "test.csv"),
    )
    parser.add_argument(
        "--ivqa_vocab_path",
        default=os.path.join(DATA_DIR, name2folder["ivqa"], "vocab.json"),
    )
    parser.add_argument(
        "--ivqa_subtitles_path",
        default=os.path.join(DATA_DIR, name2folder["ivqa"], "subtitles.pkl"),
    )
    parser.add_argument(
        "--msrvtt_features_path",
        default=os.path.join(DATA_DIR, name2folder["msrvtt"], "clipvitl14.pth"),
    )
    parser.add_argument(
        "--msrvtt_train_csv_path",
        default=os.path.join(DATA_DIR, name2folder["msrvtt"], "train.csv"),
    )
    parser.add_argument(
        "--msrvtt_val_csv_path",
        default=os.path.join(DATA_DIR, name2folder["msrvtt"], "val.csv"),
    )
    parser.add_argument(
        "--msrvtt_test_csv_path",
        default=os.path.join(DATA_DIR, name2folder["msrvtt"], "test.csv"),
    )
    parser.add_argument(
        "--msrvtt_vocab_path",
        default=os.path.join(DATA_DIR, name2folder["msrvtt"], "vocab.json"),
    )
    parser.add_argument(
        "--msrvtt_subtitles_path",
        default=os.path.join(DATA_DIR, name2folder["msrvtt"], "subtitles.pkl"),
    )
    parser.add_argument(
        "--msvd_features_path",
        default=os.path.join(DATA_DIR, name2folder["msvd"], "clipvitl14.pth"),
    )
    parser.add_argument(
        "--msvd_train_csv_path",
        default=os.path.join(DATA_DIR, name2folder["msvd"], "train.csv"),
    )
    parser.add_argument(
        "--msvd_val_csv_path",
        default=os.path.join(DATA_DIR, name2folder["msvd"], "val.csv"),
    )
    parser.add_argument(
        "--msvd_test_csv_path",
        default=os.path.join(DATA_DIR, name2folder["msvd"], "test.csv"),
    )
    parser.add_argument(
        "--msvd_vocab_path",
        default=os.path.join(DATA_DIR, name2folder["msvd"], "vocab.json"),
    )
    parser.add_argument(
        "--msvd_subtitles_path",
        default=os.path.join(DATA_DIR, name2folder["msvd"], "subtitles.pkl"),
    )
    parser.add_argument(
        "--activitynet_features_path",
        default=os.path.join(DATA_DIR, name2folder["activitynet"], "clipvitl14.pth"),
    )
    parser.add_argument(
        "--activitynet_train_csv_path",
        default=os.path.join(DATA_DIR, name2folder["activitynet"], "train.csv"),
    )
    parser.add_argument(
        "--activitynet_val_csv_path",
        default=os.path.join(DATA_DIR, name2folder["activitynet"], "val.csv"),
    )
    parser.add_argument(
        "--activitynet_test_csv_path",
        default=os.path.join(DATA_DIR, name2folder["activitynet"], "test.csv"),
    )
    parser.add_argument(
        "--activitynet_vocab_path",
        default=os.path.join(DATA_DIR, name2folder["activitynet"], "vocab.json"),
    )
    parser.add_argument(
        "--activitynet_subtitles_path",
        default=os.path.join(DATA_DIR, name2folder["activitynet"], "subtitles.pkl"),
    )
    parser.add_argument(
        "--tgif_features_path",
        default=os.path.join(DATA_DIR, name2folder["tgif"], "clipvitl14.pth"),
    )
    parser.add_argument(
        "--tgif_frameqa_train_csv_path",
        default=os.path.join(DATA_DIR, name2folder["tgif"], "train_frameqa.csv"),
    )
    parser.add_argument(
        "--tgif_frameqa_test_csv_path",
        default=os.path.join(DATA_DIR, name2folder["tgif"], "test_frameqa.csv"),
    )
    parser.add_argument(
        "--tgif_vocab_path",
        default=os.path.join(DATA_DIR, name2folder["tgif"], "vocab.json"),
    )
    parser.add_argument(
        "--how2qa_features_path",
        default=os.path.join(DATA_DIR, name2folder["how2qa"], "clipvitl14_split.pth"),
    )
    parser.add_argument(
        "--how2qa_train_csv_path",
        default=os.path.join(DATA_DIR, name2folder["how2qa"], "train.csv"),
    )
    parser.add_argument(
        "--how2qa_val_csv_path",
        default=os.path.join(DATA_DIR, name2folder["how2qa"], "public_val.csv"),
    )
    parser.add_argument(
        "--how2qa_subtitles_path",
        default=os.path.join(DATA_DIR, name2folder["how2qa"], "subtitles.pkl"),
    )
    parser.add_argument(
        "--tvqa_features_path",
        default=os.path.join(DATA_DIR, name2folder["tvqa"], "clipvitl14.pth"),
    )
    parser.add_argument(
        "--tvqa_train_csv_path",
        default=os.path.join(DATA_DIR, name2folder["tvqa"], "train.csv"),
    )
    parser.add_argument(
        "--tvqa_val_csv_path",
        default=os.path.join(DATA_DIR, name2folder["tvqa"], "val.csv"),
    )
    parser.add_argument(
        "--tvqa_test_csv_path",
        default=os.path.join(DATA_DIR, name2folder["tvqa"], "test_public.csv"),
    )
    parser.add_argument(
        "--tvqa_subtitles_path",
        default=os.path.join(DATA_DIR, name2folder["tvqa"], "subtitles.pkl"),
    )
    parser.add_argument(
        "--vqa_features_path",
        default=os.path.join(DATA_DIR, name2folder["vqa"], "clipvitl14.pth"),
    )
    parser.add_argument(
        "--vqa_train_pkl_path",
        default=os.path.join(DATA_DIR, name2folder["vqa"], "train_list.pkl"),
    )
    parser.add_argument(
        "--vqa_val_pkl_path",
        default=os.path.join(DATA_DIR, name2folder["vqa"], "val_list.csv"),
    )
    parser.add_argument(
        "--vqa_vocab_path",
        default=os.path.join(DATA_DIR, name2folder["vqa"], "vocab.json"),
    )

    # Training hyper-parameters
    parser.add_argument(
        "--mlm_prob",
        type=float,
        default=0.15,
        help="masking probability for the MLM objective",
    )
    parser.add_argument("--lr", default=3e-4, type=float, help="learning rate")
    parser.add_argument(
        "--beta1", default=0.9, type=float, help="Adam optimizer parameter"
    )
    parser.add_argument(
        "--beta2", default=0.95, type=float, help="Adam optimizer parameter"
    )
    parser.add_argument(
        "--batch_size", default=32, type=int, help="batch size used for training"
    )
    parser.add_argument(
        "--batch_size_val",
        default=32,
        type=int,
        help="batch size used for eval",
    )
    parser.add_argument("--weight_decay", default=0, type=float)
    parser.add_argument(
        "--epochs", default=10, type=int, help="number of training epochs"
    )
    parser.add_argument(
        "--lr_drop",
        default=10,
        type=int,
        help="number of epochs after which the learning rate is reduced when not using linear decay",
    )
    parser.add_argument("--optimizer", default="adam", type=str)
    parser.add_argument(
        "--clip_max_norm", default=0.1, type=float, help="gradient clipping max norm"
    )
    parser.add_argument(
        "--schedule",
        default="",
        choices=["", "linear_with_warmup"],
        help="learning rate decay schedule, default is constant",
    )
    parser.add_argument(
        "--fraction_warmup_steps",
        default=0.1,
        type=float,
        help="fraction of number of steps used for warmup when using linear schedule",
    )
    parser.add_argument(
        "--eval_skip",
        default=1,
        type=int,
        help='do evaluation every "eval_skip" epochs',
    )
    parser.add_argument(
        "--print_freq",
        type=int,
        default=100,
        help="print log every print_freq iterations",
    )

    # Model parameters
    parser.add_argument(
        "--ft_lm",
        dest="freeze_lm",
        action="store_false",
        help="whether to finetune the weights of the language model",
    )
    parser.add_argument(
        "--model_name",
        default="deberta-v2-xlarge",
        choices=(
            "bert-base-uncased",
            "bert-large-uncased",
            "deberta-v2-xlarge",
            "gpt-neo-1p3b",
            "gpt-j-6b",
            "gpt-neo-2p7b",
        ),
    )
    parser.add_argument(
        "--ds_factor_attn",
        type=int,
        default=0,
        help="downsampling factor for adapter attn",
    )
    parser.add_argument(
        "--ds_factor_ff",
        type=int,
        default=0,
        help="downsampling factor for adapter ff",
    )
    parser.add_argument(
        "--freeze_ln",
        dest="ft_ln",
        action="store_false",
        help="whether or not to freeze layer norm parameters",
    )
    parser.add_argument(
        "--ft_mlm",
        dest="freeze_mlm",
        action="store_false",
        help="whether or not to finetune the mlm head parameters",
    )
    parser.add_argument(
        "--dropout", default=0.1, type=float, help="dropout to use in the adapter"
    )
    parser.add_argument(
        "--scratch",
        action="store_true",
        help="whether to train the LM with or without language init",
    )
    parser.add_argument(
        "--n_ans",
        type=int,
        default=0,
        help="number of answers in the answer embedding module, it is automatically set",
    )
    parser.add_argument(
        "--ft_last",
        dest="freeze_last",
        action="store_false",
        help="whether to finetune answer embedding module or not",
    )

    # Run specific
    parser.add_argument(
        "--test",
        action="store_true",
        help="whether to run evaluation on val or test set",
    )
    parser.add_argument(
        "--save_dir", default="", help="path where to save, empty for no saving"
    )
    parser.add_argument(
        "--presave_dir",
        default=PRESAVE_DIR,
        help="the actual save_dir is an union of presave_dir and save_dir",
    )
    parser.add_argument("--device", default="cuda", help="device to use")
    parser.add_argument("--seed", default=42, type=int, help="random seed")
    parser.add_argument(
        "--load",
        default="",
        help="path to load checkpoint",
    )
    parser.add_argument(
        "--resume",
        action="store_true",
        help="continue training if loading checkpoint",
    )
    parser.add_argument(
        "--start-epoch", default=0, type=int, metavar="N", help="start epoch"
    )
    parser.add_argument("--eval", action="store_true", help="only run evaluation")
    parser.add_argument(
        "--num_workers", default=3, type=int, help="number of workers for dataloader"
    )

    # Distributed training parameters
    parser.add_argument(
        "--world-size", default=1, type=int, help="number of distributed processes"
    )
    parser.add_argument(
        "--dist-url", default="env://", help="url used to set up distributed training"
    )

    # Video and Text parameters
    parser.add_argument(
        "--max_feats",
        type=int,
        default=10,
        help="maximum number of video features considered, one per frame",
    )
    parser.add_argument(
        "--features_dim",
        type=int,
        default=768,
        help="dimension of the visual embedding space",
    )
    parser.add_argument(
        "--no_video",
        dest="use_video",
        action="store_false",
        help="disables usage of video",
    )
    parser.add_argument(
        "--no_context",
        dest="use_context",
        action="store_false",
        help="disables usage of speech",
    )
    parser.add_argument(
        "--max_tokens",
        type=int,
        default=64,
        help="maximum number of tokens in the input text prompt",
    )
    parser.add_argument(
        "--max_atokens",
        type=int,
        default=5,
        help="maximum number of tokens in the answer",
    )
    parser.add_argument(
        "--prefix",
        default="",
        type=str,
        help="task induction before question for videoqa",
    )
    parser.add_argument(
        "--suffix",
        default="",
        type=str,
        help="suffix after the answer mask for videoqa",
    )

    # Demo
    parser.add_argument(
        "--question_example",
        default="",
        type=str,
        help="question example for demo",
    )
    parser.add_argument(
        "--video_example",
        default="",
        type=str,
        help="path to a video example for demo",
    )

    return parser