File size: 3,065 Bytes
34f3bc9
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
#!/usr/bin/env python

# Copyright 2024 The HuggingFace Inc. team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

from dataclasses import dataclass, field

import abc
import draccus

from src.dataset.transforms import ImageTransformsConfig
from src.transforms.core import TransformGroup
from src.dataset.video_utils import get_safe_default_codec


@dataclass
class DatasetConfig(draccus.ChoiceRegistry, abc.ABC):
    # You may provide a list of datasets here. `train.py` creates them all and concatenates them. Note: only data
    # keys common between the datasets are kept. Each dataset gets and additional transform that inserts the
    # "dataset_index" into the returned item. The index mapping is made according to the order in which the
    # datasets are provided.
    repo_id: str
    # Root directory where the dataset will be stored (e.g. 'dataset/path').
    root: str | None = None
    episodes: list[int] | None = None
    image_transforms: ImageTransformsConfig = field(default_factory=ImageTransformsConfig)
    revision: str | None = None
    use_imagenet_stats: bool = True
    use_external_stats: bool = False
    external_stats_path: str | None = None
    weight_rules_path: str | None = None
    video_backend: str = field(default_factory=get_safe_default_codec)
    streaming: bool = False
    dist_loading: bool = False
    buffer_size: int = 1024
    action_mode: str = "abs"  # abs | delta
    repack_transforms: TransformGroup = field(default_factory=TransformGroup)
    data_transforms: TransformGroup = field(default_factory=TransformGroup)
    model_transforms: TransformGroup = field(default_factory=TransformGroup)

    def __post_init__(self):
        assert self.action_mode in ['abs', 'delta'], "Either abs or delta for 'action_type'. "
        # The two external-stats knobs must agree, else stats silently go
        # missing or a real custom-stats file is silently ignored.
        if self.use_external_stats and not self.external_stats_path:
            raise ValueError(
                "use_external_stats=True but external_stats_path is None. "
                "Either set external_stats_path, or set use_external_stats=False."
            )
        if (not self.use_external_stats) and self.external_stats_path:
            raise ValueError(
                f"external_stats_path={self.external_stats_path!r} is set but "
                f"use_external_stats=False. The path would be silently ignored. "
                f"Set use_external_stats=True to consume it, or clear the path."
            )