#!/usr/bin/env python3 """MultiTask Dataset module compatible with torch.utils.data.Dataset & DataLoader.""" from __future__ import annotations from pathlib import Path from argparse import Namespace, ArgumentParser from pprint import pprint from natsort import natsorted from loguru import logger import torch as tr import numpy as np from torch.utils.data import Dataset, DataLoader from lovely_tensors import monkey_patch monkey_patch() BuildDatasetTuple = tuple[dict[str, list[Path]], list[str]] MultiTaskItem = tuple[dict[str, tr.Tensor], str, list[str]] # [{task: data}, stem(name) | list[stem(name)], [tasks]] class NpzRepresentation: """Generic Task with data read from/saved to npz files. Tries to read data as-is from disk and store it as well""" def __init__(self, name: str): self.name = name def load_from_disk(self, path: Path) -> tr.Tensor: """Reads the npz data from the disk and transforms it properly""" data = np.load(path, allow_pickle=False) data = data if isinstance(data, np.ndarray) else data["arr_0"] # in case on npz, we need this as well return tr.from_numpy(data) # can be uint8, float16, float32 etc. def save_to_disk(self, data: tr.Tensor, path: Path): """stores this item to the disk which can then be loaded via `load_from_disk`""" np.save(path, data.cpu().numpy(), allow_pickle=False) def plot_fn(self, x: tr.Tensor) -> np.ndarray: """very basic implementation of converting this representation to a viewable image. You should overwrite this""" assert isinstance(x, tr.Tensor), type(x) if len(x.shape) == 2: x = x.unsqueeze(-1) assert len(x.shape) == 3, x.shape # guaranteed to be (H, W, C) at this point if x.shape[-1] != 3: x = x[..., 0:1] if x.shape[-1] == 1: x = x.repeat(1, 1, 3) x = x.nan_to_num(0).cpu().numpy() # guaranteed to be (H, W, 3) at this point hopefully _min, _max = x.min((0, 1), keepdims=True), x.max((0, 1), keepdims=True) if x.dtype != np.uint8: x = np.nan_to_num((x - _min) / (_max - _min) * 255, 0).astype(np.uint8) return x def __repr__(self): return str(self) def __str__(self): return f"{str(type(self)).split('.')[-1][0:-2]}({self.name})" class MultiTaskDataset(Dataset): """ MultiTaskDataset implementation. Reads data from npz files and returns them as a dict. Parameters: - path: Path to the directory containing the npz files. - task_names: List of tasks that are present in the dataset. If set to None, will infer from the files on disk. - handle_missing_data: Modes to handle missing data. Valid options are: - drop: Drop the data point if any of the representations is missing. - fill_none: Fill the missing data with Nones. Expected directory structure: path/ - task_1/0.npz, ..., N.npz - ... - task_n/0.npz, ..., N.npz Names can be in a different format (i.e. 2022-01-01.npz), but must be consistent and equal across all tasks. """ def __init__(self, path: Path, task_names: list[str] | None = None, handle_missing_data: str = "fill_none", files_suffix: str = "npz", task_types: dict[str, type] = None): assert Path(path).exists(), f"Provided path '{path}' doesn't exist!" assert handle_missing_data in ("drop", "fill_none"), f"Invalid handle_missing_data mode: {handle_missing_data}" assert files_suffix == "npz", "Only npz supported right now (though trivial to update)" self.path = Path(path).absolute() self.handle_missing_data = handle_missing_data self.suffix = files_suffix self.files_per_repr, self.file_names = self._build_dataset() if task_types is None: logger.debug("No explicit task types. Defaulting all of them to NpzRepresentation.") task_types = {} if task_names is None: task_names = list(self.files_per_repr.keys()) logger.debug(f"No explicit tasks provided. Using all of them as read from the paths ({len(task_names)}).") self.task_types = {k: task_types.get(k, NpzRepresentation) for k in task_names} assert all(isinstance(x, str) for x in task_names), tuple(zip(task_names, (type(x) for x in task_names))) self.task_names = sorted(task_names) self._data_shape: tuple[int, ...] | None = None self._tasks: list[NpzRepresentation] | None = None self.name_to_task = {task.name: task for task in self.tasks} logger.info(f"Tasks used in this dataset: {self.task_names}") # Public methods and properties @property def data_shape(self) -> dict[str, tuple[int, ...]]: """Returns a {task: shape_tuple} for all representations. At least one npz file must exist for each.""" first_npz = {task: [_v for _v in files if _v is not None][0] for task, files in self.files_per_repr.items()} data_shape = {task: self.name_to_task[task].load_from_disk(first_npz[task]).shape for task in self.task_names} return data_shape @property def tasks(self) -> list[NpzRepresentation]: """ Returns a list of instantiated tasks in the same order as self.task_names. Overwrite this to add new tasks and semantics (i.e. plot_fn or doing some preprocessing after loading from disk in some tasks. """ if self._tasks is not None: return self._tasks self._tasks = [self.task_types[task_name](task_name) for task_name in self.task_names] return self._tasks def collate_fn(self, items: list[MultiTaskItem]) -> MultiTaskItem: """ given a list of items (i.e. from a reader[n:n+k] call), return the item batched on 1st dimension. Nones (missing data points) are turned into zeros as per the data shape of that dim. """ assert all(item[2] == self.task_names for item in items), ((item[2] for item in items), self.task_names) items_name = [item[1] for item in items] res = {k: tr.zeros(len(items), *self.data_shape[k]).float() for k in self.task_names} # float32 always for i in range(len(items)): for k in self.task_names: res[k][i] = items[i][0][k] if items[i][0][k] is not None else 0 return res, items_name, self.task_names # Private methods def _get_all_npz_files(self) -> dict[str, list[Path]]: """returns a dict of form: {"rgb": ["0.npz", "1.npz", ..., "N.npz"]}""" in_files = {} all_repr_dirs: list[str] = [x.name for x in self.path.iterdir() if x.is_dir()] for repr_dir_name in all_repr_dirs: dir_name = self.path / repr_dir_name if all(f.is_dir() for f in dir_name.iterdir()): # dataset is stored as repr/part_x/0.npz, ..., part_k/n.npz all_files = [] for part in dir_name.iterdir(): all_files.extend(part.glob(f"*.{self.suffix}")) else: # dataset is stored as repr/0.npz, ..., repr/n.npz all_files = dir_name.glob(f"*.{self.suffix}") in_files[repr_dir_name] = natsorted(all_files, key=lambda x: x.name) # important: use natsorted() here assert not any(len(x) == 0 for x in in_files.values()), f"{ [k for k, v in in_files.items() if len(v) == 0] }" return in_files def _build_dataset_drop(self) -> BuildDatasetTuple: in_files = self._get_all_npz_files() name_to_node_path = {k: {_v.name: _v for _v in v} for k, v in in_files.items()} # {node: {name: path}} common = set(x.name for x in next(iter(in_files.values()))) nodes = in_files.keys() for node in nodes: common = common.intersection([f.name for f in in_files[node]]) assert len(common) > 0, f"Node '{node}' made the intersection null" common = natsorted(list(common)) logger.info(f"Found {len(common)} data points for each node ({len(nodes)} nodes).") files_per_repr = {node: [name_to_node_path[node][x] for x in common] for node in nodes} assert len(files_per_repr) > 0 return files_per_repr, common def _build_dataset_fill_none(self) -> BuildDatasetTuple: in_files = self._get_all_npz_files() name_to_node_path = {k: {_v.name: _v for _v in v} for k, v in in_files.items()} all_files = set(x.name for x in next(iter(in_files.values()))) nodes = in_files.keys() for node in nodes: all_files = all_files.union([f.name for f in in_files[node]]) all_files = natsorted(list(all_files)) logger.info(f"Found {len(all_files)} data points as union of all nodes' data ({len(nodes)} nodes).") files_per_repr = {node: [] for node in nodes} in_file_names = {node: [f.name for f in in_files[node]] for node in nodes} for node in nodes: for file_name in all_files: file_path = name_to_node_path[node].get(file_name, None) files_per_repr[node].append(file_path) assert len(files_per_repr) > 0 return files_per_repr, all_files def _build_dataset(self) -> BuildDatasetTuple: logger.debug(f"Building dataset from: '{self.path}'") if self.handle_missing_data == "drop": return self._build_dataset_drop() else: return self._build_dataset_fill_none() # Python magic methods (pretty printing the reader object, reader[0], len(reader) etc.) def __getitem__(self, index: int | slice | list[int] | tuple) -> MultiTaskItem: """Read the data all the desired nodes""" assert isinstance(index, (int, slice, list, tuple)), type(index) if isinstance(index, slice): assert index.start is not None and index.stop is not None and index.step is None, "Only reader[l:r] allowed" index = list(range(index.stop)[index]) if isinstance(index, (list, tuple)): return self.collate_fn([self.__getitem__(ix) for ix in index]) res = {} item_name = self.file_names[index] for _repr in self.tasks: file_path = self.files_per_repr[_repr.name][index] file_path = file_path.resolve() if file_path is not None else None assert self.handle_missing_data == "fill_none" or (file_path is not None and file_path.exists()), item_name item = _repr.load_from_disk(file_path) if file_path is not None and file_path.exists() else None res[_repr.name] = item return (res, item_name, self.task_names) def __len__(self) -> int: return len(self.files_per_repr[self.task_names[0]]) # all of them have the same number (filled with None or not) def __str__(self): f_str = f"[{str(type(self)).rsplit('.', maxsplit=1)[-1][0:-2]}]" f_str += f"\n - Path: '{self.path}'" f_str += f"\n - Only full data: {self.handle_missing_data == 'drop'}" f_str += f"\n - Representations ({len(self.tasks)}): {self.tasks}" f_str += f"\n - Length: {len(self)}" return f_str def __repr__(self): return str(self) def main(): """main fn""" parser = ArgumentParser() parser.add_argument("dataset_path", type=Path) parser.add_argument("--handle_missing_data", choices=("drop", "fill_none"), default="fill_none") args = parser.parse_args() reader = MultiTaskDataset(args.dataset_path, task_names=None, handle_missing_data=args.handle_missing_data) print(reader) print(f"Shape: {reader.data_shape}") rand_ix = np.random.randint(len(reader)) data, name, repr_names = reader[rand_ix] # get a random single data point print(f"Name: {name}. Nodes: {repr_names}") pprint({k: v for k, v in data.items()}) data, name, repr_names = reader[rand_ix: min(len(reader), rand_ix + 5)] # get a random batch print(f"Name: {name}. Nodes: {repr_names}") pprint({k: v for k, v in data.items()}) # Nones are converted to 0s automagically loader = DataLoader(reader, collate_fn=reader.collate_fn, batch_size=5, shuffle=True) data, name, repr_names = next(iter(loader)) # get a random batch using torch DataLoader print(f"Name: {name}. Nodes: {repr_names}") pprint({k: v for k, v in data.items()}) # Nones are converted to 0s automagically if __name__ == "__main__": main()