| # Copyright (c) 2024-present, Yumeow. Licensed under the MIT License. | |
| import os | |
| import random | |
| import builtins | |
| __all__ = [ | |
| "softmax", | |
| "seed_all", | |
| "bounded_value", | |
| ] | |
| def bounded_value(value, min, max, default, converter=None): | |
| """Convert a value when requested, then constrain it to an inclusive range.""" | |
| try: | |
| parsed = converter(value) if converter is not None else value | |
| except (TypeError, ValueError): | |
| parsed = default | |
| return builtins.max(min, builtins.min(max, parsed)) | |
| def softmax(x): | |
| import numpy as np | |
| x = np.exp(x - x.max()) | |
| return x / x.sum() | |
| def seed_all(seed): | |
| os.environ["PL_GLOBAL_SEED"] = str(seed) | |
| random.seed(seed) | |
| try: | |
| import numpy as np | |
| np.random.seed(seed) | |
| except ImportError: | |
| pass | |
| try: | |
| import torch | |
| torch.manual_seed(seed) | |
| torch.cuda.manual_seed_all(seed) | |
| torch.cuda.manual_seed(seed) | |
| torch.backends.cudnn.deterministic = True | |
| torch.backends.cudnn.benchmark = False | |
| except ImportError: | |
| pass | |