| import torch | |
| import torch.nn | |
| from torch import nn | |
| from torch.utils.data import Dataset | |
| import numpy as np | |
| import pytorch_lightning as pl | |
| from pytorch_lightning.cli import LightningCLI | |
| from typing import List, Dict |
| import jsonargparse | |
| import inspect | |
| from jsonargparse.parameter_resolvers import ParamData | |
| from jsonargparse.signatures import get_signature_parameters | |
| from jsonargparse.signatures import get_doc_short_description | |
| from typing import List, Set, Union, Optional, Tuple, Type, Any | |
| from jsonargparse.util import get_import_path, iter_to_set_str | |
| from argparse import SUPPRESS | |
| from jsonargparse.typing import is_final_class |
| """ | |
| make VERSION=us -j16 | |
| build/us_pc/sm64.us | |
| SeeAlso: | |
| https://github.com/TechieAndroid/sm64redrawn | |
| """ | |
| import ubelt as ub | |
| import aifc | |
| import kwimage |
| # syntax=docker/dockerfile:1.3.0-labs | |
| # This dockerfile uses new-ish buildkit syntax. | |
| # Details on how to run are on the bottom of the file. | |
| # (docker devs: todo unconsequential heredocs) | |
| FROM nvidia/cuda:11.4.3-cudnn8-devel-ubuntu20.04 | |
| ARG PYTHON_VERSION=3.10.5 | |
| ARG PYENV_VERSION=v2.3.3 |
An fully FOSS alternative to a conda environment is pyenv. The only real drawback to pyenv is that you must be able to compile Python itself.
On ubuntu this can be setup as follows:
### Based on code in https://github.com/Erotemic/local/blob/main/tools/pyenv_ext/pyenv_ext_commands.sh
| """ | |
| Add NaN parsing to normal ijson pure-python parsing backend. | |
| Example: | |
| >>> import ijson_ext as ijson | |
| >>> text = ub.codeblock( | |
| >>> ''' | |
| >>> {"id": 246, "image_id": 123, "flag": true, "empty": null, "category_id": 2, "bbox": [ NaN, 5, 130, 290 ] } | |
| >>> ''') | |
| >>> import io |
| # Abuse YAML notation to make a heredoc. This will be ignored by the CI. | |
| stages: | |
| - build | |
| - test | |
| - gpgsign | |
| - deploy | |
| ### TEMPLATES ### |
| def dvc_clean(): | |
| import ubelt as ub | |
| import yaml | |
| dvc_root = ub.Path('.').resolve() | |
| subdir = dvc_root / 'MySubdirectory' | |
| ub.Path(ub.cmd('dvc cache dir', cwd=dvc_root, check=True)['out'].strip()) | |
| def list_subdir_dvc_fpaths(subdir): | |
| dvc_fpaths = [] | |
| prog = ub.ProgIter(desc='walking') |
| # import pycodestyle | |
| import ubelt as ub | |
| VERBOSE = 3 | |
| def exec_flake8(dpaths, select=None, ignore=None, max_line_length=79): | |
| if VERBOSE > 1: | |
| print('ignore = {!r}'.format(ignore)) | |
| args_list = ['--max-line-length', f'{max_line_length:d}'] |
| 6190010320010213417794501427120152665560619838379962172624809259242413133662471762706831535160803392841378532145762285099722187617972879813538517518233083306753347260548612898631147332011693627724889751066418150260498985083827355788699194931492232877245108779894625292904047223527216984663982750718623034958051309209333582174176577416459542020450843576169829816732864243397351533895411990483910832697254726363069230161517405040607673995815031529773327058337354301354062371647075206440061759885500802331823316853484399773382885418238331848120196431594456195805808432630836739945562939048610470071338082942418797034098249235121992937130708313446300865889713718389590277346247758021822613993586445131379268791654064089880558143900268099653583909158946828975519742188414010911014106643945712874253761637932854216545020272237775646331705506726197042829531280095737537148740518436741171103518386444450294779963654796639163475149693245937407808337856106834434695344937970746060176768499672131001455338379607532097417780722737895226 |