Standalone function for uploading a checkpoint file. This function does not actually upload the checkpoint; it initiates the RemoteUploader's uploading of it Args: source_path (str): The path to the file to upload. dest_dir (str): The directory/uri to upload the file to.
(
dest_dir: str,
source_path: Optional[str]=None,
symlink_granularity: Optional[str]=None, # file, dir, or None
symlink_name: Optional[str]='latest.symlink',
async_upload: bool = True,
state: Optional[State] = None,
overwrite: bool = False,
)
| 134 | |
| 135 | |
| 136 | def upload_file( |
| 137 | dest_dir: str, |
| 138 | source_path: Optional[str]=None, |
| 139 | symlink_granularity: Optional[str]=None, # file, dir, or None |
| 140 | symlink_name: Optional[str]='latest.symlink', |
| 141 | async_upload: bool = True, |
| 142 | state: Optional[State] = None, |
| 143 | overwrite: bool = False, |
| 144 | ): |
| 145 | """Standalone function for uploading a checkpoint file. |
| 146 | |
| 147 | This function does not actually upload the checkpoint; it initiates the RemoteUploader's uploading of it |
| 148 | Args: |
| 149 | source_path (str): The path to the file to upload. |
| 150 | dest_dir (str): The directory/uri to upload the file to. |
| 151 | symlink_granularity (Optional[str]): The granularity to use for symlinking. One of 'file', 'dir', or None. |
| 152 | if None: no symlink uploaded |
| 153 | if 'file': command remoteuploader to wait until the file (specificied by source_path) is uploaded and then uploads a symlink pointing to the uploaded file |
| 154 | if 'dir': command remoteuploader to wait until all files across all ranks are uploaded to dest_dir and then uploads a symlink |
| 155 | pointing to the remote directory (prefix in object_store terminology). |
| 156 | symlink_name (Optional[str]): The name to use for the symlink. Defaults to 'latest.symlink'. |
| 157 | async_upload (bool): If True, the uploads will be done asynchronously via the RemoteUploader and this function will return immediately. |
| 158 | state (Optional[State]): If async_upload is True, then state must be specified so that the remote_uploader can be |
| 159 | either extracted from state.callbacks or initialized and added to state.callbacks. |
| 160 | overwrite (bool): If allow overwrite existing remote checkpoint files |
| 161 | """ |
| 162 | remote_uploader = RemoteUploader(remote_folder=dest_dir) |
| 163 | dest_path = '' |
| 164 | if source_path is not None: |
| 165 | _, _, dest_path = parse_uri(dest_dir) |
| 166 | remote_file_name = os.path.join(dest_path, os.path.basename(source_path)) |
| 167 | else: |
| 168 | remote_file_name = None |
| 169 | all_remote_file_names = dist.all_gather_object([remote_file_name] if remote_file_name is not None else []) |
| 170 | remote_file_names = [] |
| 171 | for filenames in all_remote_file_names: |
| 172 | remote_file_names += filenames |
| 173 | |
| 174 | if source_path is not None: |
| 175 | assert remote_uploader is not None |
| 176 | assert remote_file_name is not None |
| 177 | remote_uploader.upload_file_async( |
| 178 | remote_file_name=remote_file_name, |
| 179 | file_path=pathlib.Path(source_path), |
| 180 | overwrite=overwrite, |
| 181 | ) |
| 182 | symlink_remote_file_name = None |
| 183 | if dist.get_global_rank() == 0: |
| 184 | if symlink_name is not None: |
| 185 | symlink_remote_file_name = os.path.join(dest_path, symlink_name) |
| 186 | if symlink_granularity == 'file': |
| 187 | create_symlink_file(dest_path, symlink_name) |
| 188 | elif symlink_granularity == 'dir': |
| 189 | create_symlink_file(str(pathlib.Path(dest_path).parent), symlink_name) |
| 190 | elif symlink_granularity is not None: |
| 191 | raise ValueError(f'Unrecognized symlink granularity: {symlink_granularity}') |
| 192 | else: |
| 193 | symlink_remote_file_name = None |