Read text from a given ``filepath`` with 'r' mode. Args: filepath (str or Path): Path to read data. encoding (str): T

Infer a suitable file client based on the URI and arguments. Args: file_client_args (dict, optional): Arguments to instantiate a

r"""Concatenate all file paths. Join one or more filepath components intelligently. The return value is the concatenation of filepath and any

r"""Concatenate all file paths. Join one or more filepath components intelligently. The return value is the concatenation of filepath

Return a new data element with same type. If ``metainfo`` and ``data`` are None, the new data element will have same metainfo and data

Returns a set of submodules that were never called during the trace of the graph. This may be because they were unused, or because th

Build training loop. Examples of ``loop``:: # `EpochBasedTrainLoop` will be used loop = dict(by_epoch=True, max_epoc

Close an opened object.

The in-place version of ``get_subset`` to convert dataset to a subset of original dataset. This method will convert the original data

Initialize gradient accumulation related attributes. ``OptimWrapper`` can be used without calling ``initialize_iter_status``. However

Scan a directory to find the interested directories or files in arbitrary order. Note: :meth:`list_dir_or_file` returns t

Initialize the distributed environment.

Apply function to each element in dict, list or tuple that matches with the expression. For examples, if you want to convert each element in

Build function of OptimWrapper. If ``constructor`` is set in the ``cfg``, this method will build an optimizer wrapper constructor, and use op

Check whether the type of value is in ``valid_type``. Args: name (str): value name. value (Any): value. valid_type (Type,

Draw single or multiple circles. Args: center (Union[np.ndarray, torch.Tensor]): The x coordinate of each line' s

Dump data to json/yaml/pickle strings or files. This method provides a unified api for dumping data as strings or to files, and also supports

Load annotation file and set ``BaseDataset._fully_initialized`` to True. If ``lazy_init=False``, ``full_init`` will be called during

Import modules from the given list of strings. Args: imports (list | str | None): The given module names to be imported. allow_fa

Check whether a file path is a directory. Args: filepath (str or Path): Path to be checked whether it is a direct

A generator to get the name and corresponding :obj:`OptimWrapper`

Register a backend to FileClient. This method can be used as a normal class method or a decorator. .. code-block:: python

Sets additional operator handles, or replaces existing ones. If a handle is ``None``, the op will be explicitly ignored. Otherwise, h

Temporarily switch default scope to the target scope, and get the corresponding registry. If the registry of the corresponding scope

Initialize the distributed environment.

Initialize the distributed environment.

Recursively copy an entire directory tree rooted at src to a directory named dst and return the destination directory. src and dst sh

Return a subset of dataset. This method will return a subset of original dataset. If type of indices is int, ``get_subset_`` will ret

Check whether a file path is a directory. Args: filepath (str or Path): Path to be checked whether it is a direct

Convert all tensors to np.ndarray in data.

Write text to a given ``filepath`` with 'w' mode. Note: ``put_text`` will create a directory if the directory of ``fi

Resume model from checkpoint. Args: filename (str): Accept local filepath, URL, ``torchvision://xxx``, ``open-mml

Set or change key-value pairs in ``metainfo_field`` by parameter ``metainfo``. Args: metainfo (dict): A dict contains the

Sets which warnings to print when tracing the graph to calculate statistics. There are three modes. Defaults to 'no_tracer_warning'. A

Update progressbar. Args: num_tasks (int): Update step size.

Update parameters in :attr:`optimizer`.

Initialize the distributed environment.

Used to parse the parameter 'on_trace_ready'.

Build dataloader. The method builds three components: - Dataset - Sampler - Dataloader An example of ``data

Returns the statistics for all submodules, aggregated over all operators. Returns: Counter(str): statistics counter group

Call all hooks. Args: fn_name (str): The function name in each hook to be called, such as "before_train_epoch".

Create a symbolic link pointing to src named dst. If failed to create a symbolic link pointing to src, directly copy src to dst inste

Draw single or multiple binary masks. Args: binary_masks (np.ndarray, torch.Tensor): The binary_masks to draw wit

Return the last computed value by current scheduler. Returns: list: A list of the last computed value of the optimizer's

Overwrite the current default scope with `scope_name`

Count parameters of a model and its submodules. Adopted from https://github.com/facebookresearch/fvcore/blob/main/fvcore/nn/parameter_count.p

Register a hook into the hook list. The hook will be inserted into a priority queue, with the specified priority (See :class:`Priorit

Launch validation.

If the type of value is torch.Tensor, convert the value to np.ndarray. Args: value (np.ndarray, torch.Tensor): value. Returns:

Load checkpoint from somewhere (modelzoo, file, url). Args: filename (str): Accept local filepath, URL, ``torchvision://xxx``,

Decide whether to stop the training process. Args: runner (Runner): The runner of the training process. metrics (dict

Build test loop. Examples of ``loop``:: # `TestLoop` will be used loop = dict() # custom test loop

Returns the canonical module name of the given ``name``, which might be different from the given ``name`` if the module is shared. Th

Check whether the type of value is in ``valid_type``. If type of the ``value`` is list, check whether its length is equal with or greater than

Draw single or multiple bboxes. Args: bboxes (Union[np.ndarray, torch.Tensor]): The bboxes to draw with the forma

Draw single or multiple line segments. Args: x_datas (Union[np.ndarray, torch.Tensor]): The x coordinate of each

Draw single or multiple bboxes. Args: polygons (Union[Union[np.ndarray, torch.Tensor],\ List[Union[np.ndarray, to

Read bytes from a given ``filepath`` with 'rb' mode. Args: filepath (str or Path): Path to read data. Returns:

Get the metric value specified by an indicator, which can be either a metric name or a full name with evaluator prefix. Args: indicat

Returns True if Ascend PyTorch and npu devices exist.

r"""Concatenate all file paths. Join one or more filepath components intelligently. The return value is the concatenation of filepath

Load or resume checkpoint.

Parse the prefix of a uri. Args: uri (str | Path): Uri to be parsed that contains the file prefix. Examples:

Pop property in data and metainfo as the same as python.

Track the progress of parallel task execution with a progress bar. The built-in :mod:`multiprocessing` module is used for process pools and tasks

Sets if warnings from uncalled submodules are shown. Defaults to true. A submodule is considered "uncalled" if it is never called dur

Update the `_params_init_info` in the module if the value of parameters are changed. Args: module (obj:`nn.Module`): The module of Py

Update the parameters of the model. This method will execute the ``avg_func`` to compute the new parameters and update the model's par

If the type of ``value`` is ``valid_type``, convert the value to list and expand to ``expand_dim``. Args: value (Any): value.

Get device id of current machine.

Get package name and relative config path. Args: cfg_path (str): External relative config path with 'package::'. Returns: Tu

Return the root registry.

Record logs after training iteration. Args: runner (Runner): The runner of the training process. batch_idx (int): The