reference/generated/tensordict.TensorDictBase.to_json

Run in Google Colab

Colab

Download Notebook

Notebook

View on GitHub

GitHub

TensorDictBase.to_json

TensorDictBase.to_json(path, *, separator: str | None = None, lines: bool = False, **kwargs)

Writes the tensordict to a JSON file, as an array of records.

Each element of the tensordict becomes one record, with one field per leaf, as in to_pandas(). With pandas installed, the file is written with pandas.DataFrame.to_json() (orient="records"), and otherwise with the json module. from_json() reads the file back.

Parameters:

path (str or Path) – the path of the JSON file.

Keyword Arguments:

• separator (str, optional) – if given, joins the keys of nested entries with it to build the column names: with separator=".", the entry ("obs", "x") becomes the column "obs.x". Required when the tensordict has nested tensordicts. Defaults to None.

• lines (bool, optional) – if True, writes JSON Lines, one record per line. Defaults to False.

• **kwargs – passed to pandas.DataFrame.to_json(). Ignored without pandas.

Examples

>>> import os
>>> import tempfile
>>> import torch
>>> from tensordict import TensorDict
>>> td = TensorDict(a=torch.arange(2), b=TensorDict(c=torch.zeros(2)), batch_size=[2])
>>> with tempfile.TemporaryDirectory() as tmpdir:
...     path = os.path.join(tmpdir, "data.json")
...     td.to_json(path, separator=".")
...     with open(path) as f:
...         print(f.read())
[{"a":0,"b.c":0.0},{"a":1,"b.c":0.0}]

Every tensordict class and every tensorclass has this method. For the other methods, see Methods by task.