Tensors

Creating Tensors

Dataset.create_tensor

Creates a new tensor in the dataset.

Dataset.create_group

Creates a tensor group.

Dataset.create_tensor_like

Copies the source tensor's meta information and creates a new tensor with it.

Deleting and Renaming Tensors

Dataset.delete_tensor

Delete a tensor from the dataset.

Dataset.delete_group

Delete a tensor group from the dataset.

Dataset.rename_tensor

Renames tensor with name name to new_name

Dataset.rename_group

Renames group with name name to new_name

Adding and deleting samples

Tensor.append

Appends a single sample to the end of the tensor.

Tensor.extend

Extends the end of the tensor by appending multiple elements from a sequence.

Tensor.pop

Removes an element at the given index.

Tensor.clear

Deletes all samples from the tensor

Tensor.__setitem__

Update samples with new values.

Retrieving samples

Tensor.numpy

Computes the contents of the tensor in numpy format.

Tensor.data

Returns data in the tensor in a format based on the tensor's base htype.

Tensor.tobytes

Returns the bytes of the tensor.

Tensor.text

Return text data.

Tensor.dict

Return json data.

Tensor.list

Return list data.

Tensor._linked_sample

Returns the linked sample at the given index.

Tensor Properties

Tensor.htype

Htype of the tensor.

Tensor.base_htype

Base htype of the tensor.

Tensor.dtype

Dtype of the tensor.

Tensor.shape

Get the shape of this tensor.

Tensor.shape_interval

Returns a ShapeInterval object that describes this tensor's shape more accurately.

Tensor.ndim

Number of dimensions of the tensor.

Tensor.num_samples

Returns the length of the primary axis of the tensor.

Tensor.__len__

Returns the length of the primary axis of the tensor.

Tensor.is_dynamic

Will return True if samples in this tensor have shapes that are unequal.

Tensor.is_sequence

Whether this tensor is a sequence tensor.

Tensor.is_link

Whether this tensor is a link tensor.

Tensor.verify

Whether linked data will be verified when samples are added.

Info

Tensor.info

Returns the information about the tensor.

Tensor.sample_info

Returns info about particular samples in a tensor.

Video features

Tensor.play

Play video sample.

Tensor.timestamps

Returns timestamps (in seconds) for video sample as numpy array.