Unsupervised embeddings

Looking for usage documentation? Check out Embeddings.

`TabPFNUnsupervisedModel.get_embeddings`

Get the transformer embeddings for the test data X.

TabPFNUnsupervisedModel.get_embeddings(
    X: torch.tensor,
    per_column: bool = False,
) -> torch.tensor

Parameters

Parameter	Type	Default	Description
`X`	`torch.tensor`	Required	—
`per_column`	`bool`	`False`	—

Returns

Type	Description
`torch.tensor`	torch.Tensor of shape (n_samples, embedding_dim)

`TabPFNUnsupervisedModel.get_embeddings_`

View source

`TabPFNUnsupervisedModel.get_embeddings_per_column`

View source

Alternative implementation for get_embeddings, where we get the embeddings for each column as a label separately and concatenate the results. This alternative way needs more passes but might be more accurate.

TabPFNUnsupervisedModel.get_embeddings_per_column(
    X: torch.tensor,
) -> torch.tensor

Parameters

Parameter	Type	Default	Description
`X`	`torch.tensor`	Required	—

Returns

Type	Description
`torch.tensor`	—

Reference

tabpfn

tabpfn-client

tabpfn-extensions

REST API

Unsupervised embeddings

`TabPFNUnsupervisedModel.get_embeddings`

`TabPFNUnsupervisedModel.get_embeddings_`

`TabPFNUnsupervisedModel.get_embeddings_per_column`

​TabPFNUnsupervisedModel.get_embeddings

​TabPFNUnsupervisedModel.get_embeddings_

​TabPFNUnsupervisedModel.get_embeddings_per_column

`TabPFNUnsupervisedModel.get_embeddings`

`TabPFNUnsupervisedModel.get_embeddings_`

`TabPFNUnsupervisedModel.get_embeddings_per_column`