BERT Embedder

Maps String or Document columns to a numerical vector using the provided BERT model. The node accepts non-fine-tuned BERT models (magenta port) or fine-tuned BERT models (grey port) and utilizes them for calculation of the embeddings of the texts. Embeddings are the numerical vector representation of the texts that can be used for visualization, clustering, classification, etc.

Options

Settings

Sentence column

The column with texts that will be vectorized.

Two-sentence mode

The mode for cases when input text consists of 2 distinct parts ("sentences").

Second sentence column

The column with the second sentence for the Two-sentence mode.

Max sequence length

The maximum length of a sequence after tokenization, limit is 512.

Advanced

Batch size

The size of a chunk of the input data to process.

Include sequence embeddings

Include individual word embeddings in addition to the whole text embeddings.

Python

Python

Select one of the Python execution environment options:

• use default Python environment for the Redfield BERT Nodes (can be configured on the preference page)
• use Conda environment from a Conda flow variable (only selectable if such a flow variable is available)

Views

This node has no views

Workflows

• CI_IA_AP-Duplicate_Invoice_DetectionKNIME Hub
• CI_IA_Bank reconciliation - Outliers detectionKNIME Hub
• Content_Based_Movie_Recommenders_System_BERT_EmbeddingsKNIME Hub
• JKISeason2-6KNIME Hub
• linkedin regression treeKNIME Forum
• Show all 7 workflows

Links

• No links available

Developers