GPT-3 Info Completion

This component uses OpenAI's GPT-3 engine. You will need an authorized API key to use it.

You can get $18 credit for use the API directly from OpenAI here: https://beta.openai.com/signup

A small token calculator has been included to estimate the cost of making the call. In any case, it is at the user's discretion to take responsibility for the use of the component.

In the input table, we will select the columns to be used (Reference Columns) as data sources and those to be filled in if they have no value.

A useful example will be constructed from a row with all these columns filled with data.

From the columns selected as "Reference Columns", we will choose which ones will be filled if there are any "*" (Columns to Fill).

Options

Columns to fill
Select the columns you want to fill.
Reference columns
Select the columns you want to use as 'base' for the inference of the missed values at the columns to be filled. These columns will be used during examples definition, so ensure there are at least the same rows with all the data of those columns complete as the number of examples you select.
Max. Spending
The maximum amount of money to spend ($).
Max. tokens
Select the maximum number of tokens you expect to be generated by inferring all fields in a row in a single run.
Maximum number of examples at prompt
Select the number of examples you want to generate. Greater values could lead to less structural errors and more specific results, but it also increases the risk of 'overfeating', as well as the cost of executions.
Temperature
Choose a temperature value for text generation. The higher this value is, the more "creativity" or dispersion the model will include in the text generation.
Top P
Top P parameter. For more information, check the OpenAI documentation.
Presence penalty
Positive values penalize the generation of tokens already existing in the generated text.
Frequency penalty
How much to penalize the generation of tokens (i.e. symbolic words/characters) depending on their frequency of occurrence in the generated text.
Model
Select the GPT-3 engine you want to use. The pricing rates are:%%00010- Davinci: $60/Mtoken%%00010- Curie: $6/Mtoken%%00010- Babbage: $1.2/Mtoken%%00010- Ada: $0.8$/Mtoken)
OpenAI API-KEY
The API-key given by OpenAI to the users that have access to GPT-3

Input Ports

Icon
Input data with columns to be inferred using GPT-3

Output Ports

Icon
Output data with the results provided by GPT-3 for selected columns.

Nodes

Extensions

Links

Download

To use this component in KNIME, download it from the below URL and open it in KNIME:

Download Component
On NodePit since: 2022-03-18
Last update: 2024-05-01
Created with KNIME version: v4.5.1

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!