Batch processing
Maximize efficiency and save your time with the ability to process large volumes of data at once
Visiomera brings you the power of batch processing of data with ChatGPT. With Visiomera, users can prepare prompts and define how dynamically provided data are embedded in them. They can then upload an Excel or CSV file and process each row of data. This article will explain how to batch process your data with Visiomera.
Getting started
You first need to define an automation. You can think of automations as dynamic prompts with user defined outputs in JSON format. Here is an example of a simple automation for analyzing customer reviews:
This automation expects three variables to be provided: username, product and customer_review. Visiomera embeds these variables into the prompt before sending it to ChatGPT. The response from ChatGPT is stored into three variables satisfied, complain_reason and reply.
Batch processing
Imagine having thousands of customer reviews and being tasked to analyze them and also to create a automated reply generator. The automation should generate a reply draft that will be later checked by a human operator and sent to the customer.
We have tabular data with following structure:
Notice that the column names match the variable names in the automation.
Go to the Batch processing page, select your automation from the dropdown list and click on Upload data. You can select your .xlsx or .csv file. A prompt with import options will pop up. After importing, the data will appear in the lower part of the screen.
After clicking on Execute, batch execution results will appear in the data table. New columns will be created with the names satisfied, complain_reason and reply. Note that for this variables to appear here, they have to be selected in the Return step of the automation.
High throughput workflows
Visiomera can help you maximize efficiency and save your time by process large volumes of data at once. It enables the automation of repetitive tasks, reducing the risk of errors and increasing efficiency. Also, it allows for the integration of natural language processing capabilities into data processing workflows, which can enable more sophisticated data analysis and decision-making.