Skip to content

Upload data

Upload Document

Uploading a document works in two steps. When you click the Upload Document button, you are presented with a file selector to select a local document for upload. Some supported files are .docx, .doc, .pdf, .txt, .csv, .json, and .xlsx.

mAIstro will automatically run OCR processing if a PDF is uploaded, but returns no text contents.

After the document is successfully uploaded, it is available in the Upload Document pane:

upload_document

The uploaded document can then be used with the following syntax:

{{ doc|name:output.csv }}

Parameters

  • File Upload: The file to be processed.

Returns

  • The plain text of the document. If an image-based PDF is uploaded, returning no text from the scraper, we will automatically return OCR'd text from the document.

OCR an Image

mAIstro’s OCR feature automatically processes image-based PDFs and images, converting them into searchable, editable text.

OCR'ing a document works in two steps. When you click the OCR an Image button, you are presented with a file selector to select a local document for upload. Some supported files are .pdf, .png, .jpeg.

After the document is successfully uploaded, it is available in the Upload Document pane:

ocr_screenshot

The uploaded document or image can then be used with the following syntax:

{{ doc|name:screenshot_2024-11-05.png }}

Parameters

  • File Upload: PDF or image file to be processed with OCR.

Returns

  • The plain text of the document. If an image-based PDF is uploaded, returning no text from the scraper, we will automatically return OCR'd text from the document.
Example 1: Using OCR with PDF Files

  1. Go to Upload Data in mAIstro.

  2. Select Upload Document or OCR an Image and choose your PDF file.

  3. OCR will be automatically applied, transforming the document into searchable text.

NTL Snippet: 

{{ doc | name: "example.pdf" }}
{{ LLM | prompt: "List names in this document:" | cache: "true" }}

Returns

This example returns a text-rich version of example.pdf, with names extracted as specified in the prompt.

Example 2: Using OCR with Image Files

  1. Go to Upload Data in mAIstro.

  2. Select OCR an Image to upload an image file.

  3. OCR processing starts automatically, converting the image into searchable text.

NTL Snippet: 

{{ doc  | name: "image.png" }}
{{ LLM | prompt: "List names in this document:" | cache: "true" }}

Returns

This example returns a text-rich version of image.png, optimized for data extraction and analysis.