Data Normalization

Data processing and normalization tasks evaluate structured output from messy prose and different structures. This capability is essential for catalogue and product pipelines where data needs to be extracted from unstructured text and formatted consistently. Models must understand the desired output format and extract relevant information accurately.

100
Input Tokens
0
Output Tokens
0
Duration
0
Cost
0
Context
Input
Reformat the following context text where there is only 1 sentence per line.
Expected output
{"instructions": "Plug in the vacuum cleaner and switch it on.
Adjust the height setting for your floor type.
Move the cleaner slowly back and forth across the surface.
Empty the dust container when full.
After use, switch off, unplug, and store in a dry place."}
Model output
{
  "instructions": "Plug in the vacuum cleaner and switch it on.\nAdjust the height setting for your floor type.\nMove the cleaner slowly back and forth across the surface.\nEmpty the dust container when full.\nAfter use, switch off, unplug, and store in a dry place."
}