metadata
license: unlicense
datasets:
- vmal/jobs_dataset
language:
- en
base_model:
- google/flan-t5-xl
pipeline_tag: text2text-generation
tags:
- parsing
- jobs
This Model
- It consumes the text information, extracted from HTML job pages and outputs the formatted json which contains job title, company name, salary, location.
- It is fine tuned on base model google/flan-t5-xl
- It supports 2K context length and outputs the json up to 512 tokens.
- It has been fine tuned using huggingface dataset, vmal/jobs_dataset