datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
job_classification_dataset_ruThis dataset represents the classification of a profession by its name and description.
Real resumes from hh.ru with additional anonymization were used to form the dataset.
Claude 3 sonet was used for clasification. The total cost of training the dataset was about 500$
Данный датасет представляет с собой класификацию профессии по её названию и описанию.
Для формирования датасета использовались реальные резюме из hh.ru с дополнительной анонимизацией.
Для класификации был использован claude 3… See the full description on the dataset page: https://huggingface.co/datasets/daswer123/job_classification_dataset_ru.job-classification-llama2
🧠 Job Classification Dataset for LLaMA 2 Fine-Tuning
This dataset contains 1,000 synthetic job descriptions and their associated job categories. It is designed for fine-tuning large language models (LLMs), such as LLaMA 2, for job classification tasks.
📂 Dataset Structure
Format: JSONL (.jsonl)
Fields:
instruction: A generic instruction prompt.
input: The job description text.
output: The job type label.
🔧 Example
{
"instruction": "Classify the… See the full description on the dataset page: https://huggingface.co/datasets/saiteja001r/job-classification-llama2.job_classification_dataset_v2_ruThis dataset represents the classification of a profession by its name and description. Real resumes from hh.ru with additional anonymization were used to form the dataset. Claude 3 sonet was used for clasification. The total cost of training the dataset was about 500$
In this version, formation was done by chunks , and each new chunk was added to the RAG database, which the LLM received later, for more accurate classification.
Данный датасет представляет с собой класификацию профессии по её… See the full description on the dataset page: https://huggingface.co/datasets/daswer123/job_classification_dataset_v2_ru.job_classification_dataset_v2_ruThis dataset represents the classification of a profession by its name and description. Real resumes from hh.ru with additional anonymization were used to form the dataset. Claude 3 sonet was used for clasification. The total cost of training the dataset was about 500$
In this version, formation was done by chunks , and each new chunk was added to the RAG database, which the LLM received later, for more accurate classification.
Данный датасет представляет с собой класификацию профессии по её… See the full description on the dataset page: https://huggingface.co/datasets/art403/job_classification_dataset_v2_ru.