Download agentic_workflow_data.jsonl from bala5046/agentic-workflow: direct link, hf CLI and curl.
- Browser
- Download file 688 Bytes
-
https://huggingface.co/datasets/bala5046/agentic-workflow/resolve/main/agentic_workflow_data.jsonl
- Command line
-
hf download hf://datasets/bala5046/agentic-workflow/agentic_workflow_data.jsonl
-
curl -L -o agentic_workflow_data.jsonl https://huggingface.co/datasets/bala5046/agentic-workflow/resolve/main/agentic_workflow_data.jsonl
688 Bytes
| {"prompt":"I need to clean a tabular dataset with missing values, encode categorical columns, and train an XGBoost classifier.","tool_call":"run_ml_pipeline(task='classification', steps=['impute_missing', 'label_encode', 'xgboost'])","response":"Here is the execution plan for your classification pipeline.","license":"cc0-1.0","verified_clean":true} | |
| {"prompt":"Optimize a PyTorch training loop where GPU memory utilization is spiking and causing OOM errors.","tool_call":"inspect_gpu_memory(strategy=['gradient_accumulation', 'mixed_precision_fp16'])","response":"To resolve the OOM error, implement gradient accumulation and mixed precision.","license":"cc0-1.0","verified_clean":true} | |