Files
Broken_Code_Generation.1.0/metrics/02_json_validity.json

27 lines
734 B
JSON
Raw Normal View History

{
"metric_group": "json_validity",
"model": "Broken_Code_Generation.1.0",
"hf_model": "Vilyam888/Broken_Code_Generation.1.0",
"adapter_dir": "outputs/qwen25-coder-3b-qlora",
"evaluation_file": "prepared_data/test.json",
"evaluation_date": "2026-06-11",
"samples_evaluated": 100,
"generation_params": {
"temperature": 0.2,
"top_p": 0.95,
"max_new_tokens": 1200,
"seed": 42
},
"metrics": {
"valid_json_rate": 0.94,
"required_fields_rate": 0.92,
"difficulty_match_rate": 0.96,
"topic_tag_key_match_rate": 0.97
},
"metrics_counts": {
"valid_json": 94,
"required_fields_complete": 92,
"difficulty_match": 96,
"topic_tag_keys_match": 97
}
}