BiGRU_T_version / reports /v6_5_v2_predict_fix_eval.json
PowerMachine's picture
V7 FINAL: FASE1+FASE2 completed (10000 samples, 20 punishments)
ce4887a verified
Raw History Blame Contribute Delete
3.98 kB
{
"evaluation": "predict_fix_v65_v2",
"user_requirement": "os strings 'gato' e 'cachorro' são fixos quando deveriam ser extrações variáveis e flexíveis de rótulos proveniente de dados dos datasets anteriormente treinados",
"n_test_queries": 6,
"results": [
{
"query": "o gato dorme na cama",
"prediction": "short_text",
"probability": 0.4133017063140869,
"is_gato_hardcoded": false,
"is_cachorro_hardcoded": false,
"is_registry_label": true,
"is_default_label": false
},
{
"query": "calcule dois mais dois",
"prediction": "long_text",
"probability": 0.5006099939346313,
"is_gato_hardcoded": false,
"is_cachorro_hardcoded": false,
"is_registry_label": true,
"is_default_label": false
},
{
"query": "qual é a capital do brasil",
"prediction": "long_text",
"probability": 0.5814295411109924,
"is_gato_hardcoded": false,
"is_cachorro_hardcoded": false,
"is_registry_label": true,
"is_default_label": false
},
{
"query": "explique o que é uma rede neural",
"prediction": "long_text",
"probability": 0.5864738821983337,
"is_gato_hardcoded": false,
"is_cachorro_hardcoded": false,
"is_registry_label": true,
"is_default_label": false
},
{
"query": "olá como você está",
"prediction": "short_text",
"probability": 0.37604132294654846,
"is_gato_hardcoded": false,
"is_cachorro_hardcoded": false,
"is_registry_label": true,
"is_default_label": false
},
{
"query": "traduza hello para portugues",
"prediction": "long_text",
"probability": 0.5700568556785583,
"is_gato_hardcoded": false,
"is_cachorro_hardcoded": false,
"is_registry_label": true,
"is_default_label": false
}
],
"per_dataset_results": [
{
"dataset": "dominguesm/restore-punctuation-ptbr-dataset",
"expected_labels": [
"unpunctuated",
"punctuated"
],
"prediction": "punctuated",
"valid_for_dataset": true
},
{
"dataset": "carolina-c4ai/corpus-carolina",
"expected_labels": [
"raw_corpus",
"normalized_text"
],
"prediction": "normalized_text",
"valid_for_dataset": true
},
{
"dataset": "CEIA-POSITIVO/ultrachat_br_clustred_balanced_v1",
"expected_labels": [
"user_turn",
"assistant_turn"
],
"prediction": "assistant_turn",
"valid_for_dataset": true
},
{
"dataset": "dominguesm/Canarim-Instruct-PTBR-Dataset",
"expected_labels": [
"instruction",
"response"
],
"prediction": "response",
"valid_for_dataset": true
},
{
"dataset": "adalbertojunior/punctuation-ptbr",
"expected_labels": [
"unpunctuated",
"punctuated"
],
"prediction": "punctuated",
"valid_for_dataset": true
},
{
"dataset": "iara-project/news-articles-ptbr-dataset",
"expected_labels": [
"headline",
"body"
],
"prediction": "body",
"valid_for_dataset": true
},
{
"dataset": "manoela/noticias_ptbr",
"expected_labels": [
"headline",
"body"
],
"prediction": "body",
"valid_for_dataset": true
},
{
"dataset": "BrunoN-Dev/corpus-ptbr-v1",
"expected_labels": [
"short_text",
"long_text"
],
"prediction": "long_text",
"valid_for_dataset": true
}
],
"summary": {
"n_returns_gato": 0,
"n_returns_cachorro": 0,
"n_returns_registry_label": 6,
"n_returns_default_label": 0,
"hardcoded_bug_present": false,
"predict_fix_verified": true
},
"quality_assessment": {
"fix_status": "PASS",
"note": "predict() não retorna mais 'gato'/'cachorro' fixos — usa label_registry dinâmico."
}
}