regulated-advice / eval /regulated_advice_eval.json
bogdanraduta's picture
Upload eval/regulated_advice_eval.json with huggingface_hub
c0812d6 verified
Raw
History Blame Contribute Delete
20.6 kB
{
"task": "regulated_advice",
"threshold": 0.88,
"per_language": {
"az": {
"support": 22,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 52
},
"bg": {
"support": 22,
"precision": 1.0,
"recall": 0.9545,
"f1": 0.9767,
"fpr": 0.0,
"accuracy": 0.9615,
"n": 52
},
"cs": {
"support": 22,
"precision": 1.0,
"recall": 0.9545,
"f1": 0.9767,
"fpr": 0.0,
"accuracy": 0.9615,
"n": 52
},
"da": {
"support": 22,
"precision": 1.0,
"recall": 0.9545,
"f1": 0.9767,
"fpr": 0.0,
"accuracy": 0.9423,
"n": 52
},
"de": {
"support": 22,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 0.9808,
"n": 52
},
"el": {
"support": 22,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 0.9808,
"n": 52
},
"en": {
"support": 22,
"precision": 1.0,
"recall": 0.9545,
"f1": 0.9767,
"fpr": 0.0,
"accuracy": 0.9423,
"n": 52
},
"es": {
"support": 22,
"precision": 1.0,
"recall": 0.9545,
"f1": 0.9767,
"fpr": 0.0,
"accuracy": 0.9808,
"n": 52
},
"et": {
"support": 22,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 0.9615,
"n": 52
},
"fi": {
"support": 22,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 0.9423,
"n": 52
},
"fr": {
"support": 22,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 0.9231,
"n": 52
},
"ga": {
"support": 22,
"precision": 0.913,
"recall": 0.9545,
"f1": 0.9333,
"fpr": 0.0667,
"accuracy": 0.8269,
"n": 52
},
"hr": {
"support": 22,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 0.9231,
"n": 52
},
"hu": {
"support": 22,
"precision": 1.0,
"recall": 0.9545,
"f1": 0.9767,
"fpr": 0.0,
"accuracy": 0.9038,
"n": 52
},
"it": {
"support": 22,
"precision": 1.0,
"recall": 0.9545,
"f1": 0.9767,
"fpr": 0.0,
"accuracy": 0.9231,
"n": 52
},
"lt": {
"support": 22,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 0.9808,
"n": 52
},
"lv": {
"support": 22,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 0.9808,
"n": 52
},
"mt": {
"support": 22,
"precision": 1.0,
"recall": 0.8182,
"f1": 0.9,
"fpr": 0.0,
"accuracy": 0.8269,
"n": 52
},
"nl": {
"support": 22,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 52
},
"pl": {
"support": 22,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 0.9038,
"n": 52
},
"pt": {
"support": 22,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 0.9808,
"n": 52
},
"ro": {
"support": 22,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 0.9808,
"n": 52
},
"sk": {
"support": 22,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 0.9808,
"n": 52
},
"sl": {
"support": 22,
"precision": 1.0,
"recall": 0.9545,
"f1": 0.9767,
"fpr": 0.0,
"accuracy": 0.9231,
"n": 52
},
"sv": {
"support": 22,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 0.9808,
"n": 52
},
"tr": {
"support": 22,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 0.9615,
"n": 52
}
},
"per_register": {
"definition": {
"support": 0,
"precision": 0.0,
"recall": 0.0,
"f1": 0.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 123
},
"directive": {
"support": 133,
"precision": 1.0,
"recall": 0.9699,
"f1": 0.9847,
"fpr": 0.0,
"accuracy": 0.8947,
"n": 133
},
"general_risk": {
"support": 0,
"precision": 0.0,
"recall": 0.0,
"f1": 0.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 111
},
"historical_fact": {
"support": 0,
"precision": 0.0,
"recall": 0.0,
"f1": 0.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 107
},
"hypothetical": {
"support": 0,
"precision": 0.0,
"recall": 0.0,
"f1": 0.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 98
},
"mundane_account_access": {
"support": 0,
"precision": 0.0,
"recall": 0.0,
"f1": 0.0,
"fpr": 0.0179,
"accuracy": 0.9821,
"n": 56
},
"mundane_informational": {
"support": 0,
"precision": 0.0,
"recall": 0.0,
"f1": 0.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 48
},
"mundane_operational": {
"support": 0,
"precision": 0.0,
"recall": 0.0,
"f1": 0.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 50
},
"mundane_transactional": {
"support": 0,
"precision": 0.0,
"recall": 0.0,
"f1": 0.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 54
},
"personal_recommendation": {
"support": 328,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 0.9695,
"n": 328
},
"process_description": {
"support": 0,
"precision": 0.0,
"recall": 0.0,
"f1": 0.0,
"fpr": 0.0075,
"accuracy": 0.9925,
"n": 133
},
"suitability_claim": {
"support": 111,
"precision": 1.0,
"recall": 0.9189,
"f1": 0.9577,
"fpr": 0.0,
"accuracy": 0.6036,
"n": 111
}
},
"per_label": {
"financial_advice": {
"support": 260,
"precision": 0.8745,
"recall": 0.9115,
"f1": 0.8927,
"fpr": 0.0311,
"accuracy": 0.0,
"n": 0
},
"legal_advice": {
"support": 156,
"precision": 0.8933,
"recall": 0.859,
"f1": 0.8758,
"fpr": 0.0134,
"accuracy": 0.0,
"n": 0
},
"medical_advice": {
"support": 156,
"precision": 0.9504,
"recall": 0.859,
"f1": 0.9024,
"fpr": 0.0059,
"accuracy": 0.0,
"n": 0
}
},
"worst_languages": [
[
"mt",
0.9
],
[
"ga",
0.9333333333333332
],
[
"bg",
0.9767441860465117
],
[
"cs",
0.9767441860465117
],
[
"da",
0.9767441860465117
]
],
"calibration": {
"threshold": 0.88,
"objective": "macro_f1",
"value_on_val": 0.8921,
"default_threshold": 0.5,
"default_value_on_val": 0.8606,
"improvement": 0.0315,
"chosen_on": "validation split, never test",
"curve": [
{
"threshold": 0.02,
"macro_f1": 0.4851,
"any_label_recall": 1.0,
"any_label_fpr": 0.0577
},
{
"threshold": 0.03,
"macro_f1": 0.5211,
"any_label_recall": 1.0,
"any_label_fpr": 0.0385
},
{
"threshold": 0.04,
"macro_f1": 0.5858,
"any_label_recall": 1.0,
"any_label_fpr": 0.0295
},
{
"threshold": 0.05,
"macro_f1": 0.6416,
"any_label_recall": 1.0,
"any_label_fpr": 0.0269
},
{
"threshold": 0.06,
"macro_f1": 0.6811,
"any_label_recall": 1.0,
"any_label_fpr": 0.0244
},
{
"threshold": 0.07,
"macro_f1": 0.7085,
"any_label_recall": 1.0,
"any_label_fpr": 0.0192
},
{
"threshold": 0.08,
"macro_f1": 0.7321,
"any_label_recall": 1.0,
"any_label_fpr": 0.0192
},
{
"threshold": 0.09,
"macro_f1": 0.7437,
"any_label_recall": 1.0,
"any_label_fpr": 0.0192
},
{
"threshold": 0.1,
"macro_f1": 0.7512,
"any_label_recall": 1.0,
"any_label_fpr": 0.0192
},
{
"threshold": 0.11,
"macro_f1": 0.764,
"any_label_recall": 1.0,
"any_label_fpr": 0.0192
},
{
"threshold": 0.12,
"macro_f1": 0.7664,
"any_label_recall": 1.0,
"any_label_fpr": 0.0192
},
{
"threshold": 0.13,
"macro_f1": 0.7717,
"any_label_recall": 1.0,
"any_label_fpr": 0.0192
},
{
"threshold": 0.14,
"macro_f1": 0.774,
"any_label_recall": 1.0,
"any_label_fpr": 0.0192
},
{
"threshold": 0.15,
"macro_f1": 0.7775,
"any_label_recall": 1.0,
"any_label_fpr": 0.0192
},
{
"threshold": 0.16,
"macro_f1": 0.7822,
"any_label_recall": 1.0,
"any_label_fpr": 0.0192
},
{
"threshold": 0.17,
"macro_f1": 0.7838,
"any_label_recall": 1.0,
"any_label_fpr": 0.0192
},
{
"threshold": 0.18,
"macro_f1": 0.7844,
"any_label_recall": 1.0,
"any_label_fpr": 0.0192
},
{
"threshold": 0.19,
"macro_f1": 0.7867,
"any_label_recall": 0.9983,
"any_label_fpr": 0.0192
},
{
"threshold": 0.2,
"macro_f1": 0.7905,
"any_label_recall": 0.9983,
"any_label_fpr": 0.0192
},
{
"threshold": 0.21,
"macro_f1": 0.795,
"any_label_recall": 0.9983,
"any_label_fpr": 0.0167
},
{
"threshold": 0.22,
"macro_f1": 0.7968,
"any_label_recall": 0.9983,
"any_label_fpr": 0.0167
},
{
"threshold": 0.23,
"macro_f1": 0.7978,
"any_label_recall": 0.9983,
"any_label_fpr": 0.0167
},
{
"threshold": 0.24,
"macro_f1": 0.8015,
"any_label_recall": 0.9965,
"any_label_fpr": 0.0167
},
{
"threshold": 0.25,
"macro_f1": 0.8038,
"any_label_recall": 0.9965,
"any_label_fpr": 0.0167
},
{
"threshold": 0.26,
"macro_f1": 0.8078,
"any_label_recall": 0.9965,
"any_label_fpr": 0.0167
},
{
"threshold": 0.27,
"macro_f1": 0.8091,
"any_label_recall": 0.9965,
"any_label_fpr": 0.0167
},
{
"threshold": 0.28,
"macro_f1": 0.8135,
"any_label_recall": 0.9965,
"any_label_fpr": 0.0154
},
{
"threshold": 0.29,
"macro_f1": 0.8156,
"any_label_recall": 0.9965,
"any_label_fpr": 0.0154
},
{
"threshold": 0.3,
"macro_f1": 0.816,
"any_label_recall": 0.9965,
"any_label_fpr": 0.0154
},
{
"threshold": 0.31,
"macro_f1": 0.8186,
"any_label_recall": 0.9965,
"any_label_fpr": 0.0154
},
{
"threshold": 0.32,
"macro_f1": 0.8246,
"any_label_recall": 0.9965,
"any_label_fpr": 0.0154
},
{
"threshold": 0.33,
"macro_f1": 0.8259,
"any_label_recall": 0.9965,
"any_label_fpr": 0.0154
},
{
"threshold": 0.34,
"macro_f1": 0.8273,
"any_label_recall": 0.9965,
"any_label_fpr": 0.0154
},
{
"threshold": 0.35,
"macro_f1": 0.8274,
"any_label_recall": 0.9965,
"any_label_fpr": 0.0141
},
{
"threshold": 0.36,
"macro_f1": 0.8293,
"any_label_recall": 0.9965,
"any_label_fpr": 0.0141
},
{
"threshold": 0.37,
"macro_f1": 0.8334,
"any_label_recall": 0.9965,
"any_label_fpr": 0.0141
},
{
"threshold": 0.38,
"macro_f1": 0.8356,
"any_label_recall": 0.9965,
"any_label_fpr": 0.0141
},
{
"threshold": 0.39,
"macro_f1": 0.837,
"any_label_recall": 0.9965,
"any_label_fpr": 0.0141
},
{
"threshold": 0.4,
"macro_f1": 0.8385,
"any_label_recall": 0.9965,
"any_label_fpr": 0.0141
},
{
"threshold": 0.41,
"macro_f1": 0.8433,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0128
},
{
"threshold": 0.42,
"macro_f1": 0.8429,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0128
},
{
"threshold": 0.43,
"macro_f1": 0.845,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0128
},
{
"threshold": 0.44,
"macro_f1": 0.8458,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0128
},
{
"threshold": 0.45,
"macro_f1": 0.8458,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0128
},
{
"threshold": 0.46,
"macro_f1": 0.8474,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0128
},
{
"threshold": 0.47,
"macro_f1": 0.8512,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0115
},
{
"threshold": 0.48,
"macro_f1": 0.8528,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0115
},
{
"threshold": 0.49,
"macro_f1": 0.8558,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0115
},
{
"threshold": 0.5,
"macro_f1": 0.8606,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0115
},
{
"threshold": 0.51,
"macro_f1": 0.8604,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0115
},
{
"threshold": 0.52,
"macro_f1": 0.8627,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0115
},
{
"threshold": 0.53,
"macro_f1": 0.8641,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0103
},
{
"threshold": 0.54,
"macro_f1": 0.8641,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0103
},
{
"threshold": 0.55,
"macro_f1": 0.8635,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0103
},
{
"threshold": 0.56,
"macro_f1": 0.8643,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0103
},
{
"threshold": 0.57,
"macro_f1": 0.8661,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0103
},
{
"threshold": 0.58,
"macro_f1": 0.8661,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0103
},
{
"threshold": 0.59,
"macro_f1": 0.8683,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0103
},
{
"threshold": 0.6,
"macro_f1": 0.8683,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0103
},
{
"threshold": 0.61,
"macro_f1": 0.8699,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0103
},
{
"threshold": 0.62,
"macro_f1": 0.8707,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0103
},
{
"threshold": 0.63,
"macro_f1": 0.8686,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0103
},
{
"threshold": 0.64,
"macro_f1": 0.8717,
"any_label_recall": 0.9948,
"any_label_fpr": 0.009
},
{
"threshold": 0.65,
"macro_f1": 0.8715,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0077
},
{
"threshold": 0.66,
"macro_f1": 0.8748,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0077
},
{
"threshold": 0.67,
"macro_f1": 0.8782,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0077
},
{
"threshold": 0.68,
"macro_f1": 0.8782,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0077
},
{
"threshold": 0.69,
"macro_f1": 0.877,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0077
},
{
"threshold": 0.7,
"macro_f1": 0.8779,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0077
},
{
"threshold": 0.71,
"macro_f1": 0.8779,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0077
},
{
"threshold": 0.72,
"macro_f1": 0.8833,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0077
},
{
"threshold": 0.73,
"macro_f1": 0.8839,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0077
},
{
"threshold": 0.74,
"macro_f1": 0.8882,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0077
},
{
"threshold": 0.75,
"macro_f1": 0.8899,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0077
},
{
"threshold": 0.76,
"macro_f1": 0.8896,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0077
},
{
"threshold": 0.77,
"macro_f1": 0.8888,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0077
},
{
"threshold": 0.78,
"macro_f1": 0.8876,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0077
},
{
"threshold": 0.79,
"macro_f1": 0.8885,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0077
},
{
"threshold": 0.8,
"macro_f1": 0.885,
"any_label_recall": 0.9948,
"any_label_fpr": 0.0077
},
{
"threshold": 0.81,
"macro_f1": 0.885,
"any_label_recall": 0.993,
"any_label_fpr": 0.0077
},
{
"threshold": 0.82,
"macro_f1": 0.8847,
"any_label_recall": 0.993,
"any_label_fpr": 0.0077
},
{
"threshold": 0.83,
"macro_f1": 0.8837,
"any_label_recall": 0.9913,
"any_label_fpr": 0.0077
},
{
"threshold": 0.84,
"macro_f1": 0.8893,
"any_label_recall": 0.9913,
"any_label_fpr": 0.0064
},
{
"threshold": 0.85,
"macro_f1": 0.8886,
"any_label_recall": 0.9913,
"any_label_fpr": 0.0064
},
{
"threshold": 0.86,
"macro_f1": 0.89,
"any_label_recall": 0.9843,
"any_label_fpr": 0.0064
},
{
"threshold": 0.87,
"macro_f1": 0.8912,
"any_label_recall": 0.9825,
"any_label_fpr": 0.0064
},
{
"threshold": 0.88,
"macro_f1": 0.8921,
"any_label_recall": 0.9738,
"any_label_fpr": 0.0064
},
{
"threshold": 0.89,
"macro_f1": 0.89,
"any_label_recall": 0.9685,
"any_label_fpr": 0.0064
},
{
"threshold": 0.9,
"macro_f1": 0.8904,
"any_label_recall": 0.965,
"any_label_fpr": 0.0064
},
{
"threshold": 0.91,
"macro_f1": 0.8902,
"any_label_recall": 0.9458,
"any_label_fpr": 0.0064
},
{
"threshold": 0.92,
"macro_f1": 0.8882,
"any_label_recall": 0.9266,
"any_label_fpr": 0.0064
},
{
"threshold": 0.93,
"macro_f1": 0.8887,
"any_label_recall": 0.9003,
"any_label_fpr": 0.0064
},
{
"threshold": 0.94,
"macro_f1": 0.8878,
"any_label_recall": 0.8881,
"any_label_fpr": 0.0064
},
{
"threshold": 0.95,
"macro_f1": 0.8808,
"any_label_recall": 0.8601,
"any_label_fpr": 0.0064
}
],
"notes": []
},
"pair_accuracy": 0.8794
}