// Copyright 2023 Ant Group Co., Ltd.
// 
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
// 
//   http://www.apache.org/licenses/LICENSE-2.0
// 
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
//

syntax = "proto3";

package secretflowapis.v1.sdc.teeapps.params;

option java_package = "com.alipay.secretflow.secretflowapis.v1.sdc.teeapps.params";
option java_outer_classname = "BiClassifierEvalProto";

message BiClassifierEvalParams {
  string score_field_name = 1;
  int32 bad_label_value = 2;
  uint32 bucket_size = 3;
}

message BiClassifierEvalReport {
  message PrReport {
    // 范围: [0,1] 越小越好
    double fpr = 1;
    // 范围: [0,1] 越大越好
    double precision = 2;
    // 范围: [0,1] 越大越好
    double recall = 3;
  }
  message SummaryReport {
    int32 total_samples = 1;
    int32 positive_samples = 2;
    int32 negative_samples = 3;
    // 范围: [0,1] 0.5近似随机分布，越接近1越好
    double auc = 4;
    // 范围: [0,1] 用于区分正负样本分隔程度，越大越好
    double ks = 5;
    // 范围: [0,1] 是precision和recall的一个调和参数，越大越好
    double f1_score = 6;
  }
  message EquiBinReport {
    // score分桶，[start_value, end_value]
    double start_value = 1;
    double end_value = 2;
    // 正样本数
    int32 positive = 3;
    // 负样本数
    int32 negative = 4;
    // 样本总数
    int32 total = 5;
    // 精确率：TP/(TP+FP)
    double precision = 6;
    // 召回率：TP/(TP+FN)
    double recall = 7;
    // 误诊率(false positive rate): FP/(FP+TN)
    double fpr = 8;
    // F1值：2*P*R/(P+R)
    // 精确率和召回率的调和平均数
    // 表征了精确率和召回率的平衡程度，值越大说明越平衡
    double f1_score = 9;
    // lift值: 精确度/正样本占比
    // lift值越大，说明模型效果越好
    double lift = 10;
    // 正样本分布：预测正样本/正样本总数
    double distribution_of_positive = 11;
    // 负样本分布：预测负样本/正样本总数
    double distribution_of_negative = 12;
    // 正样本累积占比，用来计算KS值
    double cumulative_percent_of_positive = 13;
    // 负样本累积占比，用来计算KS值
    double cumulative_percent_of_negative = 14;
    // 总计累积百分比，用来表示样本占比
    double total_cumulative_percent = 15;
    // KS值：(cumulative_percent_of_positive -  cumulative_percent_of_negative)
    // 用来表明该桶内正负样本的区分度
    // 绝对值越大表明该桶正负样本区分度越大
    double ks = 16;
    // score均值
    double avg_score = 17;
  }

  SummaryReport summary_report = 1;
  repeated EquiBinReport equi_frequent_bin_report = 2;
  repeated EquiBinReport equi_range_bin_report = 3;
  // 头部准确率，
  // 即 fpr 是 0.001, 0.005, 0.01, 0.05，0.1, 0.2 的时候对应的 precision/recall
  repeated PrReport head_report = 4;
}