| 22 | return second_column_list |
| 23 | |
| 24 | def table_label_value(df, n, m): |
| 25 | label_value = {} |
| 26 | count = 0 |
| 27 | for row in df.itertuples(index=False): |
| 28 | count += 1 |
| 29 | key = row[1] |
| 30 | # key = getattr(row, df.columns[1]) |
| 31 | # key = df.iloc[row,1] |
| 32 | value = 1.0 / getattr(row, df.columns[2]) |
| 33 | # value = 1.0 / row[2] # 第三列作为值 |
| 34 | if key in label_value: |
| 35 | label_value[key] += value |
| 36 | else: |
| 37 | label_value[key] = value |
| 38 | for key, value in label_value.items(): |
| 39 | label_value[key] = (value ** int(n)) / (count ** int(m)) |
| 40 | return label_value |
| 41 | #一对向量的点乘结果 |
| 42 | def count_SEP(label_value_set1, label_value_set2): |
| 43 | common_keys = set(label_value_set1.keys()).intersection(label_value_set2.keys()) |