心跳信号分类预测Task3 特征处理

时间序列数据格式处理
  加入时间步特征time

# 包导入
import pandas as pd
import numpy as np
import tsfresh as tsf
from tsfresh import extract_features, select_features
from tsfresh.utilities.dataframe_functions import impute

data_train = pd.read_csv("train.csv")
data_test_A = pd.read_csv("testA.csv")
print(data_train.shape)
print(data_test_A.shape)

(100000, 3)
(20000, 2)

data_train.head()

data_test_A.head()

# 对心电特征进行行转列处理，同时为每个心电信号加入时间步特征time
train_heartbeat_df = data_train["heartbeat_signals"].str.split(",", expand=True).stack()
train_heartbeat_df = train_heartbeat_df.reset_index()
train_heartbeat_df = train_heartbeat_df.set_index("level_0")
train_heartbeat_df.index.name = None
train_heartbeat_df.rename(columns={"level_1":"time", 0:"heartbeat_signals"}, inplace=True)
train_heartbeat_df["heartbeat_signals"] = train_heartbeat_df["heartbeat_signals"].astype(float)
train_heartbeat_df

# 将处理后的心电特征加入到训练数据中，同时将训练数据label列单独存储
data_train_label = data_train["label"]
data_train = data_train.drop("label", axis=1)
data_train = data_train.drop("heartbeat_signals", axis=1)
data_train = data_train.join(train_heartbeat_df)
data_train

data_train[data_train["id"]==1]

data_train.head()

data_train.shape

(20500000, 3)

portion_data_train=data_train.iloc[:30,:]

portion_data_train

from tsfresh import extract_features
# 特征提取
train_features = extract_features(portion_data_train, column_id='id', column_sort='time')
train_features

Feature Extraction: 100%|████████████████████████████████████████████████████████████████| 1/1 [00:09<00:00,  9.45s/it]

from tsfresh.utilities.dataframe_functions import impute
import warnings
warnings.filterwarnings("ignore")
# 去除抽取特征中的NaN值
train_features=impute(train_features)

接下来，按照特征和响应变量之间的相关性进行特征选择，这一过程包含两步：首先单独计算每个特征和响应变量之间的相关性，然后利用Benjamini-Yekutieli procedure进行特征选择，决定哪些特征可以被保留。

## from tsfresh import select_features
# 按照特征和数据label之间的相关性进行特征选择
train_features_filtered = select_features(train_features, data_train_label[:2])
train_features_filtered

---------------------------------------------------------------------------
AssertionError                            Traceback (most recent call last)
<ipython-input-48-f2da600384d2> in <module>
      2 
      3 # 按照特征和数据label之间的相关性进行特征选择
----> 4 train_features_filtered = select_features(train_features, data_train_label[:2])
      5 
      6 train_features_filtered
E:\My_ruan_jian\anaconda3\lib\site-packages\tsfresh\feature_selection\selection.py in select_features(X, y, test_for_binary_target_binary_feature, test_for_binary_target_real_feature, test_for_real_target_binary_feature, test_for_real_target_real_feature, fdr_level, hypotheses_independent, n_jobs, show_warnings, chunksize, ml_task, multiclass, n_significant)
    151     )
    152     assert len(y) > 1, "y must contain at least two samples."
--> 153     assert len(X) == len(y), "X and y must contain the same number of samples."
    154     assert (
    155         len(set(y)) > 1
AssertionError: X and y must contain the same number of samples.