fix bug

data/3.python高阶/2.结构化数据分析工具Pandas/4.高级应用/stat.json

 {
-  "source": "stat.py",
+    "author": "huanhuilong",
+    "source": "stat.md",
     "depends": [],
     "exercise_id": 125,
     "type": "code_options"

data/3.python高阶/2.结构化数据分析工具Pandas/4.高级应用/stat.md

+# pandas dataframe之apply
+
+apply的使用
+
+
+## template
+
+```python
+import pandas as pd
+import numpy as np
+
+def add_val(num):
+    if num > 0:
+        return 1
+    else:
+        return 0
+
+
+if __name__ == '__main__':
+    data = {
+        'cloumn_one': pd.Series(np.random.randint(-10, 10, size=5)),
+        'cloumn_two': pd.Series(np.random.randint(0, 10, size=5)),
+        'cloumn_three': pd.Series(np.random.randint(0, 10, size=5))
+    }
+    data_df = pd.DataFrame(data)
+
+    data_df['label_one'] = data['cloumn_one'].apply(add_val)
+
+    print(data_df)
+```
+
+## 答案
+
+```python
+import pandas as pd
+import numpy as np
+
+if __name__ == '__main__':
+    data = {
+        'cloumn_one': pd.Series(np.random.randint(-10, 10, size=5)),
+        'cloumn_two': pd.Series(np.random.randint(0, 10, size=5)),
+        'cloumn_three': pd.Series(np.random.randint(0, 10, size=5))
+    }
+    data_df = pd.DataFrame(data)
+
+    data_df['label_one'] = data['cloumn_one'].apply['num'>1]
+
+    print(data_df)
+```
+
+## 选项
+
+### A
+
+```python
+import pandas as pd
+import numpy as np
+
+def add_val(num):
+    if num > 0:
+        return 1
+    else:
+        return 0
+
+
+if __name__ == '__main__':
+    data = {
+        'cloumn_one': pd.Series(np.random.randint(-10, 10, size=5)),
+        'cloumn_two': pd.Series(np.random.randint(0, 10, size=5)),
+        'cloumn_three': pd.Series(np.random.randint(0, 10, size=5))
+    }
+    data_df = pd.DataFrame(data)
+
+    data_df['label_one'] = data['cloumn_one'].apply(add_val)
+
+    print(data_df)
+```
+
+### B
+
+```python
+import pandas as pd
+import numpy as np
+
+if __name__ == '__main__':
+    data = {
+        'cloumn_one': pd.Series(np.random.randint(-10, 10, size=5)),
+        'cloumn_two': pd.Series(np.random.randint(0, 10, size=5)),
+        'cloumn_three': pd.Series(np.random.randint(0, 10, size=5))
+    }
+    data_df = pd.DataFrame(data)
+
+    data_df['label_one'] = data['cloumn_one'].apply(lambda num: 1 if num>0 else 0)
+
+    print(data_df)
+```
+
+### C
+
+```python
+import pandas as pd
+import numpy as np
+
+if __name__ == '__main__':
+    data = {
+        'cloumn_one': pd.Series(np.random.randint(-10, 10, size=5)),
+        'cloumn_two': pd.Series(np.random.randint(0, 10, size=5)),
+        'cloumn_three': pd.Series(np.random.randint(0, 10, size=5))
+    }
+    data_df = pd.DataFrame(data)
+
+    def add_val(num):
+        return 1 if num>0 else 0
+
+    data_df['label_one'] = data['cloumn_one'].apply(add_val)
+
+    print(data_df)
+```