Python 中的 pandas 快速上手之:概念初识

import csv

def find_nearest(target, csv_file):
    """
    根据目标数字在排序的CSV文件中查找最接近的数字及对应的值
    
    参数:
    target (int/float): 目标数字
    csv_file (str): CSV文件路径
    
    返回:
    tuple: 最接近的数字和对应的值
    """
    # 打开CSV文件
    with open(csv_file, 'r') as file:
        reader = csv.reader(file)
        
        # 将CSV文件中的数字存入列表
        nums = [float(row[0]) for row in reader]
        
        # 二分查找
        left = 0
        right = len(nums) - 1
        
        while left <= right:
            mid = (left + right) // 2
            
            if nums[mid] == target:
                # 找到目标数字
                nearest_num = nums[mid]
                nearest_val = next(row[1] for row in reader if float(row[0]) == nearest_num)
                return (nearest_num, nearest_val)
            elif nums[mid] < target:
                # 目标数字在右半部分
                left = mid + 1
            else:
                # 目标数字在左半部分
                right = mid - 1
        
        # 没有找到目标数字,返回最接近的数字及对应的值
        nearest_idx = right if right >= 0 and abs(nums[right] - target) < abs(nums[left] - target) else left
        nearest_num = nums[nearest_idx]
        nearest_val = next(row[1] for row in reader if float(row[0]) == nearest_num)
        return (nearest_num, nearest_val)

# 使用示例
target = 1687160644552479420
csv_file = 'ins_can_000000_gaspedel.csv'
nearest_num, nearest_val = find_nearest(target, csv_file)
print(f"离目标数字 {target} 最近的数字是 {nearest_num}, 对应的值为 {nearest_val}")

import pandas as pd

# 读取 csv 文件内容
pd_csv = pd.read_csv("ins_can_000000_gaspedel.csv")
print(pd_csv) # pd_csv 输出
"""
                 timestamp  gas_pedal
0      1687160644552313854     0.0000
1      1687160644552368280     0.0000
2      1687160644552419868     0.0000
3      1687160644552479422     0.0000
4      1687160644552531764     0.0000
...                    ...        ...
110759  1687160857514424803        NaN
110760  1687160857534642423    22.0472
110761  1687160857554423051    22.0472
110762  1687160857617684338    22.0472
110763  1687160857617750929        NaN

[10764 rows x 2 columns]
"""

# 通过向量操作，获取与目标值最接近那一行的索引值
idx = (pd_csv["timestamp"] - target).abs().idxmin()

# 根据索引值获取需要的结果
nearest_num, nearest_value = pd_csv.loc[idx, ['timestamp', 'gas_pedal']]
print(f"离目标数字 {target} 最近的数字是 {nearest_num}, 对应的值为 {nearest_val}")

import pandas as pd

data = {'Name': ['Alice', 'Bob', 'Claire'],
        'Age': [25, 30, 42],
        'Weight': [55.0, 72.5, 61.0]}

res = pd.DataFrame(data)

In [2]: res = pd.DataFrame(data)
   ...: print(res)
   
     Name  Age  Weight
0   Alice   25    55.0
1     Bob   30    72.5
2  Claire   42    61.0

In [3]: print(res["Name"])

0     Alice
1       Bob
2    Claire
Name: Name, dtype: object

In [5]: print(res.index)
RangeIndex(start=0, stop=3, step=1)
In [6]: print(res.columns)
Index(['Name', 'Age', 'Weight'], dtype='object')

# 将返回第0行
In [7]: res.loc[0]
Out[7]: 
Name      Alice
Age          25
Weight     55.0
Name: 0, dtype: object
# 将返回'Age'这一列
In [8]: res['Age']
Out[8]: 
0    25
1    30
2    42
Name: Age, dtype: int6