一、常見檔案類型與存取方式

01. Text

# Read Text
with open("file.txt") as file:
data = file.read()
data_in_line = file.read().splitlines()

# REwrite whole file
with open("my_file.txt", mode='w') as file:
content = file.write("new text inside")

# Append content after the original
with open("my_file.txt", mode='a') as file:
content = file.write("new text inside")

02. JSON

Write
json.dump(input_data, output_file)

Read
json.load(inptput_file)

Update
json.update(append_data)

new_data = {
        website: {
            "email": email,
            "password": password,
        }
    }

with open("data.json", "r") as file:
        Reading old data
                data = json.load(file)

    except FileNotFoundError:
        with open("data.json", "w") as file:
                json.dump(new_data, file, indent=4)
        else:
            # Updating old data with new data
            data.update(new_data)

            with open("data.json", "w") as file:
                # Saving updated data
                json.dump(data, file, indent=4)

03. CSV

# CSV => DataFrame
data = pandas.read_csv("data/word_to_learn.csv")

# DataFrame => dictionary in list
to_learn = data.to_dict("records")

# list => DataFrame
to_learn_dataframe = pandas.DataFrame(to_learn)

# DataFrame => CSV
to_learn_dataframe.to_csv("data/word_to_learn.csv", index=False)

二、常見資料型態與建立

01. List

List Comprehension

new_list = [new_item for item in list]
new_list = [new_item for item in list if test]

Create from Series

temp_list = data["temp"].to_list()

remove an item from list

to_learn.remove(current_card)

02. Dictionary

Dictionary Comprehension

new_dict = {new_key:new_value for item in list}
new_dict = {new_key:new_value for (key,value) in dict.items()}
new_dict = {new_key:new_value for (key,value) in dict.items() if test}

Create from Data Frame

data_dict = data.to_dict()

03. Data Frame

Create from Dictionary
student_data_frame = pandas.DataFrame(student_score)

student_score = {
    "students": ["Amy", "James", "Penny"],
    "scores": [76, 56, 65]
}

Create from CSV
student_data_frame = pandas.read_csv("student_score.csv")

  student  score
0     Amy     56
1   James     76
2    Lily     98

04. Series

Create from Data Frame

student_score = data["score"]
student_score = data.score

Calculate data in Series

# 求 student_score 的最大值
highest_score = data["temp"].max()

# 求 student_score 的平均值
average_score = data.temp.mean()

# 求 weekday = "Monday" 時的攝氏氣溫並轉成華氏氣溫
monday = data[data.day == "Monday"]
f_temp = int(monday.c_temp) * 9 / 5 + 32

# 求Monday的資料
monday_info = data[data.day == "Monday"]

# 求最高溫的日期
highest_temp_day = data[data.temp == data.temp.max()]

三、Data Frame 內部資料取得

# Dictionary
student_score = {
    "students": ["Amy", "James", "Penny"],
    "scores": [76, 56, 65]
}

# Data Frame
  student  score
0     Amy     56
1   James     76
2    Lily     98

01. Dictionary.items()

for(key, value) in student_dict.items():
print(key)

student
score

for(key, value) in student_dict.items():
print(value)

['Amy', 'James', 'Lily']
[56, 76, 98]

02. Data Frame.items() => Python內建語法

for(key, value) in student_data_frame.items():
print(key)

student
score

for(key, value) in student_data_frame.items():
print(value)

0    Amy
1    James
2    Lily
Name: score, dtype: int64

0    56
1    76
2    98
Name: score, dtype: int64

03. Data Frame.iterrows() => Pandas改良語法

index (直行資料)，印出表格第一直行之索引
for(index, row) in student_data_frame.iterrows():
print(index)

0
1
2

row (橫列資料)，印出表格每一橫列之內容(且保留標題)
for(index, row) in student_data_frame.iterrows():
print(row)

student    Amy
score       56
Name: 0, dtype: object

student    James
score         76
Name: 1, dtype: object

student    Lily
score        98
Name: 2, dtype: object

04. 特定條件下資料取得

# 求學生名單
for(index, row) in student_data_frame.iterrows():
    print(row.student)

# 也可用Series取得
student = tudent_data_frame.student

# output
> Amy
> James
> Lily

# 求Amy的成績
for(index, row) in student_data_frame.iterrows():
    if row.student == 'Amy'
    print(row.score)

# 也可用Series取得
amy_info = student_data_frame[student_data_frame.score == "Amy"]
amy_score = amy.info.score

# output
> 56

# 求成績最高學生
highest_score_student = student_data_frame.name[student_data_frame.score == student_data_frame.name.max()]