Python–使用虹膜数据集的熊猫基础知识

import pandas as pd


data = pd.read_csv("your downloaded dataset location ")


data.head()


data.sample(10)


data.columns


#The first one is the number of rows and
# the other one is the number of columns.
data.shape


print(data)


#data[start:end]
#start is inclusive whereas end is exclusive
print(data[10:21])
# it will print the rows from 10 to 20.
# you can also save it in a variable for further use in analysis
sliced_data=data[10:21]
print(sliced_data)


#here in the case of Iris dataset
#we will save it in a another variable named "specific_data"
specific_data=data[["Id","Species"]]
#data[["column_name1","column_name2","column_name3"]]
#now we will print the first 10 columns of the specific_data dataframe.
print(specific_data.head(10))


#here we will use iloc
data.iloc[5]
#it will display records only with species "Iris-setosa".
data.loc[data["Species"] == "Iris-setosa"]


#In this dataset we will work on the Species column, it will count number of times a particular species has occurred.
data["Species"].value_counts()
#it will display in descending order.


# data["column_name"].sum()
sum_data = data["SepalLengthCm"].sum()
mean_data = data["SepalLengthCm"].mean()
median_data = data["SepalLengthCm"].median()
print("Sum:",sum_data, "\nMean:", mean_data, "\nMedian:",median_data)


min_data=data["SepalLengthCm"].min()
max_data=data["SepalLengthCm"].max()
print("Minimum:",min_data, "\nMaximum:", max_data)


# For example, if we want to add a column let say "total_values",
# that means if you want to add all the integer value of that particular
# row and get total answer in the new column "total_values".
# first we will extract the columns which have integer values.
cols = data.columns
# it will print the list of column names.
print(cols)
# we will take that columns which have integer values.
cols = cols[1:5]
# we will save it in the new dataframe variable
data1 = data[cols]
# now adding new column "total_values" to dataframe data.
data["total_values"]=data1[cols].sum(axis=1)
# here axis=1 means you are working in rows,
# whereas axis=0 means you are working in columns.


newcols={
"Id":"id",
"SepalLengthCm":"sepallength"
"SepalWidthCm":"sepalwidth"}
data.rename(columns=newcols,inplace=True)
print(data.head())


#this is an example of rendering a datagram,
which is not visualised by any styles.
data.style


# we will here print only the top 10 rows of the dataset,
# if you want to see the result of the whole dataset remove
#.head(10) from the below code
data.head(10).style.highlight_max(color='lightgreen', axis=0)
data.head(10).style.highlight_max(color='lightgreen', axis=1)
data.head(10).style.highlight_max(color='lightgreen', axis=None)


data.isnull()
#if there is data is missing, it will display True else False.


data.isnull.sum()


import seaborn as sns
iris = sns.load_dataset("iris")
sns.heatmap(iris.corr(),camp = "YlGnBu", linecolor = 'white', linewidths = 1)


sns.heatmap(iris.corr(),camp = "YlGnBu", linecolor = 'white', linewidths = 1, annot = True )


data.corr(method='pearson')


g = sns.pairplot(data,hue="Species")