xdc_git_python_20210511/aws-task-ruleless/jam-EDA.py

# # import boto3 
# import pandas as pd
# # import s3fs

# df = pd.read_csv('s3://cse-cic-ids2018/Processed Traffic Data for ML Algorithms/Wednesday-28-02-2018_TrafficForML_CICFlowMeter.csv',low_memory=False)

# df = df[df.Label != 'Label']
# # df[100:101]
# print(df[100:101])

import s3fs
fs = s3fs.S3FileSystem(anon=True)
fs = s3fs.S3FileSystem(anon=True)
fs.ls('cse-cic-ids2018')
#s3://cse-cic-ids2018/Processed Traffic Data for ML Algorithms/Wednesday-28-02-2018_TrafficForML_CICFlowMeter.csv
with fs.open('cse-cic-ids2018/Processed Traffic Data for ML Algorithms/Wednesday-28-02-2018_TrafficForML_CICFlowMeter.csv', 'r',encoding='utf-8') as f:
    reader = csv.reader(f)
    header = next(reader)
    print(header)
    i=0
    for row in reader:
        i=i+1
        if i<102 and i>98:
            print(i,row)


len(csvdata)
print(csvdata[0:1000])
import pandas as pd
csvdata = csvdata.decode('utf-8')
data_frame = pd.read_csv(csvdata)
print(len(data_frame))
import csv
reader = csv.reader(csvdata)
print(next(reader))