pandas: টেবিল নিয়ে কাজ

ticket_id,channel,category,minutes_to_resolve,satisfaction
T1,email,billing,45,4
T2,chat,technical,12,5
T3,chat,billing,8,5
T4,phone,technical,30,
T5,email,account,60,2
T6,chat,technical,15,4
T7,phone,billing,25,3
T8,email,technical,90,1

import pandas as pd

df = pd.read_csv("tickets.csv")
print(df.head(3))
print(df.shape)
print(df.dtypes)

  ticket_id channel   category  minutes_to_resolve  satisfaction
0        T1   email    billing                  45           4.0
1        T2    chat  technical                  12           5.0
2        T3    chat    billing                   8           5.0
(8, 5)
ticket_id                 str
channel                   str
category                  str
minutes_to_resolve      int64
satisfaction          float64
dtype: object

import pandas as pd

df = pd.read_csv("tickets.csv")

print(df["minutes_to_resolve"].mean())                 # একটা কলাম হলো Series
print(df[["ticket_id", "channel"]].tail(2))             # কলামের লিস্ট দিলে DataFrame

slow = df[df["minutes_to_resolve"] > 40]                # শর্ত দিয়ে সারি ছাঁকা
print(slow[["ticket_id", "category", "minutes_to_resolve"]])

chat_billing = df[(df["channel"] == "chat") & (df["category"] == "billing")]
print(len(chat_billing), "chat billing ticket(s)")

35.625
  ticket_id channel
6        T7   phone
7        T8   email
  ticket_id   category  minutes_to_resolve
0        T1    billing                  45
4        T5    account                  60
7        T8  technical                  90
1 chat billing ticket(s)

import pandas as pd

df = pd.read_csv("tickets.csv")

df["hours"] = (df["minutes_to_resolve"] / 60).round(2)          # নতুন কলাম
df["is_slow"] = df["minutes_to_resolve"] > 40
df["label"] = df["category"].str.upper() + "/" + df["channel"]  # .str দিয়ে লেখার মেথড

print(df[["ticket_id", "hours", "is_slow", "label"]].sort_values("hours", ascending=False).head(3))
print(df["channel"].value_counts())

  ticket_id  hours  is_slow            label
7        T8   1.50     True  TECHNICAL/email
4        T5   1.00     True    ACCOUNT/email
0        T1   0.75     True    BILLING/email
channel
email    3
chat     3
phone    2
Name: count, dtype: int64

import pandas as pd

df = pd.read_csv("tickets.csv")

summary = df.groupby("channel").agg(
    tickets=("ticket_id", "count"),
    avg_minutes=("minutes_to_resolve", "mean"),
    avg_satisfaction=("satisfaction", "mean"),
).round(1)
print(summary.sort_values("avg_minutes"))

         tickets  avg_minutes  avg_satisfaction
channel                                        
chat           3         11.7               4.7
phone          2         27.5               3.0
email          3         65.0               2.3

import pandas as pd

df = pd.read_csv("tickets.csv")

print(df.isna().sum())                       # প্রতি কলামে কয়টা মান নেই
print(df[df["satisfaction"].isna()])

filled = df["satisfaction"].fillna(df["satisfaction"].median())
print(filled.tolist())
print(len(df.dropna()), "rows have no missing values")

df.to_csv("tickets_clean.csv", index=False)

ticket_id             0
channel               0
category              0
minutes_to_resolve    0
satisfaction          1
dtype: int64
  ticket_id channel   category  minutes_to_resolve  satisfaction
3        T4   phone  technical                  30           NaN
[4.0, 5.0, 5.0, 4.0, 2.0, 4.0, 3.0, 1.0]
7 rows have no missing values