Project: Getting a Good Night's Sleep

# Start coding here
# Use as many cells as you need
import pandas as pd
df = pd.read_csv('sleep_health_data.csv')
df.head(10)
df.info()
df.describe()
df.isna().any()
df.isna().sum()
pd.set_option('display.max_columns', None)

# Which occupation has the lowest average sleep duration? Save this in a string variable called lowest_sleep_occ.
# Fixing the error by calculating the mean sleep duration for each occupation and then finding the one with the lowest value.
# Ensure the column names are correct
occupation_sleep_mean = df.groupby('Occupation')['Sleep Duration'].mean()
print(occupation_sleep_mean)
#occupation_sleep_mean.plot(kind='bar')

# Find the occupation with the lowest average sleep duration
lowest_sleep_occ = occupation_sleep_mean.sort_values().iloc[0:1].index[0]
print(lowest_sleep_occ)

# Find the occupation with the highest average sleep duration
highest_sleep_occ = occupation_sleep_mean.sort_values(ascending=False).iloc[0:1].index[0]
print(highest_sleep_occ)

# Which occupation has the lowest average sleep quality? Save this in a string variable called lowest_sleep_quality_occ. Did the occupation with the lowest sleep duration also have the lowest sleep quality? If so assign a boolean value to variable same_occ variable, True if it is the same occupation, and False if it isn't.
occupation_sleep_quality = df.groupby('Occupation')['Quality of Sleep'].mean()
print(occupation_sleep_quality)
occupation_sleep_quality.plot(kind='bar')

lowest_sleep_quality_occ = occupation_sleep_quality.sort_values().iloc[0:1].index[0]
print(lowest_sleep_quality_occ)

highest_sleep_quality_occ = occupation_sleep_quality.sort_values(ascending=False).iloc[0:1].index[0]
print(highest_sleep_quality_occ)

# Check if the occupation with the lowest sleep duration is the same as the one with the lowest sleep quality
same_occ = (lowest_sleep_occ == lowest_sleep_quality_occ)
print(same_occ)

same_occ_highest = (highest_sleep_occ == highest_sleep_quality_occ)
print(same_occ_highest)

# Let's explore how BMI Category can affect sleep disorder rates. Start by finding what ratio of app users in each BMI Category have been diagnosed with Insomnia. Create a dictionary named: bmi_insomnia_ratios. The key should be the BMI Category as a string, while the value should be the ratio of people in this category with insomnia as a float rounded to two decimal places. Here is an example:
normal_insomnia = df[(df['BMI Category'] == 'Normal') & (df['Sleep Disorder'] == 'Insomnia')]
print(normal_insomnia)

print(df['BMI Category'].unique())

obese_insomnia = df[(df['BMI Category'] == 'Obese') & (df['Sleep Disorder'] == 'Insomnia')]
print(obese_insomnia)

overweight_insomnia = df[(df['BMI Category'] == 'Overweight') & (df['Sleep Disorder'] == 'Insomnia')]
print(overweight_insomnia)

# Total Normal BMI Category
total_normal = len(df[df['BMI Category'] == 'Normal'])
print(total_normal)

# Total Obese BMI Category
total_obese = len(df[df['BMI Category'] == 'Obese'])
print(total_obese)

# Total Overweight BMI Category
total_overweight= len(df[df['BMI Category']== 'Overweight'])
print(total_overweight)

# Normal and Insomnia count
normal_insomnia_len=len(normal_insomnia)
print(normal_insomnia_len)

# Obese and Insomnia count
obese_insomnia_len=len(obese_insomnia)
print(obese_insomnia_len)

# Overweight and Insomnia count
overweight_insomnia_len=len(overweight_insomnia)
print(overweight_insomnia_len)

# Calculate Ratios
ratio_normal = round((normal_insomnia_len / total_normal), 2) if total_normal > 0 else 0
ratio_obese = round((obese_insomnia_len / total_obese), 2) if total_obese > 0 else 0
ratio_overweight = round((overweight_insomnia_len / total_overweight), 2) if total_overweight > 0 else 0

print(ratio_normal, ratio_obese, ratio_overweight)

bmi_insomnia_ratios={
    "Normal": ratio_normal,
    "Overweight": ratio_overweight,
    "Obese": ratio_obese
}
print(bmi_insomnia_ratios)