df = df.assign(temp_f = lambda x: x['temp_c'] * 9 / 5 + 32)

df = df.assign(temp_f = pd.col('temp_c') * 9 / 5 + 32)

  pd.col('a') + 10  # => (col('a') + 10)
  

df.assign(
    city_upper = pd.col('city').str.upper(),
    log_temp_c = np.log(pd.col('temp_c'))
)

import matplotlib.pyplot as plt
import numpy as np

data = np.random.normal(50, 15, 1000)
plt.hist(data, bins=30, color='skyblue', edgecolor='black', alpha=0.7)
plt.xlabel('Values'); plt.ylabel('Count'); plt.title('Гистограмма данных')
plt.show()

from distfit import distfit
import numpy as np

my_data = np.random.normal(25, 8, 2000)
fitter = distfit(method='parametric')
fitter.fit_transform(my_data)

print("Лучшее распределение:", fitter.model['name'])
print("Параметры:", fitter.model['params'])

fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(15,6))
fitter.plot(chart='PDF', ax=ax1); ax1.set_title('PDF')
fitter.plot(chart='CDF', ax=ax2); ax2.set_title('CDF')
plt.show()

from scipy.stats import binom
count_data = binom(20, 0.3).rvs(1000)
discrete_fitter = distfit(method='discrete')
discrete_fitter.fit_transform(count_data)
discrete_fitter.plot()

fitter.bootstrap(my_data, n_boots=100)
print(fitter.summary[['name','score','bootstrap_score','bootstrap_pass']])