python_analysis/second_semestr/pract4-2.py - main - Nik_voz/rgpu_python_analysis

python_analysis/second_semestr/pract4-2.py
153 строки4 KB
Nick-voz
another commit.
23 дек 2025, 17:37
23 дек 2025, 17:37e0763c4
import os

import matplotlib.pyplot as plt
import pandas as pd
import seaborn as sns
from constants import Columns
from dotenv import load_dotenv

load_dotenv()

if (DATASET_PATH := os.environ.get("DATASET_PATH")) is None:
    raise RuntimeWarning("DATASET_PATH is not set in environment")

data = pd.read_csv(DATASET_PATH, sep=";")

num_cols = [
    Columns.YEAR,
    Columns.COUNT_GRADUATE,
    Columns.PERCENT_EMPLOYED,
    Columns.AVERAGE_SALARY_FACT_AVG,
    Columns.AVERAGE_SALARY_FACT_MED,
    Columns.AVERAGE_SALARY_NORM_AVG,
    Columns.AVERAGE_SALARY_NORM_MED,
    Columns.AVERAGE_AGE,
    Columns.PERCENT_COMBINE_WORK_AND_STUDY,
]

num_cols = [c for c in num_cols if c in data.columns]

df_num = data[num_cols].copy()
df_num = df_num.apply(pd.to_numeric, errors="coerce")

corr = df_num.corr()

plt.figure(figsize=(10, 8))
sns.heatmap(
    corr, annot=True, fmt=".2f", cmap="RdBu_r", vmin=-1, vmax=1, linewidths=0.5
)
plt.title("Матрица корреляций числовых признаков")
plt.tight_layout()
plt.show()

salary_cols = [
    Columns.AVERAGE_SALARY_FACT_AVG,
    Columns.AVERAGE_SALARY_NORM_AVG,
    Columns.AVERAGE_SALARY_NORM_MED,
]
salary_cols = [c for c in salary_cols if c in df_num.columns]

df_num["salary_sum_3"] = df_num[salary_cols].sum(axis=1, skipna=True)
df_num["salary_mean_3"] = df_num[salary_cols].mean(axis=1, skipna=True)

corr_updated = df_num.corr()

plt.figure(figsize=(10, 8))
sns.heatmap(
    corr_updated,
    annot=True,
    fmt=".2f",
    cmap="RdBu_r",
    vmin=-1,
    vmax=1,
    linewidths=0.5,
)
plt.title(
    "Матрица корреляций (с суммой и средним по трём зарплатным показателям)"
)
plt.tight_layout()
plt.show()

pair_cols = [
    Columns.AVERAGE_SALARY_NORM_AVG,
    Columns.PERCENT_EMPLOYED,
    Columns.AVERAGE_AGE,
    Columns.PERCENT_COMBINE_WORK_AND_STUDY,
]
pair_cols = [c for c in pair_cols if c in df_num.columns]

_ = sns.pairplot(
    df_num[pair_cols].dropna(),
    kind="scatter",
    plot_kws={"alpha": 0.3, "s": 10},
)
for ax in _.axes.flatten():
    plt.setp(ax.get_xticklabels(), rotation=45, ha="right")
    plt.setp(ax.get_yticklabels(), rotation=45, ha="right")

plt.suptitle("Парный график выбранных показателей", y=1)
plt.show()

x_col = (
    Columns.PERCENT_EMPLOYED
    if Columns.PERCENT_EMPLOYED in df_num.columns
    else pair_cols[0]
)
y_col = (
    Columns.AVERAGE_SALARY_NORM_AVG
    if Columns.AVERAGE_SALARY_NORM_AVG in df_num.columns
    else pair_cols[0]
)

plt.figure(figsize=(8, 6))
sns.regplot(
    x=df_num[x_col],
    y=df_num[y_col],
    scatter_kws={"alpha": 0.3, "s": 10},
    line_kws={"color": "red"},
)
plt.xlabel(x_col)
plt.ylabel(y_col)
plt.title(f"Регрессия: {y_col} ~ {x_col}")
plt.tight_layout()
plt.show()

dist_col = (
    Columns.AVERAGE_SALARY_NORM_AVG
    if Columns.AVERAGE_SALARY_NORM_AVG in df_num.columns
    else salary_cols[0]
)

plt.figure(figsize=(8, 5))
sns.histplot(df_num[dist_col].dropna(), bins=60, kde=True)
plt.xlabel(dist_col)
plt.title(f"Распределение {dist_col}")
plt.tight_layout()
plt.show()

z_cols = [
    Columns.AVERAGE_SALARY_NORM_AVG,
    Columns.PERCENT_EMPLOYED,
    Columns.AVERAGE_AGE,
]
z_cols = [c for c in z_cols if c in df_num.columns]

df_3d = df_num[z_cols].dropna()

fig = plt.figure(figsize=(9, 7))
ax = fig.add_subplot(111, projection="3d")
ax.scatter(
    df_3d[z_cols[0]],
    df_3d[z_cols[1]],
    df_3d[z_cols[2]],
    c=df_3d[z_cols[0]],
    cmap="viridis",
    s=8,
    alpha=0.6,
)
ax.set_xlabel(z_cols[0])
ax.set_ylabel(z_cols[1])
ax.set_zlabel(z_cols[2])
ax.set_title("3D-диаграмма рассеивания")
plt.tight_layout()
plt.show()