import pandas as pd
import numpy as np

# データ読み込み
df = pd.read_csv('data/raw/SSDSE-B-2026.csv', encoding='utf-8', skiprows=1)
print(df.shape)
print(df.dtypes)
print(df.describe())

# 「関係性の可視化」の文脈で扱う場合の例：
# 分野: 可視化
# 関連手法は同カテゴリの他用語を参照してください。

# networkx で都道府県隣接グラフ
import networkx as nx, matplotlib.pyplot as plt
G = nx.Graph()
adj = [('北海道','青森県'),('青森県','岩手県'),('青森県','秋田県'),
       ('岩手県','宮城県'),('岩手県','秋田県'),('秋田県','山形県'),
       ('宮城県','山形県'),('宮城県','福島県'),('福島県','茨城県')]
G.add_edges_from(adj)
nx.draw(G, with_labels=True, node_color='#80CBC4', node_size=900)
plt.savefig('network.png', dpi=150)

# 中心性の計算
deg = dict(G.degree())
betw = nx.betweenness_centrality(G)
import pandas as pd
rank = pd.DataFrame({'degree': deg, 'between': betw}).sort_values('degree', ascending=False)
print(rank)

# 重み付きエッジ : 都道府県人口の対数の積
import pandas as pd, numpy as np
df = pd.read_csv('data/raw/SSDSE-B-2026.csv', encoding='utf-8', skiprows=1)
pop = df.set_index('Prefecture')['A1101']
for u, v in G.edges():
    if u in pop.index and v in pop.index:
        G[u][v]['weight'] = np.log(pop[u] * pop[v])
print(list(G.edges(data=True))[:3])

# Plotly でインタラクティブ可視化
import plotly.graph_objects as go
pos = nx.spring_layout(G, seed=1)
edge_x, edge_y = [], []
for u, v in G.edges():
    edge_x += [pos[u][0], pos[v][0], None]
    edge_y += [pos[u][1], pos[v][1], None]
fig = go.Figure()
fig.add_trace(go.Scatter(x=edge_x, y=edge_y, mode='lines'))
fig.add_trace(go.Scatter(x=[pos[n][0] for n in G.nodes()],
                         y=[pos[n][1] for n in G.nodes()],
                         mode='markers+text',
                         text=list(G.nodes()), textposition='top center'))
fig.write_html('network.html')

import networkx as nx, matplotlib.pyplot as plt
G = nx.karate_club_graph()
fig, axes = plt.subplots(1,2, figsize=(10,5))
nx.draw(G, ax=axes[0], pos=nx.spring_layout(G,seed=1), node_color='#80CBC4')
axes[0].set_title('spring')
nx.draw(G, ax=axes[1], pos=nx.circular_layout(G), node_color='#FFB300')
axes[1].set_title('circular')
plt.savefig('layouts.png', dpi=150)

import networkx as nx
G = nx.karate_club_graph()
try:
    import community as cm
    part = cm.best_partition(G)
    print('コミュニティ数:', len(set(part.values())))
except ImportError:
    print('python-louvain をインストール')

import networkx as nx, pandas as pd
G = nx.karate_club_graph()
pr = nx.pagerank(G); bw = nx.betweenness_centrality(G)
df = pd.DataFrame({'pr':pr,'bw':bw}).sort_values('pr', ascending=False)
print(df.head())