ろんりかいふく
これは線形方程式を用いた分類アルゴリズムである.クラス確率は、信号関数またはソフトMax関数を使用して出力できます.

import pandas as pd

fish = pd.read_csv('https://bit.ly/fish_csv')
fish.head()

print(pd.unique(fish['Species']))

fish_input = fish[['Weight','Length','Diagonal','Height','Width']].to_numpy()
fish_target = fish['Species'].to_numpy()

from sklearn.model_selection import train_test_split

train_input, test_input, train_target, test_target = train_test_split(
    fish_input, fish_target, random_state=42)

from sklearn.preprocessing import StandardScaler

ss = StandardScaler()
ss.fit(train_input)
train_scaled = ss.transform(train_input)
test_scaled = ss.transform(test_input)

1.k最近接分類器の確率予測

from sklearn.neighbors import KNeighborsClassifier

kn = KNeighborsClassifier(n_neighbors=3)
kn.fit(train_scaled, train_target)

print(kn.score(train_scaled, train_target))
print(kn.score(test_scaled, test_target))

print(kn.classes_)

print(kn.predict(test_scaled[:5]))

import numpy as np

proba = kn.predict_proba(test_scaled[:5])
print(np.round(proba, decimals=4))

distances, indexes = kn.kneighbors(test_scaled[3:4])
print(train_target[indexes])

2.Logistic回帰（バイナリ分類）

bream_smelt_indexes = (train_target == 'Bream') | (train_target == 'Smelt')
train_bream_smelt = train_scaled[bream_smelt_indexes]
target_bream_smelt = train_target[bream_smelt_indexes]

from sklearn.linear_model import LogisticRegression
lr=LogisticRegression()
lr.fit(train_bream_smelt,target_bream_smelt)

print(lr.predict(train_bream_smelt[:5]))
print(lr.predict_proba(train_bream_smelt[:5]))

print(lr.coef_,lr.intercept_)


decisions = lr.decision_function(train_bream_smelt[:5])
print(decisions)

from scipy.special import expit
print(expit(decisions))

3.論理回帰（多分類）

lr=LogisticRegression(C=20,max_iter=1000)
lr.fit(train_scaled,train_target)
print(lr.score(train_scaled,train_target))
print(lr.score(test_scaled,test_target))

print(lr.classes_)

print(lr.predict(test_scaled[:5]))
proba=lr.predict_proba(test_scaled[:5])
print(np.round(proba,decimals=3))

print(lr.coef_,lr.intercept_)
print(lr.coef_.shape,lr.intercept_.shape)

decisions=lr.decision_function(test_scaled[:5])
print(np.round(decisions,decimals=2))

from scipy.special import softmax
proba=softmax(decisions,axis=1)
print(np.round(proba,decimals=3))