【ネタ】ロト６とロト７をPythonで予想するプログラム【ver.2】

import pandas as pd
import numpy as np
from sklearn.ensemble import RandomForestClassifier

# 過去の抽選結果データ
data = [

[2,16,23,38,40,43],
[7,8,9,19,22,38],
[25,31,33,35,37,42],
[19,29,36,37,41,42],
[6,18,29,33,35,40],
[5,12,15,16,28,40],
[7,23,27,30,34,36],
[5,19,23,26,31,42],
[2,4,9,19,33,36],
[3,9,29,30,39,42],

    #上記の数字は一部です。実際の過去の数字全て入力する場合はご自身で入力してください。
]

# データフレームに変換
df = pd.DataFrame(data, columns=[f'num_{i+1}' for i in range(6)])

# 各数字の出現頻度を計算
freq = pd.Series(np.concatenate(df.values)).value_counts().sort_index()

# 特徴量の作成
X = []
y = []
for i in range(len(data) - 1):
    current_numbers = data[i]
    next_numbers = data[i + 1]
    features = [freq[num] for num in current_numbers]
    X.append(features)
    y.append(next_numbers)

X = np.array(X)
y = np.array(y)

# モデルの訓練
model = RandomForestClassifier(n_estimators=100, random_state=42)
model.fit(X, y)

# 最新の抽選結果
latest_numbers = data[-1]
latest_features = np.array([freq[num] for num in latest_numbers]).reshape(1, -1)

# 重複しない6個の数字を予測
next_pred = model.predict(latest_features)

# 重複を避けるための処理
predicted_numbers = list(next_pred)
available_numbers = set(range(1, 44)) - set(latest_numbers)  # 1から43までの数字から最新の抽選結果を除外

predicted_numbers = np.array(predicted_numbers)
while len(predicted_numbers) < 6:
    additional_number = np.random.choice(list(available_numbers))
    predicted_numbers = np.append(predicted_numbers, additional_number)

predicted_numbers = predicted_numbers[:6]  # 必要なだけ6つに切り詰める

print(f'Next predicted numbers: {predicted_numbers}')

import pandas as pd
import numpy as np
from sklearn.ensemble import RandomForestClassifier

# データの読み込み
data = [
 
[1,3,9,12,25,26,35],
[13,26,27,29,30,34,37],
[1,2,4,19,21,26,28],
[11,12,18,19,23,26,31],
[4,5,13,16,22,23,35],
[2,12,18,20,25,28,30],
[11,13,14,18,19,23,31],
[11,13,19,25,29,36,37],
[2,15,19,20,22,31,36],
[7,19,23,33,34,35,36],
[4,13,14,15,17,27,33],
    
    #上記の数字は一部です。実際の過去の数字全て入力する場合はご自身で入力してください。
]

# データフレームに変換
df = pd.DataFrame(data, columns=[f'num_{i+1}' for i in range(7)])

# 各数字の出現頻度を計算
freq = pd.Series(np.concatenate(df.values)).value_counts().sort_index()

# 特徴量の作成
X = []
y = []
for i in range(len(data) - 1):
    current_numbers = data[i]
    next_numbers = data[i + 1]
    features = [freq[num] for num in current_numbers]
    X.append(features)
    y.append(next_numbers)

X = np.array(X)
y = np.array(y)

# モデルの訓練
model = RandomForestClassifier(n_estimators=100, random_state=42)
model.fit(X, y)

# 最新の抽選結果
latest_numbers = data[-1]
latest_features = np.array([freq[num] for num in latest_numbers]).reshape(1, -1)

# 重複しない7個の数字を予測
next_pred = model.predict(latest_features)

print(f'Next predicted numbers: {next_pred}')

import pandas as pd
import numpy as np
from sklearn.ensemble import RandomForestClassifier

# 過去の抽選結果データ
data = [
    [7, 10, 12, 17, 23, 28],
    [20, 24, 29, 31, 33, 35],
    [2, 7, 8, 11, 23, 31],
    [12, 13, 23, 24, 28, 29],
    [3, 4, 5, 16, 21, 28],
    [5, 15, 23, 30, 34, 35],
    #上記の数字は一部です。実際の過去の数字全て入力する場合はご自身で入力してください。
]

# データフレームに変換
df = pd.DataFrame(data, columns=[f'num_{i+1}' for i in range(6)])

# 各数字の出現頻度を計算
freq = pd.Series(np.concatenate(df.values)).value_counts().sort_index()

# 特徴量の作成
X = []
y = []
for i in range(len(data) - 1):
    current_numbers = data[i]
    next_numbers = data[i + 1]
    features = [freq[num] for num in current_numbers]
    X.append(features)
    y.append(next_numbers)

X = np.array(X)
y = np.array(y)

# モデルの訓練と予測
def train_and_predict(random_state):
    model = RandomForestClassifier(n_estimators=100, random_state=random_state)
    model.fit(X, y)
    latest_numbers = data[-1]
    latest_features = np.array([freq[num] for num in latest_numbers]).reshape(1, -1)
    next_pred = model.predict(latest_features)
    
    # 重複を避けるための処理
    predicted_numbers = list(next_pred[0])
    available_numbers = set(range(1, 44)) - set(latest_numbers)  # 1から43までの数字から最新の抽選結果を除外
    
    while len(predicted_numbers) < 6:
        additional_number = np.random.choice(list(available_numbers))
        if additional_number not in predicted_numbers:
            predicted_numbers.append(additional_number)
    
    return np.array(predicted_numbers[:6])  # 必要なだけ6つに切り詰める

# 予測結果を3パターン生成
predictions = []
random_states = [42, 100, 200]  # 異なるシードを設定

for state in random_states:
    pred = train_and_predict(state)
    predictions.append(pred)

print('Next predicted numbers (3 patterns):')
for i, pred in enumerate(predictions, start=1):
    print(f'Pattern {i}: {pred}')

import pandas as pd
import numpy as np
from sklearn.ensemble import RandomForestClassifier

# 過去の抽選結果データ
data = [
    [7, 10, 12, 17, 23, 28, 34],
    [11, 13, 14, 18, 19, 23, 31],
    [11, 13, 19, 25, 29, 36, 37],
    [2, 15, 19, 20, 22, 31, 36],
    [7, 19, 23, 33, 34, 35, 36],
    [4, 13, 14, 15, 17, 27, 33],
    #上記の数字は一部です。実際の過去の数字全て入力する場合はご自身で入力してください。
]

# データフレームに変換
df = pd.DataFrame(data, columns=[f'num_{i+1}' for i in range(7)])

# 各数字の出現頻度を計算
freq = pd.Series(np.concatenate(df.values)).value_counts().sort_index()

# 特徴量の作成
X = []
y = []
for i in range(len(data) - 1):
    current_numbers = data[i]
    next_numbers = data[i + 1]
    features = [freq[num] for num in current_numbers]
    X.append(features)
    y.append(next_numbers)

X = np.array(X)
y = np.array(y)

# モデルの訓練と予測
def train_and_predict(random_state):
    model = RandomForestClassifier(n_estimators=100, random_state=random_state)
    model.fit(X, y)
    latest_numbers = data[-1]
    latest_features = np.array([freq[num] for num in latest_numbers]).reshape(1, -1)
    next_pred = model.predict(latest_features)
    return next_pred[0]

# 予測結果を3パターン生成
predictions = []
random_states = [42, 100, 200]  # 異なるシードを設定

for state in random_states:
    pred = train_and_predict(state)
    predictions.append(pred)

print('Next predicted numbers (3 patterns):')
for i, pred in enumerate(predictions, start=1):
    print(f'Pattern {i}: {pred}')