로또 번호 추출 파이썬 프로그래밍

# main.py

 

import pandas as pd
from collections import Counter
import random
import os
from sklearn.linear_model import LinearRegression
import numpy as np

# 데이터 파일 경로
file_path = 'C:/Wokrspace/PycharmProjects/pythonProject/sample/lotto.txt'

# 파일 경로 확인
if not os.path.exists(file_path):
    print(f"파일을 찾을 수 없습니다: {file_path}")
    exit(1)

# 데이터 파일을 공백으로 구분하여 읽어오기
with open(file_path, 'r') as file:
    lines = file.readlines()

# 각 라인의 공백 제거 및 숫자 리스트로 변환
data = [list(map(int, line.split())) for line in lines if line.strip()]

# 7자리 숫자에서 마지막 숫자(보너스 번호)를 제외하고 6자리로 변환
cleaned_data = [numbers[:6] if len(numbers) == 7 else numbers for numbers in data]

# 데이터가 여러 행에 걸쳐 있는 경우, 데이터 프레임을 일차원 시리즈로 변환
numbers = pd.Series([num for sublist in cleaned_data for num in sublist])

# 분석을 위한 함수들
def analyze_consecutive_numbers(numbers):
    consecutive_counts = []
    for subset in cleaned_data:
        subset.sort()
        count = 0
        for i in range(1, len(subset)):
            if subset[i] == subset[i - 1] + 1:
                count += 1
            else:
                if count > 0:
                    consecutive_counts.append(count + 1)
                count = 0
        if count > 0:
            consecutive_counts.append(count + 1)
    return Counter(consecutive_counts)

def analyze_number_ranges(numbers, range_size=10):
    range_counts = Counter((num // range_size) * range_size for num in numbers)
    return range_counts

def analyze_intervals(numbers):
    intervals = []
    for subset in cleaned_data:
        subset.sort()
        for i in range(1, len(subset)):
            intervals.append(subset[i] - subset[i - 1])
    return Counter(intervals)

def analyze_specific_numbers(numbers):
    specific_counts = Counter(numbers)
    return specific_counts

def trend_analysis():
    # 예: 과거 데이터로부터 추세 분석을 위한 간단한 예제
    X = np.arange(len(cleaned_data)).reshape(-1, 1)  # 간단한 인덱스 배열
    y = [sum(subset) for subset in cleaned_data]  # 각 로또 번호 집합의 합

    model = LinearRegression()
    model.fit(X, y)

    # 예측: 다음 회차의 번호 집합의 합 예측
    next_index = np.array([[len(cleaned_data) + 1]])
    predicted_sum = model.predict(next_index)[0]

    return predicted_sum

# 패턴 분석 결과
consecutive_numbers = analyze_consecutive_numbers(numbers)
sorted_consecutive_numbers = sorted(consecutive_numbers.items(), key=lambda x: x[1], reverse=True)

number_ranges = analyze_number_ranges(numbers, range_size=10)
sorted_number_ranges = sorted(number_ranges.items(), key=lambda x: x[1], reverse=True)

specific_number_counts = analyze_specific_numbers(numbers)
sorted_specific_number_counts = sorted(specific_number_counts.items(), key=lambda x: x[1], reverse=True)

predicted_sum_trend = trend_analysis()

# 패턴 기반 예측: 분석 결과를 반영하여 6자리 번호 생성
def generate_numbers_based_on_patterns(num_predictions=10):
    all_numbers = list(range(1, 46))  # 일반적인 로또 번호 범위 1-45
    predictions = []

    def create_prediction():
        prediction = set()
        while len(prediction) < 6:
            num = random.choice(all_numbers)
            prediction.add(num)
        return sorted(prediction)

    def is_valid_prediction(prediction):
        valid = True
        if sorted_consecutive_numbers:
            most_common_length = sorted_consecutive_numbers[0][0]
            if not any(prediction[i] == prediction[i - 1] + 1 for i in range(1, len(prediction))):
                valid = False
        if sorted_number_ranges:
            most_common_range = sorted_number_ranges[0][0]
            if not any(num in range(most_common_range, most_common_range + 10) for num in prediction):
                valid = False
        return valid

    while len(predictions) < num_predictions:
        prediction = create_prediction()
        if is_valid_prediction(prediction):
            if prediction not in predictions:
                predictions.append(prediction)

    return predictions

def calculate_probability(prediction):
    # 기본 확률을 1로 시작
    probability = 1.0

    # 연속된 숫자 분석 결과의 가중치 반영 (2개 연속된 숫자에 가중치 추가)
    if sorted_consecutive_numbers:
        if 2 in dict(sorted_consecutive_numbers) and dict(sorted_consecutive_numbers)[2] >= 622:
            if any(prediction[i] == prediction[i - 1] + 1 for i in range(1, len(prediction))):
                probability *= 2.0  # 2개 연속된 숫자에 대한 가중치 증가

    # 번호 구간별 빈도 분석의 가중치 반영
    if sorted_number_ranges:
        for range_start, count in sorted_number_ranges:
            if range_start in [10, 20, 30, 0]:
                if any(num in range(range_start, range_start + 10) for num in prediction):
                    # 각 구간에 대한 가중치 증가
                    if range_start == 10:
                        probability *= 1.5  # 10-19 구간에 가중치 증가
                    elif range_start == 20:
                        probability *= 1.4  # 20-29 구간에 가중치 증가
                    elif range_start == 30:
                        probability *= 1.3  # 30-39 구간에 가중치 증가
                    elif range_start == 0:
                        probability *= 1.2  # 0-9 구간에 가중치 증가

    # 특정 숫자 빈도 분석의 가중치 반영
    if sorted_specific_number_counts:
        for num in prediction:
            if num in dict(sorted_specific_number_counts):
                probability *= (1 + dict(sorted_specific_number_counts)[num] / 500)  # 빈도에 따른 가중치 증가

    # 트렌드 분석 결과 반영
    if predicted_sum_trend:
        if sum(prediction) in range(int(predicted_sum_trend - 5), int(predicted_sum_trend + 5)):
            probability *= 1.3  # 예측된 합과 비슷한 경우 가중치 증가

    return probability

predicted_numbers = generate_numbers_based_on_patterns(num_predictions=10)
print("\n예측된 로또 번호 조합과 각 조합의 예상 확률:")
for prediction in predicted_numbers:
    probability = calculate_probability(prediction)
    if probability >= 1:
        print(f"{prediction} - 예상 확률: {probability:.4f}")

print("\n연속된 숫자 분석 결과 (많은 순서):")
for length, count in sorted_consecutive_numbers:
    print(f"{length}개 연속된 숫자: {count}회")

print("\n번호 구간별 빈도 분석 결과 (많은 순서):")
for range_start, count in sorted_number_ranges:
    print(f"{range_start}-{range_start + 9}: {count}회")

print("\n특정 숫자 빈도 분석 결과 (많은 순서):")
for num, count in sorted_specific_number_counts:
    print(f"{num}: {count}회")

 

#main.py END

-----------------------------------------------------------------------------------------------------------------------------------

Github URL

https://github.com/hyoha/Python_Lotto

GitHub - hyoha/Python_Lotto

 

아래 파이썬 로또 분석 프로그램 실행 화면 

 

 

소스의 기준데이터인

lotto.txt 파일은

실제 지금까지 당첨된 실제 로또 번호를

메모장에 전부 넣어서 사용하였습니다.

 

파싱처리한 실제 로또 번호를 분석후 추출하며

 pandas numpy 를 사용하여

학습시 가중치에 변화를 줄수 있는 소스코드입니다.

 

실제 상기 프로그램으로 번호 추출해서

 

동네 편의점에서

로또 번호 3장 구입하여

15(+3)개의 테스트를 해보았으나
실제 맞는 숫자는 없었습니다.

테스트로만 참고하시기 바랍니다.