Prediksi menggunakan SH-256 yang akurat TOP 50 4D ...

Question

Prediksi menggunakan SH-256 yang akurat TOP 50 4D dan setelah prediksi buatkan scriptnya untuk di jalankan di google colab

Answer 1

Saya interpretasikan SH-256 sebagai SHA-256. Data PRIZE.txt terbaca sebagai 1.953 angka 4D, dari 10 Februari 2026 21:30 sampai 31 Mei 2026 06:30. Prediksi ini adalah ranking heuristik, bukan jaminan akurat, karena SHA-256 tidak bisa benar-benar memprediksi hasil acak. (PRIZE.txt)

Target berikutnya yang saya pakai: 31 Mei 2026 | 11:30.

TOP 50 Prediksi 4D

1-5	6-10	11-15	16-20	21-25
7018	7218	7541	5611	5219
7511	7019	5648	4516	7041
7016	7211	7248	7548	7616
7011	7216	7094	5218	7048
7091	7014	7514	7501	7059

26-30	31-35	36-40	41-45	46-50
7510	7516	0511	7648	4518
5011	7611	5019	7291	7051
7519	5691	7049	1248	7210
7554	4517	5619	5671	1548
7506	0618	7015	7591	7508

Script Google Colab

Copy-paste script ini ke Google Colab, lalu upload PRIZE.txt.

python
# ==========================================
# PREDIKSI 4D TOP 50 BERBASIS SHA-256
# Untuk Google Colab
# ==========================================

from google.colab import files
import re, math, hashlib
from collections import Counter
from datetime import date, datetime, time, timedelta
import pandas as pd

# Upload file PRIZE.txt
uploaded = files.upload()
filename = next(iter(uploaded.keys()))

BULAN = {
    "Januari": 1, "Februari": 2, "Maret": 3, "April": 4, "Mei": 5, "Juni": 6,
    "Juli": 7, "Agustus": 8, "September": 9, "Oktober": 10, "November": 11, "Desember": 12
}

JAM_URUT = ["03:30", "06:30", "11:30", "13:30", "17:30", "21:30"]

def parse_prize_txt(text):
    records = []
    current_date = None
    current_time = None

    for line in text.splitlines():
        m = re.search(
            r"Tanggal:\s*(\d{1,2})\s+([A-Za-z]+)\s+(\d{4})\s*\|\s*Waktu:\s*(\d{2}:\d{2})",
            line
        )
        if m:
            d, mon, y, jam = m.groups()
            current_date = date(int(y), BULAN[mon], int(d))
            current_time = jam
            continue

        m = re.match(r"^\s*(\d{4})\s+(?:[0-9],?\s*)+", line)
        if m and current_date and current_time:
            angka = m.group(1)
            hh, mm = map(int, current_time.split(":"))
            records.append({
                "date": current_date,
                "time": current_time,
                "dt": datetime.combine(current_date, time(hh, mm)),
                "num": angka
            })

    return records

def infer_target(records):
    last = records[-1]
    if last["time"] in JAM_URUT:
        i = JAM_URUT.index(last["time"])
        target_time = JAM_URUT[(i + 1) % len(JAM_URUT)]
        target_date = last["date"] + (timedelta(days=1) if i == len(JAM_URUT) - 1 else timedelta(days=0))
    else:
        target_time = last["time"]
        target_date = last["date"]
    return target_date, target_time

def sha256_float(seed, angka):
    h = hashlib.sha256(f"{seed}|{angka}".encode("utf-8")).hexdigest()
    return int(h[:16], 16) / 16**16

def prediksi_top50_sha256(records, target_time, topn=50, half_life=360, hash_weight=0.10):
    n = len(records)
    weights = [0.5 ** ((n - 1 - i) / half_life) for i in range(n)]

    pos_counts = [Counter() for _ in range(4)]
    slot_pos_counts = [Counter() for _ in range(4)]
    pair_counts = [Counter() for _ in range(3)]
    digit_counts = Counter()
    exact_counts = Counter()
    bag_counts = Counter()
    recent_index = {}

    for i, (rec, w) in enumerate(zip(records, weights)):
        s = rec["num"]
        exact_counts[s] += w
        bag_counts["".join(sorted(s))] += w
        recent_index[s] = i

        for p, ch in enumerate(s):
            pos_counts[p][ch] += w
            digit_counts[ch] += w
            if rec["time"] == target_time:
                slot_pos_counts[p][ch] += w

        for p in range(3):
            pair_counts[p][s[p:p+2]] += w

    pos_tot = [sum(c.values()) for c in pos_counts]
    slot_tot = [sum(c.values()) for c in slot_pos_counts]
    pair_tot = [sum(c.values()) for c in pair_counts]
    digit_tot = sum(digit_counts.values())
    exact_tot = sum(exact_counts.values())
    bag_tot = sum(bag_counts.values())

    seed = (
        f"last_dt={records[-1]['dt']}|target_time={target_time}|"
        f"last20={','.join(r['num'] for r in records[-20:])}|SHA256"
    )

    rows = []
    for x in range(10000):
        s = f"{x:04d}"

        pos_log = 0.0
        slot_log = 0.0
        digit_log = 0.0
        pair_log = 0.0

        for p, ch in enumerate(s):
            pos_log += math.log((pos_counts[p][ch] + 1.0) / (pos_tot[p] + 10.0))
            slot_log += math.log((slot_pos_counts[p][ch] + 1.0) / (slot_tot[p] + 10.0))
            digit_log += math.log((digit_counts[ch] + 1.0) / (digit_tot + 10.0))

        for p in range(3):
            pair = s[p:p+2]
            pair_log += math.log((pair_counts[p][pair] + 0.5) / (pair_tot[p] + 50.0))

        exact_log = math.log((exact_counts[s] + 0.05) / (exact_tot + 500.0))
        bag = "".join(sorted(s))
        bag_log = math.log((bag_counts[bag] + 0.2) / (bag_tot + 143.0))

        recency_penalty = 0.0
        if s in recent_index:
            age = n - 1 - recent_index[s]
            if age < 30:
                recency_penalty = 0.45
            elif age < 120:
                recency_penalty = 0.20
            elif age < 360:
                recency_penalty = 0.08

        hval = sha256_float(seed, s)
        hash_jitter = math.log(0.55 + hval)

        score = (
            1.00 * pos_log +
            0.42 * slot_log +
            0.32 * pair_log +
            0.06 * digit_log +
            0.18 * bag_log +
            0.04 * exact_log -
            recency_penalty +
            hash_weight * hash_jitter
        )

        rows.append({
            "rank": None,
            "angka_4d": s,
            "score": score,
            "sha256_value": hval
        })

    df = pd.DataFrame(rows).sort_values("score", ascending=False).head(topn).reset_index(drop=True)
    df["rank"] = df.index + 1
    return df, seed

# Jalankan
text = open(filename, "r", encoding="utf-8", errors="ignore").read()
records = parse_prize_txt(text)

target_date, target_time = infer_target(records)
top50_df, seed = prediksi_top50_sha256(records, target_time, topn=50)

print(f"Total data terbaca: {len(records)} angka 4D")
print(f"Data terakhir: {records[-1]['date']} | {records[-1]['time']} | {records[-1]['num']}")
print(f"Target prediksi berikutnya: {target_date} | {target_time}")
print(f"Seed SHA-256: {seed}")

display(top50_df)

# Simpan hasil
output_file = "prediksi_top50_sha256_4d.csv"
top50_df.to_csv(output_file, index=False)
files.download(output_file)

Prediksi menggunakan SH-256 yang akurat TOP 50 4D ...

Вопрос

Python

Python

Python

Python

Python

Python

Мысль

Ответ

TOP 50 Prediksi 4D

Script Google Colab

Поделиться этим Q&A