PROSTLE commited on
Commit
ddc5f97
·
unverified ·
1 Parent(s): b263487

Add files via upload

Browse files
backend/trained_models/fast_report.py ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import os
2
+ import sys
3
+ import time
4
+ import numpy as np
5
+ import pandas as pd
6
+
7
+ # Path setup
8
+ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
9
+ from stock_data import fetch_stock_data, add_technical_indicators
10
+ from model import train_and_predict, _prediction_cache
11
+
12
+ def fast_report(ticker):
13
+ print(f"\n🚀 GENERATING INSTANT ACCURACY REPORT FOR: {ticker}")
14
+ print("-" * 50)
15
+
16
+ try:
17
+ # 1. Get Data
18
+ df = fetch_stock_data(ticker, period="1y", interval="1d")
19
+ df = add_technical_indicators(df)
20
+ df = df.dropna().reset_index(drop=True)
21
+
22
+ # 2. Setup 1 Window (Most Recent)
23
+ cutoff = len(df) - 5
24
+ hist_df = df.iloc[:cutoff].copy()
25
+ future_df = df.iloc[cutoff:cutoff+5].copy()
26
+ actual_prices = future_df["Close"].values.tolist()
27
+ dates = [str(d)[:10] for d in future_df["Date"]]
28
+ current_price = float(hist_df["Close"].iloc[-1])
29
+
30
+ # 3. Predict
31
+ _prediction_cache.clear()
32
+ print(f"⏳ Training LSTM on latest data (takes ~40s)...")
33
+ t0 = time.time()
34
+ pred = train_and_predict(hist_df)
35
+ elapsed = time.time() - t0
36
+
37
+ # 4. Score
38
+ predicted = pred["predicted_prices"]
39
+ mape = np.mean([abs(p - a) / a * 100 for p, a in zip(predicted, actual_prices)])
40
+ accuracy = 100 - mape
41
+
42
+ print(f"\n✅ REPORT COMPLETE IN {elapsed:.1f}s")
43
+ print(f"{'='*70}")
44
+ print(f" STOCK: {ticker.upper()}")
45
+ print(f"{'='*70}")
46
+ print(f" {'DATE':<12} | {'PREDICTED':<12} | {'ACTUAL':<12} | {'DIFF %':<8}")
47
+ print(f" {'-'*12}-|-{'-'*12}-|-{'-'*12}-|-{'-'*8}")
48
+
49
+ for d, p, a in zip(dates, predicted, actual_prices):
50
+ p_val = f"₹{p:.2f}"
51
+ a_val = f"₹{a:.2f}"
52
+ diff = abs(p - a) / a * 100
53
+ print(f" {d:<12} | {p_val:<12} | {a_val:<12} | {diff:>7.2f}%")
54
+
55
+ print(f"{'='*70}")
56
+ print(f" OVERALL ACCURACY: {accuracy:.2f}%")
57
+ print(f" CONFIDENCE : {pred.get('confidence', 0)}%")
58
+ print(f"{'='*70}\n")
59
+
60
+ except Exception as e:
61
+ print(f"❌ ERROR: {e}")
62
+
63
+ if __name__ == "__main__":
64
+ ticker = sys.argv[1] if len(sys.argv) > 1 else "RELIANCE.NS"
65
+ fast_report(ticker)