phamluan commited on
Commit
4a1bfc7
Β·
verified Β·
1 Parent(s): 1e7f8ec

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +89 -0
README.md ADDED
@@ -0,0 +1,89 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: en
3
+ tags:
4
+ - cryptocurrency
5
+ - polkadot
6
+ - price-prediction
7
+ - machine-learning
8
+ - time-series
9
+ license: mit
10
+ ---
11
+
12
+ # Polkadot (DOT) Price Prediction Models
13
+
14
+ Trained ML models for predicting Polkadot (DOT) cryptocurrency prices.
15
+
16
+ ## πŸ“Š Model Performance
17
+
18
+ | Model | RMSE | MAE |
19
+ |-------|------|-----|
20
+ | Random Forest | 0.2348 | 0.1403 |
21
+ | Gradient Boosting | 0.2031 | 0.1157 |
22
+ | Linear Regression | 0.0200 | 0.0149 |
23
+ | LSTM | 0.2332 | 0.1725 |
24
+
25
+ ## 🎯 Training Details
26
+
27
+ - **Trained on**: 2025-10-24 07:45:22
28
+ - **Data Source**: CoinGecko API
29
+ - **Historical Days**: 365
30
+ - **Features**: 23 technical indicators
31
+ - **GPU**: Accelerated with TensorFlow
32
+
33
+ ## πŸ“¦ Files Included
34
+
35
+ - `polkadot_sklearn_models.pkl`: Scikit-learn models (RF, GB, LR)
36
+ - `polkadot_scaler.pkl`: Feature scaler
37
+ - `polkadot_lstm_model.h5`: LSTM neural network
38
+ - `polkadot_metadata.json`: Training metadata
39
+
40
+ ## πŸš€ Usage
41
+
42
+ ```python
43
+ from huggingface_hub import hf_hub_download
44
+ import joblib
45
+ from tensorflow.keras.models import load_model
46
+
47
+ # Download models
48
+ sklearn_path = hf_hub_download(
49
+ repo_id="YOUR_USERNAME/YOUR_REPO",
50
+ filename="polkadot_sklearn_models.pkl"
51
+ )
52
+ scaler_path = hf_hub_download(
53
+ repo_id="YOUR_USERNAME/YOUR_REPO",
54
+ filename="polkadot_scaler.pkl"
55
+ )
56
+ lstm_path = hf_hub_download(
57
+ repo_id="YOUR_USERNAME/YOUR_REPO",
58
+ filename="polkadot_lstm_model.h5"
59
+ )
60
+
61
+ # Load models
62
+ models = joblib.load(sklearn_path)
63
+ scaler = joblib.load(scaler_path)
64
+ lstm = load_model(lstm_path)
65
+
66
+ # Make predictions
67
+ # (prepare your features first)
68
+ predictions = models['RandomForest'].predict(scaled_features)
69
+ ```
70
+
71
+ ## πŸ“ˆ Features
72
+
73
+ The models use 23 technical indicators including:
74
+ - Moving Averages (SMA 7, 25, 99)
75
+ - Exponential Moving Averages (EMA 12, 26)
76
+ - RSI (Relative Strength Index)
77
+ - MACD & Signal Line
78
+ - Bollinger Bands
79
+ - Stochastic Oscillator
80
+ - Volatility measures
81
+ - Lag features
82
+
83
+ ## ⚠️ Disclaimer
84
+
85
+ These models are for educational and research purposes only. Cryptocurrency markets are highly volatile and unpredictable. Do not use these predictions for actual trading decisions without proper risk management.
86
+
87
+ ## πŸ“„ License
88
+
89
+ MIT License