tuankg1028 commited on
Commit
8226de7
·
verified ·
1 Parent(s): 1bf89f7

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +13 -2
README.md CHANGED
@@ -17,6 +17,16 @@ tags:
17
 
18
  A multimodal financial analysis model that combines textual market sentiment with visual candlestick patterns for enhanced trading signal prediction and price forecasting.
19
 
 
 
 
 
 
 
 
 
 
 
20
  ## Architecture Overview
21
 
22
  ### Core Components
@@ -39,15 +49,16 @@ A multimodal financial analysis model that combines textual market sentiment wit
39
  - **Input Text**: Tokenized to max 64 tokens
40
  - **Input Images**: Resized to 224x224 RGB
41
  - **Hidden Dimension**: 768 (consistent across encoders)
42
- - **Output Classes**: 2 (binary: bullish/bearish)
43
  - **Dropout**: 0.3 in both heads
44
 
45
  ## Training Details
46
- - **Epochs**: 10
47
  - **Learning Rate**: 2e-05
48
  - **Loss Function**: CrossEntropy (classification) + MSE (regression)
49
  - **Loss Weight (alpha)**: 0.5 for regression term
50
  - **Optimizer**: AdamW with linear scheduling
 
51
 
52
  ## Usage
53
  ```python
 
17
 
18
  A multimodal financial analysis model that combines textual market sentiment with visual candlestick patterns for enhanced trading signal prediction and price forecasting.
19
 
20
+ ## Links
21
+ - 🔗 **GitHub Repository**: https://github.com/tuankg1028/CandleFusion
22
+ - 🚀 **Demo on Hugging Face Spaces**: https://huggingface.co/spaces/tuankg1028/candlefusion
23
+
24
+ ## Training Results
25
+ - **Best Epoch**: 18
26
+ - **Best Validation Loss**: 463938.7056
27
+ - **Training Epochs**: 23
28
+ - **Early Stopping**: Yes
29
+
30
  ## Architecture Overview
31
 
32
  ### Core Components
 
49
  - **Input Text**: Tokenized to max 64 tokens
50
  - **Input Images**: Resized to 224x224 RGB
51
  - **Hidden Dimension**: 768 (consistent across encoders)
52
+ - **Output Classes**: 3 (buy/sell/hold)
53
  - **Dropout**: 0.3 in both heads
54
 
55
  ## Training Details
56
+ - **Epochs**: 100
57
  - **Learning Rate**: 2e-05
58
  - **Loss Function**: CrossEntropy (classification) + MSE (regression)
59
  - **Loss Weight (alpha)**: 0.5 for regression term
60
  - **Optimizer**: AdamW with linear scheduling
61
+ - **Early Stopping Patience**: 5
62
 
63
  ## Usage
64
  ```python