top of page
Project Flow
Technologies: Google Colab, TensorFlow Keras, Python3
Retrieves data from database or csv hosted on github
​
Training Features:
-
11 epochs, Restore Best Weights, Patience = 5
-
Adam Optimizer Training Rate = 0.01
-
Microbatch size = 1024
-
2 data channels, 6 periodic waveforms
![pseudoflow.png](https://static.wixstatic.com/media/b49bcb_677930ae8e2849dbbb712f69e839a66a~mv2.png/v1/fill/w_600,h_795,al_c,q_90,usm_0.66_1.00_0.01,enc_avif,quality_auto/pseudoflow.png)
Project Flow
LSTM vs Bidirectional LSTM
![LSTMnets.png](https://static.wixstatic.com/media/b49bcb_c73f63b7b7c74fc68a378cf2ffe144bd~mv2.png/v1/fill/w_600,h_682,al_c,q_90,usm_0.66_1.00_0.01,enc_avif,quality_auto/LSTMnets.png)
LSTM vs BI-LSTM
Analysis Flow: Generating the 96-Point Forecast
Input
Raw Output
Apply Weights
Aggregate
Window and Score
Example Shown with Bitcoin (BTC)
Input
![BTC Input.png](https://static.wixstatic.com/media/b49bcb_300985c7dafa42c28934138d41627969~mv2.png/v1/fill/w_659,h_424,al_c,q_85,usm_0.66_1.00_0.01,enc_avif,quality_auto/BTC%20Input.png)
Analysis Flow
Raw Output
![BTC Raw Output.png](https://static.wixstatic.com/media/b49bcb_197c04952d564dc693cae9f5a49a3fd5~mv2.png/v1/fill/w_661,h_350,al_c,q_85,usm_0.66_1.00_0.01,enc_avif,quality_auto/BTC%20Raw%20Output.png)
![BTC Apply Weights.png](https://static.wixstatic.com/media/b49bcb_86a032eaf10c4a56bff61f92f254e26f~mv2.png/v1/fill/w_632,h_315,al_c,q_85,usm_0.66_1.00_0.01,enc_avif,quality_auto/BTC%20Apply%20Weights.png)
Apply Weights
Aggregate
![BTC Aggregate.png](https://static.wixstatic.com/media/b49bcb_17177791bcdf4ff69f1a16796f26789d~mv2.png/v1/fill/w_531,h_350,al_c,q_85,usm_0.66_1.00_0.01,enc_avif,quality_auto/BTC%20Aggregate.png)
![BTC PredvsActual.png](https://static.wixstatic.com/media/b49bcb_bca41f0725c247dbb13a26426bec5b16~mv2.png/v1/fill/w_527,h_339,al_c,q_85,usm_0.66_1.00_0.01,enc_avif,quality_auto/BTC%20PredvsActual.png)
Window and Score
bottom of page