Skip to content

dunetz/CS230Final

BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

24 Commits
code
code
 
 
data
data
 
 
input
input
 
 
model
model
 
 
papers
papers
 
 
1000 - LSTM 10-day target- Batch size 500 with l2 - 100 epoch.ipynb
1000 - LSTM 10-day target- Batch size 500 with l2 - 100 epoch.ipynb
 
 
1000Test - LSTM 10-day target- Batch size 500 with l2 - 100 epoch.ipynb
1000Test - LSTM 10-day target- Batch size 500 with l2 - 100 epoch.ipynb
 
 
1001 - LSTM 10-day target- Batch size 500 with l2 - 100 epochs - reduce learning rate.ipynb
1001 - LSTM 10-day target- Batch size 500 with l2 - 100 epochs - reduce learning rate.ipynb
 
 
12357452.pdf
12357452.pdf
 
 
12445636.pdf
12445636.pdf
 
 
2000 -CONV2D Base.ipynb
2000 -CONV2D Base.ipynb
 
 
2010 -CONV2D 10-day target.ipynb
2010 -CONV2D 10-day target.ipynb
 
 
2011-CONV2D 10-day target-removing starndarization improves results.ipynb
2011-CONV2D 10-day target-removing starndarization improves results.ipynb
 
 
2014 CONV2D 10-day target- Increase dropout to 60.ipynb
2014 CONV2D 10-day target- Increase dropout to 60.ipynb
 
 
2015 CONV2D 10-day target- Increase epochs to 100.ipynb
2015 CONV2D 10-day target- Increase epochs to 100.ipynb
 
 
2016 - CONV2D 10-day target- Smaller batch size (50) with l2 - 100 epoch (1).ipynb
2016 - CONV2D 10-day target- Smaller batch size (50) with l2 - 100 epoch (1).ipynb
 
 
2016Hybrid - CONV2D 10-day target with hand crafted- Smaller batch size (50) with l2 - 100 epoch.ipynb
2016Hybrid - CONV2D 10-day target with hand crafted- Smaller batch size (50) with l2 - 100 epoch.ipynb
 
 
2016HybridTest- CONV2D 10-day target with hand crafted- Smaller batch size (50) with l2 - 100 epoch.ipynb
2016HybridTest- CONV2D 10-day target with hand crafted- Smaller batch size (50) with l2 - 100 epoch.ipynb
 
 
2016Test CONV2D 10-day target- Smaller batch size (50) with l2 - 100 epoch.ipynb
2016Test CONV2D 10-day target- Smaller batch size (50) with l2 - 100 epoch.ipynb
 
 
2017 - CONV2D 10-day target- increase learning rate epoch.ipynb
2017 - CONV2D 10-day target- increase learning rate epoch.ipynb
 
 
3016 - CONV1D Dilation 10-day target- Smaller batch size (50) with l2 - 100 epoch.ipynb
3016 - CONV1D Dilation 10-day target- Smaller batch size (50) with l2 - 100 epoch.ipynb
 
 
3016Test - CONV1D Dilation 10-day target- Smaller batch size (50) with l2 - 100 epoch-Copy1.ipynb
3016Test - CONV1D Dilation 10-day target- Smaller batch size (50) with l2 - 100 epoch-Copy1.ipynb
 
 
4016 - CONV2D-LSTM 10-day target- 500 batch size with l2 - 100 epoch.ipynb
4016 - CONV2D-LSTM 10-day target- 500 batch size with l2 - 100 epoch.ipynb
 
 
4017- CONV2D-LSTM 10-day target- increase dropout to 60 - 100 epoch.ipynb
4017- CONV2D-LSTM 10-day target- increase dropout to 60 - 100 epoch.ipynb
 
 
4018- CONV2D-LSTM 10-day target - reduce learning rate to 0.005 - 100 epoch.ipynb
4018- CONV2D-LSTM 10-day target - reduce learning rate to 0.005 - 100 epoch.ipynb
 
 
4019- CONV2D-LSTM 10-day decrease batch size to 250 - 100 epoch.ipynb
4019- CONV2D-LSTM 10-day decrease batch size to 250 - 100 epoch.ipynb
 
 
4019Test- CONV2D-LSTM 10-day decrease batch size to 250 - 100 epoch-Copy1.ipynb
4019Test- CONV2D-LSTM 10-day decrease batch size to 250 - 100 epoch-Copy1.ipynb
 
 
5002- LSTM 10-day target- Transfer Learning - For each stock.ipynb
5002- LSTM 10-day target- Transfer Learning - For each stock.ipynb
 
 
5002Test LSTM 10-day target- Transfer Learning - For each stock.ipynb
5002Test LSTM 10-day target- Transfer Learning - For each stock.ipynb
 
 
Model.pdf
Model.pdf
 
 
README.md
README.md
 
 

Repository files navigation

Visualization and Prediction for a Limit Order Book
Folders:
Data: Contains 10 zip files, one for each day of data
Input: Contains classes for reading the zip files and creating Pandas files
Model: Contains classes for generating batches of data
Code: Contains 2 jupyter notebooks for performing basic visualization of the dataset and analysis with a variety of classifiers including logistic regression, random forest, SVM and feed forward neural network.
Papers: Contains papers referenced by the project including a paper describing the dataset
Model.pdf - provides a summary of the model runs performed over the course of the project
Key notebooks
1000: LSTM Training
1000Test: LSTM Test
2016: Conv2d Training
2016Test: Conv2d Test
2016Hybrid: Conv2d with hybrid features
2016HybridTest: Conv2d with hybrid features -- Test
3016: Conv1d with Dilation and Causal Padding (Wavenet-type Model)
3016Test: Wavenet-type Model - Test
4016: Conv2d-LSTM - Training
4016Test: Conv2d-LSTM Test
5002: Transfer Learning
5002Test: Transfer Learning Test

About

No description, website, or topics provided.

Resources

Readme
Activity

Stars

11 stars

Watchers

0 watching

Forks

6 forks
Report repository

Releases

No releases published

Packages

No packages published

Languages