For the second phase, you will be expected to implement a recurring neural network (e.g., LSTM or CNN-LSTM) for learning features from the data and improving on your results from the first part of this project. Results of your baseline methods including and specific metric that we are using will be posted on a separate document. The groups with the best performance on this second phase will receive extra credit.
Deliverables:
1. Predictions for the provided test set
2. Technical report
3. name.csv file containing the unityID of group members
For your report, you should follow the following guidelines:
1. The length should not be more than 3 pages long. No introduction or abstract are needed. Your report should have three sections.
2. Section 1 - Methodology: Include a description of your approach including citations to papers describing the methodology, and references to toolboxes (and specific functions) used for the implementation.
3. Section 2 – Model Training and Hyperparameter Selection: Provide details about your procedure for hyper-parameter tuning.
4. Section 3 – Evaluation: Perform an evaluation of your method including error metrics. We will be giving you a test set for which you will give us your prediction for evaluation.
5. Include several plots showing results of your inference and the groundtruth.
