Some files still can not be derived from the code base

[TronYY]

Dear authors,

I found some files still can not be derived from the code base.

For example, LSTMPrediction.ipynb needs to load some files (e.g., 'ignored_assets/to_lstm_3b_3f.csv', 'ignored_assets/new_lstm_to_pred.csv') to generate different predictor for Matlab simulation part, I searched the whole cs341 code base but found nowhere to generate those files.

Now I can only use LSTMPrediction.ipynb to generate empty predictor, and simulate no prediction scenario.

I want to simulate with prediction part, could you please give me some advice?

(message to your stanford mail couldn't be delivered...)

[federico3]

The LSTM prediction is more up @huevosabio 's alley, and I hope he will be able to comment.

However, looking at the code, I strongly suspect that the predictions we used in the paper are actually the ones in GoodPrediction.ipynb, and not the ones in LSTM_prediction.ipynb.

The GoodPrediction notebook depends on two files, trainPredict.npy and testPredict.npy, which are generated by build_lstm.py.

build_lstm, in turn, needs X.npy and Y.npy, which are generated by data_preprocess.ipynb.

The notebook LSTM_prediction was a previous attempt at LSTM predictions by another coauthor - I am fairly sure we ended up not using them, but Ramon can confirm.

Thanks for the patience, and sorry you have to dig through our unorganized code!

Also, sorry about the Stanford email. My affiliation finally expired last summer.

[federico3]

Closed per communication in #3