vincekellner / demandforecasting

Forecast taxi demand for given areas in New York City
15 stars 2 forks source link

Simple Autoencoder & UBER Autoencoder notebook refers to functions and structures not declared #1

Closed juhariis closed 4 years ago

juhariis commented 4 years ago

In section "Start Training of Models" / "Vanilla Auroencoder" ts_largest has not been defined when called

ts_series_09_11_area237 = ts_largest['2009':'2011'].iloc[:,0].copy()

Perhaps one should have created that earlier when creating ts_10largest and ts_20largest?

The function used below has not been defined.

all_results_batch512_drop03_W24_area237__2011 = get_full_autoencoder_pred_model_2H_128_32(ts_series_09_11_area237, multivariate, row_train_test_split, n_timesteps, n_features, stacked_encoder, n_batch_size_autoencoder, n_batch_size_pred_model, dropout_rate, n_epochs, model_name)

Furthermore, the csv file referred to in the notebook does not have the same name as the one in the repo.

I presume the repo is under active further development and you are aware of internal inconsistencoes? Anyway, it looks very interesting project, Good luck!

vincekellner commented 4 years ago

Hi, thank you for your helpful suggestions. I've updated the code and uploaded the missing csv-files.