This message was deleted.
# neural-forecasts
Slackbot
08/15/2023, 9:33 PMThis message was deleted.
c
Cristian (Nixtla)
08/15/2023, 9:43 PMHi @Phil. A couple of questions:
• Are you using the latest code from the main branch (with today's PR), or the pip version?
• Do you have "short" time series? In which total length < input_size +h ?
• Have you tried larger
mlp_unitsp
Phil
08/15/2023, 9:50 PMHi Cristian,
• In these experiments, I am not using the latest version. Unfortunately, I am limited to pip version for now. specifically version 1.6.1.
• The length of the ten timeseries in the training set is 676 days. In this case, the input size is 360 and horizon is 180. I have a test portion of 270 days.
• I did try a larger number of mlp_units but the perfomance decreases.
For reference, I hashed out the names of the timeseries and messed around with the scale of the numbers so I don't get in trouble but the general look of the timeseries look like this.
Phil
08/15/2023, 9:52 PMFrom my experiments, the training loss decreases most from adding blocks. The higher the number of blocks the lower the training loss.
c
Cristian (Nixtla)
08/15/2023, 9:53 PMthe validation set is just a classic split right? After the training set of these variables
p
Phil
08/15/2023, 9:54 PMThat's right. I simply call
nf.fit(df=Y_train_df.reset_index(), val_size=180)6 Views