How to show the progress of training process when using model.fit

[hieutv85]

Hi there,

I am a new user of tsai. I want to apply time series classification methods of tsai to multivariate time series data.
My time series data is so large so when I apply model.fit it take a lot of time to run.

Is there anyway to show the progress of the training process? (such as the verbose parameter of keras model.fit)
Since it take a lot of time to run so I want to estimate how much time will it take.

Thank you very much,

[hieutv85]

Hi guys, please help me.
Please support me @oguiza . Thank you very much.

Below is my code:

from tsai.basics import *
tfms = [None, TSClassification()]
batch_tfms = TSStandardize(by_sample=True)
mv_clf = TSClassifier(X_sw,yr, splits=split, path='models', arch="InceptionTimePlus", tfms=tfms, batch_tfms=batch_tfms, metrics=accuracy, cbs=ShowGraph())
mv_clf.fit_one_cycle(10, 1e-2)

[albertoazzari]

How long the timeseries is ?
The fit_one_cycle() method should print a "summary" of the training process for every epoch.

[hieutv85]

Hi @albertoazzari,

The length of the time series is 1.5 million time lags. And I have 15 similar time series like that.
I already trained one time series on a high performance computer and I cannot see the results after 3 days.
If I train all of 15 time series and do not know about the progress then it will very inconvenient.

[oguiza]

Hi @hieutv85 ,
When training a model you get an updated with every batch.
It doesn't make sense to train a model with a time series of 1.5 time steps. I don't know what you are trying to learn, but you should probably apply a sliding window.