[FEAT] Reduce minimum required size for finetuning

[elephaint]

This reduces the minimum required size for finetuning, by reducing the values in the checks.

[review-notebook-app[bot]]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[github-actions[bot]]

Experiment Results

Experiment 1: air-passengers

Description:

variable	experiment
h	12
season_length	12
freq	MS
level	None
n_windows	1

Results:

metric	timegpt-1	timegpt-1-long-horizon	SeasonalNaive	Naive
mae	12.6793	11.0623	47.8333	76
mape	0.027	0.0232	0.0999	0.1425
mse	213.936	199.132	2571.33	10604.2
total_time	1.3922	0.7272	0.0043	0.0032

Plot:

Experiment 2: air-passengers

Description:

variable	experiment
h	24
season_length	12
freq	MS
level	None
n_windows	1

Results:

metric	timegpt-1	timegpt-1-long-horizon	SeasonalNaive	Naive
mae	58.1031	58.4587	71.25	115.25
mape	0.1257	0.1267	0.1552	0.2358
mse	4040.21	4110.79	5928.17	18859.2
total_time	0.9956	0.8062	0.0037	0.0034

Plot:

Experiment 3: electricity-multiple-series

Description:

variable	experiment
h	24
season_length	24
freq	H
level	None
n_windows	1

Results:

metric	timegpt-1	timegpt-1-long-horizon	SeasonalNaive	Naive
mae	178.293	268.137	269.23	1331.02
mape	0.0234	0.0311	0.0304	0.1692
mse	121588	219496	213677	4.68961e+06
total_time	0.8968	1.8881	0.0046	0.0042

Plot:

Experiment 4: electricity-multiple-series

Description:

variable	experiment
h	168
season_length	24
freq	H
level	None
n_windows	1

Results:

metric	timegpt-1	timegpt-1-long-horizon	SeasonalNaive	Naive
mae	465.532	346.984	398.956	1119.26
mape	0.062	0.0437	0.0512	0.1583
mse	835120	403787	656723	3.17316e+06
total_time	1.0992	1.733	0.0048	0.0044

Plot:

Experiment 5: electricity-multiple-series

Description:

variable	experiment
h	336
season_length	24
freq	H
level	None
n_windows	1

Results:

metric	timegpt-1	timegpt-1-long-horizon	SeasonalNaive	Naive
mae	558.702	459.769	602.926	1340.95
mape	0.0697	0.0566	0.0787	0.17
mse	1.22728e+06	739135	1.61572e+06	6.04619e+06
total_time	1.1119	2.0892	0.0049	0.0044

Plot:

[elephaint]

Can you add a small test finetuning with horizon + 1 samples?

Check, good point, added

[CLAassistant]

All committers have signed the CLA.

[elephaint]

Ray in conjunction with latest Pyarrow gives an error (see this issue). Fixed the pyarrow version to <21.0.0 for now.

[elephaint]

A lot of CI tests are failing, and we also are only testing for very old Python versions.