Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

can't generate dataset "pts_m5" #156

Open
binaryiii opened this issue Dec 6, 2023 · 5 comments
Open

can't generate dataset "pts_m5" #156

binaryiii opened this issue Dec 6, 2023 · 5 comments

Comments

@binaryiii
Copy link

Excuse me,when i run the code like this:
dataset = get_dataset("pts_m5", regenerate=True)
there was an error like:
RuntimeError: M5 data is available on Kaggle (https://www.kaggle.com/c/m5-forecasting-accuracy/data). You first need to agree to the terms of the competition before being able to download the data. After you have done that, please copy the files into C:\Users\bzn.mxnet\gluon-ts\datasets\pts_m5.
But i have copied the files into C:\Users\bzn.mxnet\gluon-ts\datasets\pts_m5.

@kashif
Copy link
Collaborator

kashif commented Dec 6, 2023

and can you try:

dataset = get_dataset("m5", regenerate=True)

@binaryiii
Copy link
Author

dataset = get_dataset("m5", regenerate=True)
Yes! I tried that in the Implicit-Quantile-Network-Example.ipynb and run the code successfully.
But in the m5.ipynb and m5-tft.ipynb ,if i try
dataset = get_dataset("m5", regenerate=False)
the result will be very different from the initial result in ipynb.

@Usama-Samad
Copy link

Usama-Samad commented Dec 6, 2023

Getting the same issue, even after copying the files and also putting the regenerate to True.
getting this error
FileNotFoundError: [Errno 2] No such file or directory: '/home/abdul/.mxnet/gluon-ts/datasets/m5/metadata.json'
dont know how to get this metadata file.

@kashif
Copy link
Collaborator

kashif commented Dec 6, 2023

@hanlaoshi
Copy link

Excuse me, Dr. Kashif, I noticed in your publicly available code that the following code is used:
dataset = get_dataset("traffic_nips", regenerate=True)
The training time steps for the "traffic_nips" dataset is set to 4001 in your code. However, in the "Conditioned Normalizing Flow" paper, it is mentioned that the time step for this dataset is 10,413. On the other hand, the "TimeGrad" paper indicates a time step of 4001. Could you please clarify how the dataset with a time step of 10,413 was obtained?

Here is the example of pytorch-ts-0.7.0
https://github.com/zalandoresearch/pytorch-ts/blob/version-0.7.0/examples/Traffic.ipynb

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants