search

proceduralia / pytorch-GAN-timeseries

GANs for time series generation in pytorch
273 stars 77 forks source link

csv format #2

Open Maayan19 opened 4 years ago

Maayan19 commented 4 years ago

Hi, Can you please give more information regarding the input csv file? How the data should be organized? or adding an example of such file? Thanks

weilaimingche commented 4 years ago

Hello, can you tell me where to download the data, or is there a dataset that can be used publicly?

petruccs commented 4 years ago

Hello, can you tell me where to download the data, or is there a dataset that can be used publicly?

I think the csv sould be organized in this way:

data_column;orario_column;BTP_Price

where data_column is the date, orario_column is the time, BTP_Price is the price.

Even with it, I got some errors I am working on:

RuntimeError: Assertion x >= 0. && x <= 1.' failed. input value should be between 0~1, but got -nan at /pytorch/aten/src/THNN/generic/BCECriterion.c:62

weilaimingche commented 4 years ago

Hello, can you tell me where to download the data, or is there a dataset that can be used publicly?

I think the csv sould be organized in this way:

data_column;orario_column;BTP_Price

where data_column is the date, orario_column is the time, BTP_Price is the price.

Even with it, I got some errors I am working on:

RuntimeError: Assertion x >= 0. && x <= 1.' failed. input value should be between 0~1, but got -nan at /pytorch/aten/src/THNN/generic/BCECriterion.c:62

can you share your data with me? I want to try it. my emal is 2676102759@qq.com. We can discuss the code.

petruccs commented 4 years ago

I didn't do much. I created a file with the following values:

data_column;orario_column;BTP_Price 2020/01/01;00:00;0.1 2020/01/01;00:01;0.2 2020/01/01;00:02;0.3 2020/01/01;00:03;0.4 2020/01/01;00:04;0.5 2020/01/01;00:05;0.6

I think that should be the right content of the csv file...