About the dataset setting

[cccccj-03]

I want to reproduce the experimental results, but it seems some hyper-parameters setting is not clear, the best performance of the Covid dataset is mae:0.134224995970726, rmse:0.1728876382112503, with a setting as below:

data loader

parser.add_argument('--data', type=str, default='Covid', help='dataset type')
parser.add_argument('--root_path', type=str, default='./dataset/', help='root path of the data file')
parser.add_argument('--data_path', type=str, default='covid.csv', help='data file')
parser.add_argument('--channel_independence', type=int, default=0, help='1: channel dependence 0: channel independence')parser.add_argument('--features', type=str, default='MS',
help='forecasting task, options:[M, S, MS]; M:multivariate predict multivariate, S:univariate predict univariate, MS:multivariate predict univariate')
parser.add_argument('--target', type=str, default='OT', help='target feature in S or MS task')
parser.add_argument('--freq', type=str, default='m',
help='freq for time features encoding, options:[s:secondly, t:minutely, h:hourly, d:daily, b:business days, w:weekly, m:monthly], you can also use more detailed freq like 15min or 3h')
parser.add_argument('--checkpoints', type=str, default='./checkpoints/', help='location of model checkpoints')

forecasting task

parser.add_argument('--seq_len', type=int, default=12, help='input sequence length')
parser.add_argument('--label_len', type=int, default=0, help='start token length')
parser.add_argument('--pred_len', type=int, default=12, help='prediction sequence length')

how can I obtain the result as shown in the paper? Thanks a lot if the experimental setting of each dataset can be provided.

[abaklykov]

Did you manage to launch the project?
I can't even start it.

[cccccj-03]

Did you manage to launch the project? I can't even start it.

I can train it by setting the hyper-parameters as my first comment, but it is quite different from the result in the paper. Some of the hyper-parameters are unclear.

[abaklykov]

I also managed to run the training on this dataset.

[lidexf]

I want to reproduce the experimental results, but it seems some hyper-parameters setting is not clear, the best performance of the Covid dataset is mae:0.134224995970726, rmse:0.1728876382112503, with a setting as below:

data loader

parser.add_argument('--data', type=str, default='Covid', help='dataset type') parser.add_argument('--root_path', type=str, default='./dataset/', help='root path of the data file') parser.add_argument('--data_path', type=str, default='covid.csv', help='data file') parser.add_argument('--channel_independence', type=int, default=0, help='1: channel dependence 0: channel independence')parser.add_argument('--features', type=str, default='MS', help='forecasting task, options:[M, S, MS]; M:multivariate predict multivariate, S:univariate predict univariate, MS:multivariate predict univariate') parser.add_argument('--target', type=str, default='OT', help='target feature in S or MS task') parser.add_argument('--freq', type=str, default='m', help='freq for time features encoding, options:[s:secondly, t:minutely, h:hourly, d:daily, b:business days, w:weekly, m:monthly], you can also use more detailed freq like 15min or 3h') parser.add_argument('--checkpoints', type=str, default='./checkpoints/', help='location of model checkpoints')

forecasting task

parser.add_argument('--seq_len', type=int, default=12, help='input sequence length') parser.add_argument('--label_len', type=int, default=0, help='start token length') parser.add_argument('--pred_len', type=int, default=12, help='prediction sequence length')

how can I obtain the result as shown in the paper? Thanks a lot if the experimental setting of each dataset can be provided.

I guess maybe you could get the parameter setting from other conparison mdel code such as FEDformer.