laiguokun / multivariate-time-series-data Goto Github PK

What do the 8 columns in exchange_rate.txt represent?
Where did this data come from?
Thank you for your response.

我去看了一下原始的数据集,比对了一下数据.发现时间轴确实是从12年开始,在这个界面下载的数据时间轴有误,起始应该为2012/01/01 02:00

I went and looked at the original data set and compared the data. It is found that the timeline does start from 2012. The data timeline downloaded on this page is wrong, and it should start from 2012/01/01 02:00

Hi! Can you give details about which is the format of the files? How to modify a file to replicate that extension .gz?

Traceback (most recent call last):
File "main.py", line 150, in
val_loss, val_rae, val_corr = evaluate(Data, Data.valid[0], Data.valid[1], model, evaluateL2, evaluateL1, args.batch_size);
File "main.py", line 24, in evaluate
if not predict:
File "/usr/lib64/python2.7/site-packages/torch/autograd/variable.py", line 125, in bool
torch.typename(self.data) + " is ambiguous")
how to deal it

Many thanks for the pre-processed data, really save the day.

Be careful that one year only contains 12 months, so it should be 24 months (rows), though I hope we can have more time to spend in our life.

从2011年到2014年，每15分钟记录一次用电量(以千瓦时为单位)。因为有些维度等于0。因此，我们在2011年消除了这些记录。
最终我们得到的数据包含了321个客户从2012年到2014年的用电情况。我们对数据进行了转换，以反映每小时的消费情况。 我想请问下这个数据集的日期开始是从2011年1月1日0时0分开始的吗

hello!

The electricity dataset as pre-processed by the lstnet paper drops all time series which are zero in the window of 2011 expect for some reason the time series index 118... making the multivariate dim of the dataset 320 if the index 118 is removed, rather than 321.