search

liyaguang / DCRNN

Implementation of Diffusion Convolutional Recurrent Neural Network in Tensorflow
MIT License
1.21k stars 400 forks source link

Why is not an integer in the h5 file #58

Closed simplehx closed 3 years ago

simplehx commented 3 years ago

Thanks for your work.

In the readme file, the data is like this, all integers.

sensor_0 sensor_1 sensor_2 sensor_n
2018/01/01 00:00:00 60.0 65.0 70.0 ...
2018/01/01 00:05:00 61.0 64.0 65.0 ...
2018/01/01 00:10:00 63.0 65.0 60.0 ...
... ... ... ... ...

But when I read the h5 file, why the value is not an integer. such as 71.4 67.8 70.5 67.4 68.8. What does the decimal represent?

df1 = pd.read_hdf("pems-bay.h5")
print(df1.head())

2017-01-01 00:00:00 71.4 67.8 70.5 67.4 68.8 66.6 66.8 68.0 66.8 69.0 ... 68.8 67.9 68.8 68.0 69.2 68.9 70.4 68.8 71.1 68.0 2017-01-01 00:05:00 71.6 67.5 70.6 67.5 68.7 66.6 66.8 67.8 66.5 68.2 ... 68.4 67.3 68.4 67.6 70.4 68.8 70.1 68.4 70.8 67.4 2017-01-01 00:10:00 71.6 67.6 70.2 67.4 68.7 66.1 66.8 67.8 66.2 67.8 ... 68.4 67.4 68.4 67.5 70.2 68.3 69.8 68.4 70.5 67.9 2017-01-01 00:15:00 71.1 67.5 70.3 68.0 68.5 66.7 66.6 67.7 65.9 67.8 ... 68.5 67.5 68.5 67.5 70.4 68.7 70.2 68.4 70.8 67.6 2017-01-01 00:20:00 71.7 67.8 70.2 68.1 68.4 66.9 66.1 67.7 66.1 67.8 ... 68.5 67.7 68.5 67.4 69.6 69.1 70.0 68.4 71.0 67.9

liyaguang commented 3 years ago

Thanks for the question. The numbers in Readme is just for the purposes of illustration. The actual numbers represent the averaged speed (mile/h) from unevenly distributed observations with finer granularity.