Using the Python code

In this section, we are going to use the Python scripts that we have created.

Running apprMPdist.py using the two time series with 10,000 elements each that we created earlier in this chapter generates the following kind of output:

$ ./apprMPdist.py 10k1.gz 10k2.gz -s 3 -c 64 -t 500 -w 120
Max Cardinality: 64 Segments: 3 Sliding Window: 120 Threshold: 500 Default Promotion: False
MPdist: 351.27 seconds
Approximate MPdist: 12.603

Using a bigger sliding window size generates the following output:

$ ./apprMPdist.py 10k1.gz 10k2.gz -s 3 -c 64 -t 500 -w 300
Max Cardinality: 64 Segments: 3 Sliding Window: 300 Threshold: 500 Default Promotion: False
MPdist: 384.74 seconds
Approximate MPdist: 21.757

So, bigger sliding window sizes require more time. As before, this is because calculating Euclidean distances for bigger sliding window sizes is slower.

Executing joinMPdist.py produces the following output:

$ ./joinMPdist.py 10k1.gz 10k2.gz -s 3 -c 64...