Skip to main content
. 2019 Aug 5;19(15):3424. doi: 10.3390/s19153424

Table 1.

Structure of training datasets for period t.

Dataset → 1 2 ds
Input data
Turnstile 1–Period t − 1 c 1 1,t−1 c 2 1,t−1 cds 1,t−1
Turnstile 2–Period t − 1 c 1 2,t−1 c 2 2,t−1 cds 2,t−1
Turnstile ts–Period t − 1 c 1 ts ,t−1 c 2 ts ,t−1 cdsts ,t−1
Turnstile 1–Period t − 2 c 1 1,t−2 c 2 1,t−2 cds 1,t−2
Turnstile 2–Period t − 2 c 1 2,t−2 c 2 2,t−2 cds 2,t−2
Turnstile ts–Period t − 2 c 1 ts ,t−2 c 2 ts ,t−2 cdsts ,t−2
Turnstile 1–Period t − 3 c 1 1,t−3 c 2 1,t−2 cds 1,t−2
Turnstile 2–Period t − 3 c 1 2,t−2 c 2 2,t−3 cds 2,t−3
Turnstile ts–Period t − 3 c 1 ts ,t−3 c 2 ts ,t−3 cdsts ,t−3
Turnstile 1–Period t − 4 c 1 1,t−4 c 2 1,t−4 cds 1,t−4
Turnstile 2–Period t − 4 c 1 2,t−4 c 2 2,t−4 cds 2,t−4
Turnstile ts–Period t − 4 c 1 ts ,t−4 c 2 ts ,t−4 cdsts ,t−4
Output data
Railway section 1–Period t f 1 1,t f 2 1,t fds 1,t
Railway section 2–Period t f 1 2,t f 2 2,t fds 2,t
Railway section rs–Period t f 1 rs ,t f 2 rs ,t fdsrs ,t