In RandomizedPcaTrainer:-Is it essential that Each data generate on same time interval?

**Thanks for your reply.
I have one more query/issue.**
In my Application
I have density data around 57000 rows
Data set column: -Density, Hour, Minute, Second
Requirement: - find out inline /out liner density abnormal data
so, I have used "RandomizedPcaTrainer" with option parameters.
input column name "Density"
Output column name "Label"
Rank=2
SEED=28500
Feature column list: - Hour, Minute, Second
**Is it essential that Each Density data generate on same time interval?**
Example: -
Density Hour Minute Second
1) 777.8 12 10 10
2) 786.5 12 10 20
3) 786.7 12 10 30
So same way each 10 second gap data will generate.
**I have tried feature column use with single data type then it is working.
If I am using feature column use with double data type, then it will generate exception.
So why it is generating exception?**

_Originally posted by @jigarhshah17 in https://github.com/dotnet/machinelearning-samples/issues/698#issuecomment-544962800_