- `SubscriptionSynchronizer` will emit a `TimeSlice.TimePulse` before
performing any universe selection on each time loop. This will advance
`Algorithm.Time` which will allow universe selection data time and
`Algorithm.Time` to be aligned.
- Updating Regression algorithms that were using `algorithm.Time` in the
selection method.
- Coarse selection will start from the algorithms start date (not in the
next day)
- Adding regression algorithm
This flag indicates whether or not the local regression test system,
via RegressionTests.AlgorithmStatisticsRegression should run a given
IRegressionAlgorithmDefinition
It's important that we keep the factor files consistent with respect to
the date that they were generated. This enables us to run the regression
algorithms in the cloud and get the same results by using the factor files
from the correct date.
A mechanical refactoring was performed to make algorithms currently used in
regression algorithms to implement IRegressionAlgorithmDefinition, which allows
algorithms to define their own expected statistics and what languages should be
run as part of regression. The type name of the C# type is used to determine the
file/model name for python. This was for simplicity, but if needed, could later be
refactored to expose more information, but for now the convention of keeping names
the same makes sense and just works easily.
If we pull data and on the same time step that security gets removed,
we can still get that data in OnData(Slice) even though it was removed.
This change filters out removed securities by tracking a reference to
the subscription's disposed flag. Another change was made to wait until
the end of the time step to dispose of subscriptions.
Add IDataFeed.GetSubscription(SubscriptionDataConfig)
The fine enumerator factor accepts a 'date'. This date is that day
that the data is produced, so the emit time on that data is midnight
the following day. When we go to read the fine data we can't use the
current time because that's the 'emit time' -- instead we need to back
the time up a full day to find the correct start time. In addition,
the fine data was being read using UTC time stamps which led to even
more confusion here -- this change resolves both issues.
In addition to the above, the security.Fundamentals property wasn't being
properly set for the first time step when a security is selected due to
a bug in the subscription synchronizer not clearing out the universeData
in preparation for another loop (in the event changes != None).
Martin Molinero