Build & Test Lean / build (push) Has been cancelled
Regression Tests / build (push) Has been cancelled
Research Regression Tests / build (push) Has been cancelled
* Live T-1 selection warmup
- Only use BaseDataCollection class
- Reuse collection enumerator
- Remove FuturesChainUniverseSubscriptionEnumeratorFactory
- Universe selection will use Cache providers
- Add null reference check
- Add more tests
- Fixes for warmup fill forwarding. Add more tests
* Address reviews. Add missing comments
* Refactor chain providers
- ChainProviders will use quote, open interest and trade information,
which ever is first. Updating regression algorithm changing contract
being added.
- Add base BacktestingChainProvider for shared logic, reducing
duplication.
- ChainProviders will now use the given IDataCacheProvider and
IMapFileProvider
- If Providers are called for a date the exchange is closed they will
search for the previous available date where the exchange is open.
Adding unit tests.
* Address reviews. Logging improvements
Regression Tests / build (push) Has been cancelled
Build & Test Lean / build (push) Has been cancelled
* Remove user plan enum
- Get maximum order and runtime from job controls
- Remove user plan enum definitions
* Add log for BacktestingSetupHandler maximums
Regression Tests / build (push) Has been cancelled
Build & Test Lean / build (push) Has been cancelled
* Improve python exception parsing
- Improve python exception parsing adding support for line shift. Adding
unit tests
* PythonException revert change
* Centralized and normalize algorithm runtime handling
* Adding support for C# line and file exception report
Build & Test Lean / build (push) Has been cancelled
* Use lean data key as param for request
* key -> filePath rename and some cleanup
* Refactor
* Add Organizations Endpoints
* Add some organization api wrapper objects
* Address namespace issue
* Reorganize Api Test into seperate files using one ApiTestBase
* Add Organization tests
* Use capitalized "API" test namespace to reduce amount of file changes
* Add License to test base
* Update /data endpoint functions and response objects
* Update ApiDataProvider Logic
* Handle deserialization of organization products
* Simplify converter
* Only throw for equity requests when not subscribed to map/factor files
* Add missing header
* Make arguement exception
* Api adjustments
* Add Zip factor and map file providers
- Common project will now reference Compression project and not the other way
round.
- Adding Zip FactorFile and MapFile providers
* Refactor FactorFileProvider to use DataProvider to fetch files
* Use resulting MinimumDate in construction of FactorFile
* Nit FactorFile comments and arrangement
* Refactor MapFileProviders to use DataProvider for fetching files
* Refactor ZipFileProvider
* Clean up
* Refactor Backtesting Future/Option chain providers to use dataprovider
* Fixes for data/ endpoints and test adjustments
* Response objects adjustments/cleanups
* ApiDateProvider fixes and testing
* Add LocalZipFactorFileTests
* Update ApiDataProvider download test to verify stream is not null
* Implement posting of agreement summary and signed time
* Mark all Api related tests as explicit and document details on running
* Clarify default token on ApiTestBase
* Adjust summary
* Update Api responses for QCC, except org products which are sold in USD
* Implement cache expiration for zip MapFile and FactorFiles. Adding unit tests
* Fix multiple markets for ZipFactorFile provider
* Use Symbol as cache key
* Api.cs review
* Dispose of factorFileStream after reading
* Use zip.EntryFileNames
* Address a few reviews
* Few more fixes
* Address Api Review
* Add Job Org id to config
* Minor tweaks
* Compare with invariant culture
* Fixes Option Universe selection
* ZipEntryNameSubscriptionDataSourceReader will use IDataProvider
* Fix research
* Fix null reference exception
* Make duplicate log debug
Co-authored-by: Martin-Molinero <martin@quantconnect.com>
Build & Test Lean / build (push) Has been cancelled
* Reconcile and consolidate SetupHandlers
* Centralize GetConfiguredDataFeeds()
* nit - extra space
* Check for null, also allow null to be passed back if no config value
* Fix breaking test
* Cleanup
* fix return var
* remove unneeded if/else
* Minor changes
Co-authored-by: Martin-Molinero <martin@quantconnect.com>
* Adds Static AlgorithmCreationTimeout Property to BaseSetupHandler
This value will be used to set maximum time that the creation of an algorithm can take.
* Adds Additional Logging to AlgorithmPythonWrapper
It will informing the user how long it takes to import the module.
* Changes How AlgorithmCreationTimeout is Initialized
It will prevent callling Config.GetDouble on every call.
- When BacktestNodePacket has the inital `CashAmount` set we will clear
all existing cash amounts and set the account currency
- Adding more unit tests
This commit is squashed from iterative development:
- More consistent method naming
- Storage root path updated to be absolute and include algorithm name
- Storage root path created only if object store is actually used
- Implemented XML save/load
- Added missing unit tests
- Replaced Log.Trace with Log.Error calls
- Added the object store name logging in Engine.Main
- Read storage root from config
- Create algorithm storage root folder in Initialize
- Remove empty folder in Dispose
- Added null checks in all methods
- Added missing XML parameter docs
- make Initialize and Dispose virtual
- make AlgorithmStorageRoot protected
The IObjectStore abstraction provides algorithms with a persistent
storage mechanism. While the algorithm is running, data is maintained
in memory as a dictionary of raw bytes (string -> byte[]). This ensures
we avoid any reference type shenanigans. Periodically, the data in the
object store is persisted and additionally, when the algorithm shuts
down, the object store's data will again be persisted. This ensures that
when the algorithm starts up again, it will have access to any state
that has been saved into the object store.
A great use case for IObjectStore is saving a compute heavy model.
For example, computing the weights of a deep neural network is very
CPU intensive, but after the weights are computed, evaluation is fairly
quick. An initial backtest can be used to solved for the network's weights
and then subsequent backtests or even in live mode, the weights will be
available to the algorithm provided they were saved into the object store.
Also, some libraries require a file path to load model data. The object
store provides a `GetFilePath(key)` method which will copy the data for
the provided key to the disk and return that path so the library can load
the model data.
- Adding `FreePortfolioValue` to be set after algorithm initialize based
on the `TotalPortfolioValue` and the `FreePortfolioValuePercentage`
- Updating regression tests
- Adding new regression test
- Adding check for minimum order value at `BuyingPowerModel`
- Adding `WorkerThread` class, wrapper for a worker thread that will
execute given `Actions`.
- Algorithm related code (`Construction`, `Initialization`,
`Execution` will be executed by the same `WorkerThread` instance,
this is required for `Python` debugging.
- Moving `UniverseSelection.EnsureCurrencyDataFeeds` call into the
`IResultHandler` implementation through usage of the new `SetupHandlerHelper`
class, that will also set an initial conversion rate if none present.
- Adding regression test, that reproduces original issue
- ISetupHandler implementations will now use 50ms as sleep interval
while initializing the algorithm
- If the sleep interval is >= to 1s (default value) it will divide the
sleep operations into 5
Improves the message when the Loader cannot resolve the algorithm to load. It happens when the assemblies don't have a QCAlgorithm class that match the algorithm name or you have 2-of them so Lean doesn't know which one to backtest.
The possible Loader exceptions are thrown as `AlgorithmSetupException` to mach the pattern for exceptions during initialization.
This is performed w/in a try/catch which esures that we won't call PostInitialize
if Initialize throws an error, thereby preventing potential confusing in the reported
error message
Fixes#1778
The engine defines securities for each universe to properly track them within
the data feed. These securities are not tradable and have no price data associated
with them, and as such, we should not be sending history requests for these symbols.
This change removes all universe symbols from history requests.
NOTE: Requests made directly to the history provider are not filtered out, as the
filtering happens within the QCAlgorithm implementation.
This PR is an attempt to reduce contention in concurrent dictionaries, replacing method calls using full locks with lock-free equivalents:
- dictionary.Count -> dictionary.Skip(0).Count()
- dictionary.Keys -> dictionary.Select(x => x.Key)
- dictionary.Values -> dictionary.Select(x => x.Value)
The most frequent usages of these methods are: CashBook, SecurityManager, UniverseManager and indirectly, SecurityPortfolioManager.
The reasons for this update are explained very clearly in this article:
https://arbel.net/2013/02/03/best-practices-for-using-concurrentdictionary/
- Move BacktestingFutureChainProvider provider to Lean.Engine.DataFeeds along with its options equivalent.
- EmptyFutureChainProvider: provider that returns an empty list of symbols
- CachingFutureChainProvider: implements caching by date
- BacktestingFutureChainProvider: provider that gets chain from local files
- LiveFutureChainProvider: provider that gets chain from external source (empty list of symbols for now)
- Moved provider implementations out of brokerages into their own classes
- Removed DefaultOptionChainProvider
- Added BacktestingOptionChainProvider and LiveOptionChainProvider
- Moved SetOptionChainProvider call from Engine to setup handlers
Lean will not set the start and end date in the BacktestingSetupHandler.
The start and end dates are specified in the user code in
IAlgorithm.Initialize.
When using BrokerageHistoryProvider with InteractiveBrokers, GetOrderByBrokerageId calls on open orders were logging NullReferenceExceptions because SetOrderProcessor is called later, in BrokerageSetupHandler.Setup.
Martin-Molinero