public void TestEpsilonGreedy()
{
var epsilonGreedyTests = new EpsilonGreedyTestConfiguration[]
{
// No exploration
new EpsilonGreedyTestConfiguration
{
AppId = TestContext.TestName + "NoExplorationFixedActionContext",
ContextType = ContextType.FixedAction, // test fixed-action context
Epsilon = 0f,
NumberOfActions = 20,
ExperimentalUnitIdList = Enumerable.Range(1, 100).Select(i => i.ToString()).ToList(),
PolicyConfiguration = new FixedPolicyConfiguration { Action = 10 }
},
new EpsilonGreedyTestConfiguration
{
AppId = TestContext.TestName + "NoExplorationVariableActionContext",
ContextType = ContextType.VariableAction, // test variable-action context
Epsilon = 0f,
NumberOfActions = 20,
ExperimentalUnitIdList = Enumerable.Range(1, 100).Select(i => i.ToString()).ToList(),
PolicyConfiguration = new FixedPolicyConfiguration { Action = 10 }
},
// Regular exploration
new EpsilonGreedyTestConfiguration
{
AppId = TestContext.TestName + "RegularExplorationFixedActionContext",
ContextType = ContextType.FixedAction,
Epsilon = 0.2f,
NumberOfActions = 10,
ExperimentalUnitIdList = Enumerable.Range(1, 100).Select(i => i.ToString()).ToList(),
PolicyConfiguration = new FixedPolicyConfiguration { Action = 9 }
},
new EpsilonGreedyTestConfiguration
{
AppId = TestContext.TestName + "RegularExplorationVariableActionContext",
ContextType = ContextType.VariableAction,
Epsilon = 0.2f,
NumberOfActions = 10,
ExperimentalUnitIdList = Enumerable.Range(1, 100).Select(i => i.ToString()).ToList(),
PolicyConfiguration = new FixedPolicyConfiguration { Action = 9 }
},
// Heavy exploration
new EpsilonGreedyTestConfiguration
{
AppId = TestContext.TestName + "HeavyExplorationFixedActionContext",
ContextType = ContextType.FixedAction,
Epsilon = 0.9f,
NumberOfActions = 90, // test many actions
ExperimentalUnitIdList = Enumerable.Range(1, 100).Select(i => i.ToString()).ToList(),
PolicyConfiguration = new FixedPolicyConfiguration { Action = 81 }
},
new EpsilonGreedyTestConfiguration
{
AppId = TestContext.TestName + "HeavyExplorationVariableActionContext",
ContextType = ContextType.VariableAction,
Epsilon = 0.9f,
NumberOfActions = 90,
ExperimentalUnitIdList = Enumerable.Range(1, 100).Select(i => i.ToString()).ToList(),
PolicyConfiguration = new FixedPolicyConfiguration { Action = 81 }
}
};
Run(outputFilePatternExpected, outputFilePatternActual, outputJsonConfigFile, epsilonGreedyTests);
for (uint i = 0; i < epsilonGreedyTests.Length; i++)
{
CompareExplorationData(FormatPath(outputFilePatternExpected, i), FormatPath(outputFilePatternActual, i));
}
}
public void TestEpsilonGreedy()
{
var epsilonGreedyTests = new EpsilonGreedyTestConfiguration[]
{
// No exploration
new EpsilonGreedyTestConfiguration
{
AppId = TestContext.TestName + "NoExplorationFixedActionContext",
ContextType = ContextType.FixedAction, // test fixed-action context
Epsilon = 0f,
NumberOfActions = 20,
ExperimentalUnitIdList = Enumerable.Range(1, 100).Select(i => i.ToString()).ToList(),
PolicyConfiguration = new FixedPolicyConfiguration {
Action = 10
}
},
new EpsilonGreedyTestConfiguration
{
AppId = TestContext.TestName + "NoExplorationVariableActionContext",
ContextType = ContextType.VariableAction, // test variable-action context
Epsilon = 0f,
NumberOfActions = 20,
ExperimentalUnitIdList = Enumerable.Range(1, 100).Select(i => i.ToString()).ToList(),
PolicyConfiguration = new FixedPolicyConfiguration {
Action = 10
}
},
// Regular exploration
new EpsilonGreedyTestConfiguration
{
AppId = TestContext.TestName + "RegularExplorationFixedActionContext",
ContextType = ContextType.FixedAction,
Epsilon = 0.2f,
NumberOfActions = 10,
ExperimentalUnitIdList = Enumerable.Range(1, 100).Select(i => i.ToString()).ToList(),
PolicyConfiguration = new FixedPolicyConfiguration {
Action = 9
}
},
new EpsilonGreedyTestConfiguration
{
AppId = TestContext.TestName + "RegularExplorationVariableActionContext",
ContextType = ContextType.VariableAction,
Epsilon = 0.2f,
NumberOfActions = 10,
ExperimentalUnitIdList = Enumerable.Range(1, 100).Select(i => i.ToString()).ToList(),
PolicyConfiguration = new FixedPolicyConfiguration {
Action = 9
}
},
// Heavy exploration
new EpsilonGreedyTestConfiguration
{
AppId = TestContext.TestName + "HeavyExplorationFixedActionContext",
ContextType = ContextType.FixedAction,
Epsilon = 0.9f,
NumberOfActions = 90, // test many actions
ExperimentalUnitIdList = Enumerable.Range(1, 100).Select(i => i.ToString()).ToList(),
PolicyConfiguration = new FixedPolicyConfiguration {
Action = 81
}
},
new EpsilonGreedyTestConfiguration
{
AppId = TestContext.TestName + "HeavyExplorationVariableActionContext",
ContextType = ContextType.VariableAction,
Epsilon = 0.9f,
NumberOfActions = 90,
ExperimentalUnitIdList = Enumerable.Range(1, 100).Select(i => i.ToString()).ToList(),
PolicyConfiguration = new FixedPolicyConfiguration {
Action = 81
}
}
};
Run(outputFilePatternExpected, outputFilePatternActual, outputJsonConfigFile, epsilonGreedyTests);
for (uint i = 0; i < epsilonGreedyTests.Length; i++)
{
CompareExplorationData(FormatPath(outputFilePatternExpected, i), FormatPath(outputFilePatternActual, i));
}
}
