Globals /
- AbstractAgent
- Academy
- ConvolutionalNeuralNetwork
- DQAgent
- DataSource
- LearningDataLogger
- Memory
- Model
- NeuralNetwork
- QAction
- QAgent
- QMatrix
- QState
- QTransition
- Result
- Teacher
- TypedWindow
- AcademyConfig
- AcademyStepInput
- AgentConfig
- AgentTrackingInformation
- BuildAgentConfig
- ConvolutionalLayer
- ConvolutionalNetworkLayer
- DQAgentConfig
- DenseLayer
- DropoutLayer
- FlattenLayer
- GraphEdge
- GraphNode
- Layer
- LayerConfig
- MaxPooling2DLayer
- Memento
- MementoTensor
- MemoryConfig
- NeuralNetworkLayer
- QActionData
- QAgentConfig
- QStateData
- TeacherTrackingInformation
- TeachingConfig
- ToTfLayerConfig
- DEFAULT_ACADEMY_CONFIG
- DEFAULT_AGENT_CONFIG
- DEFAULT_LAYER_CONFIG
- DEFAULT_MODEL_FIT_CONFIG
- DEFAULT_QAGENT_CONFIG
- DEFAULT_TEACHING_CONFIG
● HIST_WINDOW_MIN_SIZE: 0
= 0
Defined in reimprove/algorithms/deepq/dqagent.ts:11
● HIST_WINDOW_SIZE: 100
= 100
Defined in reimprove/algorithms/deepq/dqagent.ts:10
● MEM_WINDOW_MIN_SIZE: 2
= 2
Defined in reimprove/algorithms/deepq/dqagent.ts:9
▸ getColor(value
: number): string
Defined in reimprove/algorithms/q/qmatrix.ts:205
Parameters:
Name | Type |
---|---|
value |
number |
Returns: string
Defined in reimprove/academy.ts:8
● agentsLogs: false = false
Defined in reimprove/academy.ts:10
● memoryLogs: false = false
Defined in reimprove/academy.ts:11
● parentLogsElement: null = null
Defined in reimprove/academy.ts:9
Defined in reimprove/algorithms/abstract_agent.ts:5
Defined in reimprove/algorithms/deepq/dqagent.ts:13
● batchSize: number = 32
Defined in reimprove/algorithms/deepq/dqagent.ts:15
● memorySize: number = 30000
Defined in reimprove/algorithms/abstract_agent.ts:6
Defined in reimprove/algorithms/deepq/dqagent.ts:14
● temporalWindow: number = 1
Defined in reimprove/algorithms/deepq/dqagent.ts:16
Defined in reimprove/model.ts:31
● activation: string = "relu"
Defined in reimprove/model.ts:33
● units: number = 32
Defined in reimprove/model.ts:32
● useBias: false = false
Defined in reimprove/model.ts:34
Defined in reimprove/model.ts:7
● epochs: number = 10
Defined in reimprove/model.ts:8
● stepsPerEpoch: number = 200
Defined in reimprove/model.ts:9
Defined in reimprove/algorithms/q/qagent.ts:7
● createMatrixDynamically: false = false
Defined in reimprove/algorithms/q/qagent.ts:8
● dataHash: null = null
Defined in reimprove/algorithms/q/qagent.ts:9
● gamma: number = 0.9
Defined in reimprove/algorithms/q/qagent.ts:10
Defined in reimprove/teacher.ts:5
● alpha: number = 1
Defined in reimprove/teacher.ts:13
● epsilon: number = 1
Defined in reimprove/teacher.ts:9
● epsilonDecay: number = 0.95
Defined in reimprove/teacher.ts:11
● epsilonMin: number = 0.05
Defined in reimprove/teacher.ts:10
● gamma: number = 0.9
Defined in reimprove/teacher.ts:12
● lessonLength: number = 1000
Defined in reimprove/teacher.ts:6
● lessonsQuantity: number = 30
Defined in reimprove/teacher.ts:7
● lessonsWithRandom: number = 2
Defined in reimprove/teacher.ts:8