> ## Documentation Index > Fetch the complete documentation index at: https://mintlify.com/SeanCole02/doom-neuron/llms.txt > Use this file to discover all available pages before exploring further. # PPOConfig Reference > Complete configuration dataclass for PPO training and neural interface parameters in DOOM Neuron # PPOConfig The `PPOConfig` dataclass defines all configuration parameters for PPO reinforcement learning training and the CL1 neural hardware interface. ## Environment Configuration Path to the VizDoom configuration file that defines the game scenario Screen resolution for the DOOM game buffer. Valid values are VizDoom resolution constants like `RES_320X240`, `RES_640X480` Whether to enable the screen buffer for visual observations Maximum absolute degrees for TURN\_LEFT\_RIGHT\_DELTA action Discrete turn step size in degrees when using turn buttons Initial standard deviation (in degrees) for camera delta distribution Toggle for single categorical action space vs. combinatorial action space ## Neural Interface - Channel Configuration Total number of channels available on the CL1 hardware Channel indices used for encoding game state into stimulation patterns Channel indices that decode to forward movement actions Channel indices that decode to backward movement actions Channel indices that decode to strafe left actions Channel indices that decode to strafe right actions Channel indices that decode to turn left actions Channel indices that decode to turn right actions Channel indices that decode to attack/fire actions ## Stimulation Design Parameters Parameters used to construct `cl.StimDesign` objects for biphasic electrical stimulation. Duration of the first (negative) phase in microseconds (μs) Duration of the second (positive) phase in microseconds (μs) Minimum stimulation amplitude in microamps (μA). Used as the magnitude for phase1 (negative) Maximum stimulation amplitude in microamps (μA). Used as the magnitude for phase2 (positive) ## Burst Design Parameters Parameters used to construct `cl.BurstDesign` objects that define stimulation frequency and duration. Minimum burst frequency in Hertz (Hz) Maximum burst frequency in Hertz (Hz) Number of pulses per burst. Set to 500 to ensure stimulation lasts between game ticks ## PPO Hyperparameters Learning rate for the Adam optimizer Discount factor for future rewards Lambda parameter for Generalized Advantage Estimation (GAE) Clipping parameter for PPO policy updates Coefficient for value function loss in the total loss Coefficient for entropy bonus to encourage exploration Maximum gradient norm for gradient clipping. Can be reduced to 1 or 0.5 for more conservative updates Whether to normalize returns for critic training. Stabilizes the critic but may affect learning dynamics ## Training Configuration Number of parallel environments to run Number of environment steps to collect per policy update (per environment) Minibatch size for PPO updates Number of epochs to train on each batch of collected experience Maximum number of episodes to train for Whether to use CL1 hardware or run in simulation mode ## Network Architecture Hidden layer size for encoder, decoder, and value networks ## Logging and Checkpointing Directory for TensorBoard logs Directory for saving model checkpoints Save checkpoint every N episodes Evaluate policy every N episodes ## Reward Shaping Reward threshold above which positive feedback is triggered. Requires tuning Reward threshold below which negative feedback is triggered. Requires tuning Terminal reward bonus for armor-related objectives. Requires tuning Gain multiplier for aim alignment reward shaping Maximum distance at which aim alignment reward is computed Bonus reward for accurate aim alignment Angle threshold in degrees for aim alignment bonus Scaling factor for movement velocity rewards Use simplified reward function. When false, factors in manually shaped aim alignment and velocity rewards ## Feedback Stimulation - Step-level Rewards Stimulation amplitude (μA) for positive step-level feedback Stimulation frequency (Hz) for positive step-level feedback Number of pulses for positive step-level feedback Stimulation amplitude (μA) for negative step-level feedback Stimulation frequency (Hz) for negative step-level feedback Number of pulses for negative step-level feedback ## Feedback Stimulation - Episode-level Rewards Number of pulses for positive episode-level feedback Stimulation frequency (Hz) for positive episode-level feedback Number of pulses for negative episode-level feedback Stimulation frequency (Hz) for negative episode-level feedback Whether to provide feedback only at episode end (disables step-level feedback) ## Reward Feedback Channels Whether to enable reward-based feedback stimulation Channel indices for positive reward feedback Channel indices for negative reward feedback ## Event-based Feedback Distance threshold for triggering movement-based events Dictionary mapping event names to EventFeedbackConfig objects. Contains configurations for: * `enemy_kill`: Feedback when killing an enemy * `armor_pickup`: Feedback when collecting armor * `took_damage`: Feedback when taking damage * `ammo_waste`: Feedback when wasting ammo * `approach_target`: Feedback when moving closer to target * `retreat_target`: Feedback when moving away from target See [EventFeedbackConfig](/api/event-feedback) for detailed field documentation. ## Decoder Configuration Whether to enforce non-negative weights in decoder linear readout heads. Experimental - requires testing Whether to freeze decoder weights during training. Experimental - requires testing Whether to zero out decoder biases. Recommended to be true - bias can cause decoder to generate its own predictions Whether to use MLP decoder instead of linear readout. Prefer false - MLP can learn to play the game instead of relying on neural responses Hidden layer size for MLP decoder. Only used if `decoder_use_mlp` is true. Experimental value - requires testing L2 regularization coefficient for decoder weights. Untuned L2 regularization coefficient for decoder biases. Untuned Ablation mode for testing decoder behavior. Valid values: * `none`: Normal operation * `random`: Replace spike features with random values * `zero`: Replace spike features with zeros ## Wall Detection Number of raycasts for wall detection. Probably less necessary with CNN encoder Maximum range for wall detection raycasts. Keep as is Maximum depth distance for wall detection normalization. Already calibrated - keep as is ## Encoder Configuration Whether encoder network weights are trainable. Recommended to be true for reasonable PPO policy gradients, especially with `decoder_use_mlp: false` Entropy penalty coefficient for the encoder (uses Beta distribution sampling) Whether to use CNN for visual feature extraction. Testing shows CNN does not overfit/learn on its own - useful to keep true Number of base channels for CNN encoder. Arbitrary value - can be adjusted Downsampling factor for CNN input. Arbitrary value - can be adjusted ## Episode Feedback Events Specific event name to use for positive episode feedback. If None, uses overall episode reward Specific event name to use for negative episode feedback. If None, uses overall episode reward ## Surprise-based Feedback Scaling Gain for surprise-based feedback scaling. Tune as needed based on neuron responses Maximum scaling factor for surprise-based feedback. Tune as needed based on neuron responses Frequency gain for surprise-based feedback. Tune as needed based on neuron responses Amplitude gain for surprise-based feedback. Tune as needed based on neuron responses Maximum frequency scale for surprise-based feedback. Tune as needed based on neuron responses Maximum amplitude scale for surprise-based feedback. Tune as needed based on neuron responses ## Distance Normalization Normalization constant for enemy distance features. Already calibrated - do not change ## Usage Example ```python theme={null} from ppo_doom import PPOConfig # Create default configuration config = PPOConfig() # Override specific parameters config = PPOConfig( learning_rate=1e-4, max_episodes=5000, encoder_use_cnn=True, decoder_use_mlp=False, use_hardware=True ) # Access channel configuration print(config.encoding_channels) # [8, 9, 10, 17, 18, 25, 27, 28, 57] print(config.attack_channels) # [32, 33, 34] # Access event feedback settings kill_feedback = config.event_feedback_settings['enemy_kill'] print(kill_feedback.base_frequency) # 20.0 ```