Index

Symbols

**kwargs (rl4co.models.rl.reinforce.reinforce.REINFORCE parameter), [1] (rl4co.models.zoo.active_search.search.ActiveSearch parameter) (rl4co.models.zoo.am.model.AttentionModel parameter) (rl4co.models.zoo.am.policy.AttentionModelPolicy parameter) (rl4co.models.zoo.ham.model.HeterogeneousAttentionModel parameter) (rl4co.models.zoo.ham.policy.HeterogeneousAttentionModelPolicy parameter) (rl4co.models.zoo.mdam.model.MDAM parameter) (rl4co.models.zoo.pomo.model.POMO parameter) (rl4co.models.zoo.pomo.policy.POMOPolicy parameter) (rl4co.models.zoo.ppo.policy.PPOPolicy parameter) (rl4co.models.zoo.ptrnet.model.PointerNetwork parameter) (rl4co.models.zoo.symnco.model.SymNCO parameter) (rl4co.models.zoo.symnco.policy.SymNCOPolicy parameter) (rl4co.tasks.eval.EvalBase parameter)

**unused_kw (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy parameter)

_

__call__() (rl4co.data.transforms.StateAugmentation method) (rl4co.tasks.eval.EvalBase method)

A

action (rl4co.models.zoo.ppo.decoder.PPODecoder.evaluate_action parameter) (rl4co.models.zoo.ppo.policy.PPOPolicy.evaluate_action parameter) ActiveSearch (class in rl4co.models.zoo.active_search.search) alpha (rl4co.models.zoo.symnco.model.SymNCO parameter) ATSPEnv (class in rl4co.envs.atsp) attention_dropout (rl4co.models.nn.attention.MultiHeadAttention parameter) (rl4co.models.nn.flash_attention.CrossAttention parameter) (rl4co.models.nn.flash_attention.FlashCrossAttention parameter) (rl4co.models.nn.flash_attention.FlashSelfAttention parameter) (rl4co.models.nn.flash_attention.SelfAttention parameter)

AttentionModel (class in rl4co.models.zoo.am.model) AttentionModelPolicy (class in rl4co.models.zoo.am.policy) augment_dihedral (rl4co.models.zoo.active_search.search.ActiveSearch parameter) (rl4co.models.zoo.eas.search.EAS parameter) augment_size (rl4co.models.zoo.active_search.search.ActiveSearch parameter) (rl4co.models.zoo.eas.search.EAS parameter) AugmentationEval (class in rl4co.tasks.eval) AutoregressiveDecoder (class in rl4co.models.zoo.common.autoregressive.decoder) AutoregressivePolicy (class in rl4co.models.zoo.common.autoregressive.policy)

B

baseline (rl4co.models.rl.reinforce.baselines.WarmupBaseline parameter), [1] (rl4co.models.rl.reinforce.reinforce.REINFORCE parameter), [1] (rl4co.models.zoo.am.model.AttentionModel parameter) (rl4co.models.zoo.eas.search.EAS parameter) (rl4co.models.zoo.ham.model.HeterogeneousAttentionModel parameter) (rl4co.models.zoo.mdam.model.MDAM parameter) (rl4co.models.zoo.pomo.model.POMO parameter) (rl4co.models.zoo.ptrnet.model.PointerNetwork parameter) baseline_kwargs (rl4co.models.rl.reinforce.reinforce.REINFORCE parameter), [1] (rl4co.models.zoo.am.model.AttentionModel parameter) (rl4co.models.zoo.ham.model.HeterogeneousAttentionModel parameter) (rl4co.models.zoo.mdam.model.MDAM parameter) (rl4co.models.zoo.ptrnet.model.PointerNetwork parameter) batch (rl4co.models.rl.common.base.RL4COLitModule.test_step parameter) (rl4co.models.rl.common.base.RL4COLitModule.training_step parameter) (rl4co.models.rl.common.base.RL4COLitModule.validation_step parameter) (rl4co.models.rl.reinforce.reinforce.REINFORCE.calculate_loss parameter), [1]

batch_idx (rl4co.models.rl.common.base.RL4COLitModule.test_step parameter) (rl4co.models.rl.common.base.RL4COLitModule.training_step parameter) (rl4co.models.rl.common.base.RL4COLitModule.validation_step parameter) batch_locked (rl4co.envs.common.base.RL4COEnvBase attribute), [1] batch_size (rl4co.envs.ffsp.FFSPEnv parameter) (rl4co.models.rl.common.base.RL4COLitModule parameter) (rl4co.models.zoo.active_search.search.ActiveSearch parameter) (rl4co.models.zoo.eas.search.EAS parameter) beta (rl4co.models.rl.reinforce.baselines.ExponentialBaseline parameter), [1] (rl4co.models.zoo.symnco.model.SymNCO parameter) bias (rl4co.models.nn.attention.MultiHeadAttention parameter) bl_alpha (rl4co.models.rl.reinforce.baselines.RolloutBaseline parameter), [1]

C

calc_logits() (rl4co.models.zoo.ptrnet.decoder.Decoder method) calc_reward (rl4co.models.zoo.common.autoregressive.decoder.AutoregressiveDecoder.forward parameter) (rl4co.models.zoo.eas.decoder.forward_eas parameter) calculate_loss() (rl4co.models.rl.reinforce.reinforce.REINFORCE method), [1] capacity (rl4co.envs.cvrp.CVRPEnv parameter) (rl4co.envs.sdvrp.SDVRPEnv parameter) causal (rl4co.models.nn.attention.MultiHeadAttention parameter) (rl4co.models.nn.flash_attention.CrossAttention.forward parameter) (rl4co.models.nn.flash_attention.FlashCrossAttention.forward parameter) (rl4co.models.nn.flash_attention.FlashSelfAttention.forward parameter) (rl4co.models.nn.flash_attention.SelfAttention.forward parameter) change() (rl4co.models.zoo.mdam.encoder.GraphAttentionEncoder method) check_extension() (in module rl4co.data.utils) check_solution (rl4co.envs.common.base.RL4COEnvBase parameter), [1] (rl4co.envs.pctsp.PCTSPEnv parameter) check_solution_validity() (rl4co.envs.common.base.RL4COEnvBase method), [1] (rl4co.envs.cvrp.CVRPEnv static method) (rl4co.envs.op.OPEnv static method) (rl4co.envs.pctsp.PCTSPEnv static method) (rl4co.envs.sdvrp.SDVRPEnv static method) check_unused_kwargs() (in module rl4co.tasks.eval)

chip_file (rl4co.envs.dpp.DPPEnv parameter) config (rl4co.models.nn.env_embeddings.context.env_context_embedding parameter) (rl4co.models.nn.env_embeddings.dynamic.env_dynamic_embedding parameter) (rl4co.models.nn.env_embeddings.init.env_init_embedding parameter) configure_optimizers() (rl4co.models.rl.common.base.RL4COLitModule method) (rl4co.models.rl.ppo.ppo.PPO method), [1] context (rl4co.models.zoo.ptrnet.decoder.Decoder.forward parameter) context_embedding (rl4co.models.zoo.common.autoregressive.decoder.AutoregressiveDecoder parameter) (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy parameter) cost_type (rl4co.envs.mtsp.MTSPEnv parameter) critic (rl4co.models.rl.reinforce.baselines.CriticBaseline parameter), [1] (rl4co.models.zoo.ppo.model.PPOModel parameter) critic_kwargs (rl4co.models.zoo.ppo.model.PPOModel parameter) CriticBaseline (class in rl4co.models.rl.reinforce.baselines), [1] CriticNetwork (class in rl4co.models.rl.common.critic) CriticNetworkLSTM (class in rl4co.models.zoo.ptrnet.critic) CrossAttention (class in rl4co.models.nn.flash_attention) cu_seqlens (rl4co.models.nn.flash_attention.FlashCrossAttention.forward parameter) (rl4co.models.nn.flash_attention.FlashSelfAttention.forward parameter) (rl4co.models.nn.flash_attention.MHA.forward parameter) cu_seqlens_k (rl4co.models.nn.flash_attention.FlashCrossAttention.forward parameter) CVRPEnv (class in rl4co.envs.cvrp)

D

data_dir (rl4co.envs.common.base.RL4COEnvBase parameter), [1] (rl4co.envs.dpp.DPPEnv parameter) (rl4co.models.rl.common.base.RL4COLitModule parameter) dataloader_id (rl4co.models.rl.common.base.RL4COLitModule.test_step parameter) dataloader_idx (rl4co.models.rl.common.base.RL4COLitModule.validation_step parameter) dataloader_num_workers (rl4co.models.rl.common.base.RL4COLitModule parameter) dataset (rl4co.models.zoo.active_search.search.ActiveSearch parameter) (rl4co.models.zoo.eas.search.EAS parameter) dataset() (rl4co.envs.common.base.RL4COEnvBase method), [1] decap_file (rl4co.envs.dpp.DPPEnv parameter) decode_probs() (in module rl4co.models.nn.utils) decode_type (rl4co.models.zoo.common.autoregressive.decoder.AutoregressiveDecoder.forward parameter) (rl4co.models.zoo.eas.decoder.forward_eas parameter) Decoder (class in rl4co.models.zoo.mdam.decoder) (class in rl4co.models.zoo.ptrnet.decoder) decoder (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy parameter)

decoder_input (rl4co.models.zoo.ptrnet.decoder.Decoder.forward parameter) decoder_kwargs (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy.forward parameter) (rl4co.models.zoo.mdam.policy.MDAMPolicy.forward parameter) (rl4co.models.zoo.symnco.policy.SymNCOPolicy.forward parameter) device (rl4co.envs.atsp.ATSPEnv parameter) (rl4co.envs.common.base.RL4COEnvBase parameter), [1] (rl4co.envs.pdp.PDPEnv parameter) (rl4co.envs.tsp.TSPEnv parameter) (rl4co.models.nn.attention.MultiHeadAttention parameter) dihedral_8_augmentation() (in module rl4co.data.transforms) dihedral_8_augmentation_wrapper() (in module rl4co.data.transforms) DPPContext (class in rl4co.models.nn.env_embeddings.context) DPPEnv (class in rl4co.envs.dpp) DPPInitEmbedding (class in rl4co.models.nn.env_embeddings.init) dtype (rl4co.models.nn.attention.MultiHeadAttention parameter) dynamic_embedding (rl4co.models.zoo.common.autoregressive.decoder.AutoregressiveDecoder parameter) (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy parameter)

E

EAS (class in rl4co.models.zoo.eas.search) eas_emb_cache_keys (rl4co.models.zoo.eas.search.EAS parameter) eas_lambda (rl4co.models.zoo.eas.search.EAS parameter) EASEmb (class in rl4co.models.zoo.eas.search) EASLay (class in rl4co.models.zoo.eas.search) EASLayerNet (class in rl4co.models.zoo.eas.nn) edge_update() (rl4co.models.nn.graph.mpnn.MessagePassingEncoder method) (rl4co.models.nn.graph.mpnn.MessagePassingLayer method) emb_dim (rl4co.models.zoo.eas.nn.EASLayerNet parameter) embed_dim (rl4co.models.nn.attention.LogitAttention parameter) (rl4co.models.nn.attention.MultiHeadAttention parameter) (rl4co.models.nn.graph.attnnet.MultiHeadAttentionLayer parameter) embedded_inputs (rl4co.models.zoo.ptrnet.decoder.Decoder.forward parameter) embedding_dim (rl4co.models.nn.graph.attnnet.GraphAttentionNetwork parameter) (rl4co.models.nn.graph.gcn.GCNEncoder parameter) (rl4co.models.rl.common.critic.CriticNetwork parameter) (rl4co.models.zoo.am.policy.AttentionModelPolicy parameter) (rl4co.models.zoo.common.autoregressive.decoder.AutoregressiveDecoder parameter) (rl4co.models.zoo.common.autoregressive.encoder.GraphAttentionEncoder parameter) (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy parameter) (rl4co.models.zoo.ham.policy.HeterogeneousAttentionModelPolicy parameter) (rl4co.models.zoo.pomo.policy.POMOPolicy parameter) (rl4co.models.zoo.ppo.policy.PPOPolicy parameter) (rl4co.models.zoo.symnco.policy.SymNCOPolicy parameter) embeddings (rl4co.models.zoo.common.autoregressive.decoder.AutoregressiveDecoder.forward parameter) (rl4co.models.zoo.eas.decoder.forward_eas parameter) (rl4co.models.zoo.ppo.decoder.PPODecoder.evaluate_action parameter) Encoder (class in rl4co.models.zoo.ptrnet.encoder) encoder (rl4co.models.rl.common.critic.CriticNetwork parameter) (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy parameter) (rl4co.models.zoo.ham.policy.HeterogeneousAttentionModelPolicy parameter) env (rl4co.models.nn.env_embeddings.context.env_context_embedding parameter) (rl4co.models.nn.env_embeddings.dynamic.env_dynamic_embedding parameter) (rl4co.models.nn.env_embeddings.init.env_init_embedding parameter) (rl4co.models.rl.common.base.RL4COLitModule parameter) (rl4co.models.rl.reinforce.reinforce.REINFORCE parameter), [1] (rl4co.models.zoo.active_search.search.ActiveSearch parameter) (rl4co.models.zoo.am.model.AttentionModel parameter) (rl4co.models.zoo.common.autoregressive.decoder.AutoregressiveDecoder.forward parameter) (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy.forward parameter) (rl4co.models.zoo.eas.decoder.forward_eas parameter) (rl4co.models.zoo.eas.search.EAS parameter) (rl4co.models.zoo.ham.model.HeterogeneousAttentionModel parameter) (rl4co.models.zoo.mdam.model.MDAM parameter) (rl4co.models.zoo.mdam.policy.MDAMPolicy.forward parameter) (rl4co.models.zoo.pomo.model.POMO parameter) (rl4co.models.zoo.ppo.decoder.PPODecoder.evaluate_action parameter) (rl4co.models.zoo.ppo.model.PPOModel parameter) (rl4co.models.zoo.ppo.policy.PPOPolicy.evaluate_action parameter) (rl4co.models.zoo.ptrnet.model.PointerNetwork parameter) (rl4co.models.zoo.symnco.model.SymNCO parameter) (rl4co.models.zoo.symnco.policy.SymNCOPolicy.forward parameter) (rl4co.tasks.eval.EvalBase parameter)

env_aug_feats() (in module rl4co.data.transforms) env_context_embedding() (in module rl4co.models.nn.env_embeddings.context) env_dynamic_embedding() (in module rl4co.models.nn.env_embeddings.dynamic) env_init_embedding() (in module rl4co.models.nn.env_embeddings.init) env_name (rl4co.data.transforms.StateAugmentation parameter) (rl4co.models.rl.common.critic.CriticNetwork parameter) (rl4co.models.zoo.am.policy.AttentionModelPolicy parameter) (rl4co.models.zoo.common.autoregressive.decoder.AutoregressiveDecoder parameter) (rl4co.models.zoo.common.autoregressive.encoder.GraphAttentionEncoder parameter) (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy parameter) (rl4co.models.zoo.ham.policy.HeterogeneousAttentionModelPolicy parameter) (rl4co.models.zoo.pomo.policy.POMOPolicy parameter) (rl4co.models.zoo.ppo.policy.PPOPolicy parameter) (rl4co.models.zoo.symnco.policy.SymNCOPolicy parameter) EnvContext (class in rl4co.models.nn.env_embeddings.context) epoch_callback() (rl4co.models.rl.reinforce.baselines.REINFORCEBaseline method), [1] (rl4co.models.rl.reinforce.baselines.RolloutBaseline method), [1] (rl4co.models.rl.reinforce.baselines.WarmupBaseline method), [1] eval() (rl4co.models.rl.reinforce.baselines.CriticBaseline method), [1] (rl4co.models.rl.reinforce.baselines.ExponentialBaseline method), [1] (rl4co.models.rl.reinforce.baselines.NoBaseline method), [1] (rl4co.models.rl.reinforce.baselines.REINFORCEBaseline method), [1] (rl4co.models.rl.reinforce.baselines.RolloutBaseline method), [1] (rl4co.models.rl.reinforce.baselines.SharedBaseline method), [1] (rl4co.models.rl.reinforce.baselines.WarmupBaseline method), [1] eval_fn (rl4co.tasks.eval.get_automatic_batch_size parameter) EvalBase (class in rl4co.tasks.eval) evaluate_action() (rl4co.models.zoo.ppo.decoder.PPODecoder method) (rl4co.models.zoo.ppo.policy.PPOPolicy method) evaluate_policy() (in module rl4co.tasks.eval) ExponentialBaseline (class in rl4co.models.rl.reinforce.baselines), [1] ExtraKeyDataset (class in rl4co.data.dataset)

F

feed_forward_hidden (rl4co.models.nn.graph.attnnet.GraphAttentionNetwork parameter) (rl4co.models.nn.graph.attnnet.MultiHeadAttentionLayer parameter) (rl4co.models.zoo.common.autoregressive.encoder.GraphAttentionEncoder parameter) FFSPEnv (class in rl4co.envs.ffsp) first_augment (rl4co.data.transforms.symmetric_augmentation parameter) flash_attn_wrapper() (in module rl4co.models.nn.attention) (in module rl4co.models.nn.flash_attention) (rl4co.models.nn.attention.LogitAttention method) (rl4co.models.nn.attention.MultiHeadAttention method) (rl4co.models.nn.flash_attention.MHA method) FlashCrossAttention (class in rl4co.models.nn.flash_attention) FlashSelfAttention (class in rl4co.models.nn.flash_attention) force_dihedral_8 (rl4co.tasks.eval.AugmentationEval parameter) (rl4co.tasks.eval.GreedyMultiStartAugmentEval parameter) force_flash_attn (rl4co.models.nn.attention.LogitAttention parameter) (rl4co.models.nn.attention.MultiHeadAttention parameter) (rl4co.models.nn.graph.attnnet.GraphAttentionNetwork parameter) (rl4co.models.nn.graph.attnnet.MultiHeadAttentionLayer parameter) (rl4co.models.rl.common.critic.CriticNetwork parameter) (rl4co.models.zoo.common.autoregressive.encoder.GraphAttentionEncoder parameter) (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy parameter) forward() (rl4co.models.nn.attention.LogitAttention method) (rl4co.models.nn.attention.MultiHeadAttention method) (rl4co.models.nn.env_embeddings.context.DPPContext method) (rl4co.models.nn.env_embeddings.context.EnvContext method) (rl4co.models.nn.env_embeddings.context.PDPContext method) (rl4co.models.nn.env_embeddings.context.TSPContext method) (rl4co.models.nn.env_embeddings.dynamic.SDVRPDynamicEmbedding method) (rl4co.models.nn.env_embeddings.dynamic.StaticEmbedding method) (rl4co.models.nn.env_embeddings.init.DPPInitEmbedding method) (rl4co.models.nn.env_embeddings.init.MDPPInitEmbedding method) (rl4co.models.nn.env_embeddings.init.MTSPInitEmbedding method) (rl4co.models.nn.env_embeddings.init.OPInitEmbedding method) (rl4co.models.nn.env_embeddings.init.PCTSPInitEmbedding method) (rl4co.models.nn.env_embeddings.init.PDPInitEmbedding method) (rl4co.models.nn.env_embeddings.init.TSPInitEmbedding method) (rl4co.models.nn.env_embeddings.init.VRPInitEmbedding method) (rl4co.models.nn.flash_attention.CrossAttention method) (rl4co.models.nn.flash_attention.FlashCrossAttention method) (rl4co.models.nn.flash_attention.FlashSelfAttention method) (rl4co.models.nn.flash_attention.LinearResidual method) (rl4co.models.nn.flash_attention.MHA method) (rl4co.models.nn.flash_attention.SelfAttention method) (rl4co.models.nn.graph.attnnet.GraphAttentionNetwork method) (rl4co.models.nn.graph.gcn.GCNEncoder method) (rl4co.models.nn.graph.mpnn.MessagePassingEncoder method) (rl4co.models.nn.graph.mpnn.MessagePassingLayer method) (rl4co.models.nn.mlp.MLP method) (rl4co.models.nn.ops.Normalization method) (rl4co.models.nn.ops.SkipConnection method) (rl4co.models.nn.utils.RandomPolicy method) (rl4co.models.rl.common.base.RL4COLitModule method) (rl4co.models.rl.common.critic.CriticNetwork method) (rl4co.models.zoo.common.autoregressive.decoder.AutoregressiveDecoder method) (rl4co.models.zoo.common.autoregressive.encoder.GraphAttentionEncoder method) (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy method) (rl4co.models.zoo.eas.nn.EASLayerNet method) (rl4co.models.zoo.ham.attention.HeterogenousMHA method) (rl4co.models.zoo.ham.encoder.GraphHeterogeneousAttentionEncoder method) (rl4co.models.zoo.mdam.decoder.Decoder method) (rl4co.models.zoo.mdam.encoder.GraphAttentionEncoder method) (rl4co.models.zoo.mdam.encoder.MultiHeadAttention method) (rl4co.models.zoo.mdam.encoder.Normalization method) (rl4co.models.zoo.mdam.encoder.SkipConnection method) (rl4co.models.zoo.mdam.policy.MDAMPolicy method) (rl4co.models.zoo.ptrnet.critic.CriticNetworkLSTM method) (rl4co.models.zoo.ptrnet.decoder.Decoder method) (rl4co.models.zoo.ptrnet.decoder.SimpleAttention method) (rl4co.models.zoo.ptrnet.encoder.Encoder method) (rl4co.models.zoo.ptrnet.policy.PointerNetworkPolicy method) (rl4co.models.zoo.symnco.policy.SymNCOPolicy method)

forward_eas() (in module rl4co.models.zoo.eas.decoder) forward_logit_attn_eas_lay() (in module rl4co.models.zoo.eas.decoder) freq_file (rl4co.envs.dpp.DPPEnv parameter)

G

GCNEncoder (class in rl4co.models.nn.graph.gcn) generate_data (rl4co.models.rl.common.base.RL4COLitModule parameter) generate_data() (rl4co.envs.atsp.ATSPEnv method) (rl4co.envs.common.base.RL4COEnvBase method), [1] (rl4co.envs.cvrp.CVRPEnv method) (rl4co.envs.dpp.DPPEnv method) (rl4co.envs.ffsp.FFSPEnv method) (rl4co.envs.mdpp.MDPPEnv method) (rl4co.envs.mtsp.MTSPEnv method) (rl4co.envs.op.OPEnv method) (rl4co.envs.pctsp.PCTSPEnv method) (rl4co.envs.pdp.PDPEnv method) (rl4co.envs.tsp.TSPEnv method) generate_dataset() (in module rl4co.data.generate_data) generate_default_datasets() (in module rl4co.data.generate_data) generate_env_data() (in module rl4co.data.generate_data) generate_mdpp_data() (in module rl4co.data.generate_data) generate_op_data() (in module rl4co.data.generate_data) generate_pctsp_data() (in module rl4co.data.generate_data) generate_tsp_data() (in module rl4co.data.generate_data) generate_vrp_data() (in module rl4co.data.generate_data) get_action_mask() (rl4co.envs.common.base.RL4COEnvBase method), [1] (rl4co.envs.cvrp.CVRPEnv static method) (rl4co.envs.op.OPEnv static method) (rl4co.envs.pctsp.PCTSPEnv static method) (rl4co.envs.sdvrp.SDVRPEnv static method)

get_automatic_batch_size() (in module rl4co.tasks.eval) get_log_likelihood() (in module rl4co.models.nn.utils) get_reinforce_baseline() (in module rl4co.models.rl.reinforce.baselines), [1] get_reward() (rl4co.envs.atsp.ATSPEnv method) (rl4co.envs.common.base.RL4COEnvBase method), [1] (rl4co.envs.cvrp.CVRPEnv method) (rl4co.envs.dpp.DPPEnv method) (rl4co.envs.ffsp.FFSPEnv method) (rl4co.envs.mdpp.MDPPEnv method) (rl4co.envs.mtsp.MTSPEnv method) (rl4co.envs.op.OPEnv method) (rl4co.envs.pctsp.PCTSPEnv method) (rl4co.envs.pdp.PDPEnv static method) (rl4co.envs.tsp.TSPEnv static method) glimpse_key (rl4co.models.zoo.common.autoregressive.decoder.PrecomputedCache attribute) (rl4co.models.zoo.mdam.decoder.PrecomputedCache attribute) glimpse_val (rl4co.models.zoo.common.autoregressive.decoder.PrecomputedCache attribute) (rl4co.models.zoo.mdam.decoder.PrecomputedCache attribute) graph_context (rl4co.models.zoo.common.autoregressive.decoder.PrecomputedCache attribute) (rl4co.models.zoo.mdam.decoder.PrecomputedCache attribute) GraphAttentionEncoder (class in rl4co.models.zoo.common.autoregressive.encoder) (class in rl4co.models.zoo.mdam.encoder) GraphAttentionNetwork (class in rl4co.models.nn.graph.attnnet) GraphHeterogeneousAttentionEncoder (class in rl4co.models.zoo.ham.encoder) GreedyEval (class in rl4co.tasks.eval) GreedyMultiStartAugmentEval (class in rl4co.tasks.eval) GreedyMultiStartEval (class in rl4co.tasks.eval)

H

h (rl4co.models.zoo.ham.attention.HeterogenousMHA.forward parameter) (rl4co.models.zoo.mdam.encoder.MultiHeadAttention.forward parameter) HeterogeneousAttentionModel (class in rl4co.models.zoo.ham.model) HeterogeneousAttentionModelPolicy (class in rl4co.models.zoo.ham.policy)

HeterogeneuousMHALayer (class in rl4co.models.zoo.ham.encoder) HeterogenousMHA (class in rl4co.models.zoo.ham.attention) hidden (rl4co.models.zoo.ptrnet.decoder.Decoder.forward parameter) hidden_dim (rl4co.models.rl.common.critic.CriticNetwork parameter) https (rl4co.models.nn.flash_attention.MHA.forward parameter)

I

inference_params (rl4co.models.nn.flash_attention.MHA.forward parameter) init_embedding (rl4co.models.zoo.common.autoregressive.encoder.GraphAttentionEncoder parameter) (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy parameter) init_hidden() (rl4co.models.zoo.ptrnet.encoder.Encoder method) init_parameters() (rl4co.models.nn.ops.Normalization method) (rl4co.models.zoo.ham.attention.HeterogenousMHA method) (rl4co.models.zoo.mdam.encoder.MultiHeadAttention method) (rl4co.models.zoo.mdam.encoder.Normalization method)

inputs (rl4co.models.zoo.ptrnet.critic.CriticNetworkLSTM.forward parameter) instantiate_metrics() (rl4co.models.rl.common.base.RL4COLitModule method) invariance_loss() (in module rl4co.models.zoo.symnco.losses)

K

key_padding_mask (rl4co.models.nn.flash_attention.CrossAttention.forward parameter) (rl4co.models.nn.flash_attention.MHA.forward parameter) (rl4co.models.nn.flash_attention.SelfAttention.forward parameter)

kv (rl4co.models.nn.flash_attention.CrossAttention.forward parameter) (rl4co.models.nn.flash_attention.FlashCrossAttention.forward parameter)

L

linear_bias (rl4co.models.zoo.common.autoregressive.decoder.AutoregressiveDecoder parameter) LinearResidual (class in rl4co.models.nn.flash_attention) litmodule_kwargs (rl4co.models.rl.common.base.RL4COLitModule parameter) load_data() (rl4co.envs.common.base.RL4COEnvBase static method), [1] (rl4co.envs.cvrp.CVRPEnv static method) (rl4co.envs.dpp.DPPEnv method) load_from_checkpoint() (rl4co.models.rl.reinforce.reinforce.REINFORCE class method), [1] load_npz_to_tensordict() (in module rl4co.data.utils)

log_likelihood (rl4co.models.rl.reinforce.reinforce.REINFORCE.calculate_loss parameter), [1] log_metrics() (rl4co.models.rl.common.base.RL4COLitModule method) logit_key (rl4co.models.zoo.common.autoregressive.decoder.PrecomputedCache attribute) (rl4co.models.zoo.mdam.decoder.PrecomputedCache attribute) LogitAttention (class in rl4co.models.nn.attention) lr_scheduler (rl4co.models.rl.common.base.RL4COLitModule parameter) lr_scheduler_interval (rl4co.models.rl.common.base.RL4COLitModule parameter) lr_scheduler_kwargs (rl4co.models.rl.common.base.RL4COLitModule parameter) lr_scheduler_monitor (rl4co.models.rl.common.base.RL4COLitModule parameter)

M

mask (rl4co.models.nn.graph.attnnet.GraphAttentionNetwork.forward parameter) (rl4co.models.nn.graph.gcn.GCNEncoder.forward parameter) (rl4co.models.zoo.common.autoregressive.encoder.GraphAttentionEncoder.forward parameter) (rl4co.models.zoo.ham.attention.HeterogenousMHA.forward parameter) (rl4co.models.zoo.mdam.encoder.MultiHeadAttention.forward parameter) mask_inner (rl4co.models.nn.attention.LogitAttention parameter) (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy parameter) mask_logits (rl4co.models.nn.attention.LogitAttention parameter) max_batch_size (rl4co.tasks.eval.get_automatic_batch_size parameter) max_decaps (rl4co.envs.dpp.DPPEnv parameter) max_demand (rl4co.envs.cvrp.CVRPEnv parameter) (rl4co.envs.sdvrp.SDVRPEnv parameter) max_iters (rl4co.models.zoo.active_search.search.ActiveSearch parameter) (rl4co.models.zoo.eas.search.EAS parameter) max_length (rl4co.envs.op.OPEnv parameter) max_loc (rl4co.envs.cvrp.CVRPEnv parameter) (rl4co.envs.dpp.DPPEnv parameter) (rl4co.envs.mtsp.MTSPEnv parameter) (rl4co.envs.op.OPEnv parameter) (rl4co.envs.pctsp.PCTSPEnv parameter) (rl4co.envs.sdvrp.SDVRPEnv parameter) max_num_agents (rl4co.envs.mtsp.MTSPEnv parameter) max_runtime (rl4co.models.zoo.active_search.search.ActiveSearch parameter) (rl4co.models.zoo.eas.search.EAS parameter) max_seqlen (rl4co.models.nn.flash_attention.FlashCrossAttention.forward parameter) (rl4co.models.nn.flash_attention.FlashSelfAttention.forward parameter) (rl4co.models.nn.flash_attention.MHA.forward parameter) max_seqlen_k (rl4co.models.nn.flash_attention.FlashCrossAttention.forward parameter) max_time (rl4co.envs.ffsp.FFSPEnv parameter) MDAM (class in rl4co.models.zoo.mdam.model) MDAMPolicy (class in rl4co.models.zoo.mdam.policy) MDPPEnv (class in rl4co.envs.mdpp) MDPPInitEmbedding (class in rl4co.models.nn.env_embeddings.init) message() (rl4co.models.nn.graph.mpnn.MessagePassingEncoder method) (rl4co.models.nn.graph.mpnn.MessagePassingLayer method) MessagePassingEncoder (class in rl4co.models.nn.graph.mpnn) MessagePassingLayer (class in rl4co.models.nn.graph.mpnn) metrics (rl4co.models.rl.common.base.RL4COLitModule parameter) MHA (class in rl4co.models.nn.flash_attention) min_demand (rl4co.envs.cvrp.CVRPEnv parameter) (rl4co.envs.sdvrp.SDVRPEnv parameter) min_loc (rl4co.envs.cvrp.CVRPEnv parameter) (rl4co.envs.dpp.DPPEnv parameter) (rl4co.envs.mtsp.MTSPEnv parameter) (rl4co.envs.op.OPEnv parameter) (rl4co.envs.pctsp.PCTSPEnv parameter) (rl4co.envs.sdvrp.SDVRPEnv parameter) min_max_normalize() (in module rl4co.data.transforms) min_num_agents (rl4co.envs.mtsp.MTSPEnv parameter) min_time (rl4co.envs.ffsp.FFSPEnv parameter) mixer_subset (rl4co.models.nn.flash_attention.MHA.forward parameter) MLP (class in rl4co.models.nn.mlp) module rl4co.data.dataset rl4co.data.generate_data rl4co.data.transforms rl4co.data.utils rl4co.envs.atsp rl4co.envs.common.base, [1] rl4co.envs.cvrp rl4co.envs.dpp rl4co.envs.ffsp rl4co.envs.mdpp rl4co.envs.mtsp rl4co.envs.op rl4co.envs.pctsp rl4co.envs.pdp rl4co.envs.sdvrp rl4co.envs.spctsp rl4co.envs.tsp rl4co.models.nn.attention rl4co.models.nn.env_embeddings.context rl4co.models.nn.env_embeddings.dynamic rl4co.models.nn.env_embeddings.init rl4co.models.nn.flash_attention rl4co.models.nn.graph.attnnet rl4co.models.nn.graph.gcn rl4co.models.nn.graph.mpnn rl4co.models.nn.mlp rl4co.models.nn.ops rl4co.models.nn.utils rl4co.models.rl.common.base rl4co.models.rl.common.critic rl4co.models.rl.ppo.ppo, [1] rl4co.models.rl.reinforce.baselines, [1] rl4co.models.rl.reinforce.reinforce, [1] rl4co.models.zoo.active_search.search rl4co.models.zoo.am.model rl4co.models.zoo.am.policy rl4co.models.zoo.common.autoregressive.decoder rl4co.models.zoo.common.autoregressive.encoder rl4co.models.zoo.common.autoregressive.policy rl4co.models.zoo.eas.decoder rl4co.models.zoo.eas.nn rl4co.models.zoo.eas.search rl4co.models.zoo.ham.attention rl4co.models.zoo.ham.encoder rl4co.models.zoo.ham.model rl4co.models.zoo.ham.policy rl4co.models.zoo.mdam.decoder rl4co.models.zoo.mdam.encoder rl4co.models.zoo.mdam.model rl4co.models.zoo.mdam.policy rl4co.models.zoo.pomo.model rl4co.models.zoo.pomo.policy rl4co.models.zoo.ppo.decoder rl4co.models.zoo.ppo.model rl4co.models.zoo.ppo.policy rl4co.models.zoo.ptrnet.critic rl4co.models.zoo.ptrnet.decoder rl4co.models.zoo.ptrnet.encoder rl4co.models.zoo.ptrnet.model rl4co.models.zoo.ptrnet.policy rl4co.models.zoo.symnco.losses rl4co.models.zoo.symnco.model rl4co.models.zoo.symnco.policy rl4co.tasks.eval rl4co.tasks.train

MTSPContext (class in rl4co.models.nn.env_embeddings.context) MTSPEnv (class in rl4co.envs.mtsp) MTSPInitEmbedding (class in rl4co.models.nn.env_embeddings.init) MultiHeadAttention (class in rl4co.models.nn.attention) (class in rl4co.models.zoo.mdam.encoder) MultiHeadAttentionLayer (class in rl4co.models.nn.graph.attnnet) (class in rl4co.models.zoo.mdam.encoder)

N

n_epochs (rl4co.models.rl.reinforce.baselines.WarmupBaseline parameter), [1] name (rl4co.envs.atsp.ATSPEnv attribute) (rl4co.envs.cvrp.CVRPEnv attribute) (rl4co.envs.dpp.DPPEnv attribute) (rl4co.envs.ffsp.FFSPEnv attribute) (rl4co.envs.mdpp.MDPPEnv attribute) (rl4co.envs.mtsp.MTSPEnv attribute) (rl4co.envs.op.OPEnv attribute) (rl4co.envs.pctsp.PCTSPEnv attribute) (rl4co.envs.pdp.PDPEnv attribute) (rl4co.envs.sdvrp.SDVRPEnv attribute) (rl4co.envs.spctsp.SPCTSPEnv attribute) (rl4co.envs.tsp.TSPEnv attribute) (rl4co.tasks.eval.AugmentationEval attribute) (rl4co.tasks.eval.EvalBase attribute) (rl4co.tasks.eval.GreedyEval attribute) (rl4co.tasks.eval.GreedyMultiStartAugmentEval attribute) (rl4co.tasks.eval.GreedyMultiStartEval attribute) (rl4co.tasks.eval.SamplingEval attribute) NoBaseline (class in rl4co.models.rl.reinforce.baselines), [1] node_embeddings (rl4co.models.zoo.common.autoregressive.decoder.PrecomputedCache attribute) (rl4co.models.zoo.mdam.decoder.PrecomputedCache attribute) Normalization (class in rl4co.models.nn.ops) (class in rl4co.models.zoo.mdam.encoder) normalization (rl4co.models.nn.graph.attnnet.GraphAttentionNetwork parameter) (rl4co.models.nn.graph.attnnet.MultiHeadAttentionLayer parameter) (rl4co.models.rl.common.critic.CriticNetwork parameter) (rl4co.models.zoo.am.policy.AttentionModelPolicy parameter) (rl4co.models.zoo.common.autoregressive.encoder.GraphAttentionEncoder parameter) (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy parameter) (rl4co.models.zoo.ham.policy.HeterogeneousAttentionModelPolicy parameter) (rl4co.models.zoo.pomo.policy.POMOPolicy parameter) (rl4co.models.zoo.ppo.policy.PPOPolicy parameter) (rl4co.models.zoo.symnco.policy.SymNCOPolicy parameter) normalize (rl4co.data.transforms.StateAugmentation parameter) (rl4co.models.nn.attention.LogitAttention parameter) num_augment (rl4co.data.transforms.StateAugmentation parameter) (rl4co.data.transforms.symmetric_augmentation parameter) (rl4co.models.zoo.pomo.model.POMO parameter) (rl4co.models.zoo.symnco.model.SymNCO parameter) (rl4co.tasks.eval.AugmentationEval parameter) (rl4co.tasks.eval.AugmentationEval property) (rl4co.tasks.eval.GreedyMultiStartAugmentEval parameter) (rl4co.tasks.eval.GreedyMultiStartAugmentEval property) num_encoder_layers (rl4co.models.zoo.am.policy.AttentionModelPolicy parameter) (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy parameter) (rl4co.models.zoo.ham.policy.HeterogeneousAttentionModelPolicy parameter) (rl4co.models.zoo.pomo.policy.POMOPolicy parameter) (rl4co.models.zoo.ppo.policy.PPOPolicy parameter) (rl4co.models.zoo.symnco.policy.SymNCOPolicy parameter)

num_gcn_layer (rl4co.models.nn.graph.gcn.GCNEncoder parameter) num_heads (rl4co.models.nn.attention.LogitAttention parameter) (rl4co.models.nn.attention.MultiHeadAttention parameter) (rl4co.models.nn.graph.attnnet.GraphAttentionNetwork parameter) (rl4co.models.nn.graph.attnnet.MultiHeadAttentionLayer parameter) (rl4co.models.rl.common.critic.CriticNetwork parameter) (rl4co.models.zoo.am.policy.AttentionModelPolicy parameter) (rl4co.models.zoo.common.autoregressive.decoder.AutoregressiveDecoder parameter) (rl4co.models.zoo.common.autoregressive.encoder.GraphAttentionEncoder parameter) (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy parameter) (rl4co.models.zoo.ham.policy.HeterogeneousAttentionModelPolicy parameter) (rl4co.models.zoo.pomo.policy.POMOPolicy parameter) (rl4co.models.zoo.ppo.policy.PPOPolicy parameter) (rl4co.models.zoo.symnco.policy.SymNCOPolicy parameter) num_instances (rl4co.models.zoo.eas.nn.EASLayerNet parameter) num_job (rl4co.envs.ffsp.FFSPEnv parameter) num_keepout_max (rl4co.envs.dpp.DPPEnv parameter) num_keepout_min (rl4co.envs.dpp.DPPEnv parameter) num_layers (rl4co.models.nn.graph.attnnet.GraphAttentionNetwork parameter) (rl4co.models.rl.common.critic.CriticNetwork parameter) (rl4co.models.zoo.common.autoregressive.encoder.GraphAttentionEncoder parameter) num_loc (rl4co.envs.atsp.ATSPEnv parameter) (rl4co.envs.cvrp.CVRPEnv parameter) (rl4co.envs.mtsp.MTSPEnv parameter) (rl4co.envs.op.OPEnv parameter) (rl4co.envs.pctsp.PCTSPEnv parameter) (rl4co.envs.pdp.PDPEnv parameter) (rl4co.envs.sdvrp.SDVRPEnv parameter) (rl4co.envs.tsp.TSPEnv parameter) num_machine (rl4co.envs.ffsp.FFSPEnv parameter) num_nodes (rl4co.models.nn.graph.gcn.GCNEncoder parameter) num_probes_max (rl4co.envs.mdpp.MDPPEnv parameter) num_probes_min (rl4co.envs.mdpp.MDPPEnv parameter) num_stage (rl4co.envs.ffsp.FFSPEnv parameter) num_starts (rl4co.models.zoo.common.autoregressive.decoder.AutoregressiveDecoder.forward parameter) (rl4co.models.zoo.eas.decoder.forward_eas parameter) (rl4co.models.zoo.pomo.model.POMO parameter) (rl4co.models.zoo.symnco.model.SymNCO parameter) (rl4co.tasks.eval.GreedyMultiStartAugmentEval parameter) (rl4co.tasks.eval.GreedyMultiStartEval parameter)

O

offset (rl4co.data.transforms.symmetric_transform parameter) on_train_batch_end() (rl4co.models.zoo.active_search.search.ActiveSearch method) (rl4co.models.zoo.eas.search.EAS method) on_train_batch_start() (rl4co.models.zoo.active_search.search.ActiveSearch method) (rl4co.models.zoo.eas.search.EAS method) on_train_epoch_end() (rl4co.models.rl.common.base.RL4COLitModule method) (rl4co.models.rl.ppo.ppo.PPO method), [1] (rl4co.models.rl.reinforce.reinforce.REINFORCE method), [1] (rl4co.models.zoo.active_search.search.ActiveSearch method) (rl4co.models.zoo.eas.search.EAS method)

OPContext (class in rl4co.models.nn.env_embeddings.context) OPEnv (class in rl4co.envs.op) OPInitEmbedding (class in rl4co.models.nn.env_embeddings.init) optimizer (rl4co.models.rl.common.base.RL4COLitModule parameter) (rl4co.models.zoo.active_search.search.ActiveSearch parameter) (rl4co.models.zoo.eas.search.EAS parameter) optimizer_kwargs (rl4co.models.rl.common.base.RL4COLitModule parameter) (rl4co.models.zoo.active_search.search.ActiveSearch parameter) (rl4co.models.zoo.eas.search.EAS parameter)

P

parallel_runs (rl4co.models.zoo.active_search.search.ActiveSearch parameter) (rl4co.models.zoo.eas.search.EAS parameter) parameters (rl4co.models.rl.common.base.RL4COLitModule.configure_optimizers parameter) (rl4co.models.rl.ppo.ppo.PPO.configure_optimizers parameter), [1] PCTSPContext (class in rl4co.models.nn.env_embeddings.context) PCTSPEnv (class in rl4co.envs.pctsp) PCTSPInitEmbedding (class in rl4co.models.nn.env_embeddings.init) PDPContext (class in rl4co.models.nn.env_embeddings.context) PDPEnv (class in rl4co.envs.pdp) PDPInitEmbedding (class in rl4co.models.nn.env_embeddings.init) penalty_factor (rl4co.envs.pctsp.PCTSPEnv parameter) phase (rl4co.models.rl.reinforce.reinforce.REINFORCE.set_decode_type_multistart parameter), [1] (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy.forward parameter) (rl4co.models.zoo.mdam.policy.MDAMPolicy.forward parameter) (rl4co.models.zoo.symnco.policy.SymNCOPolicy.forward parameter) phi (rl4co.data.transforms.symmetric_transform parameter) PointerNetwork (class in rl4co.models.zoo.ptrnet.model) PointerNetworkPolicy (class in rl4co.models.zoo.ptrnet.policy) policy (rl4co.models.rl.common.base.RL4COLitModule parameter) (rl4co.models.rl.reinforce.reinforce.REINFORCE parameter), [1] (rl4co.models.zoo.active_search.search.ActiveSearch parameter) (rl4co.models.zoo.am.model.AttentionModel parameter) (rl4co.models.zoo.eas.search.EAS parameter) (rl4co.models.zoo.ham.model.HeterogeneousAttentionModel parameter) (rl4co.models.zoo.mdam.model.MDAM parameter) (rl4co.models.zoo.pomo.model.POMO parameter) (rl4co.models.zoo.ppo.model.PPOModel parameter) (rl4co.models.zoo.ptrnet.model.PointerNetwork parameter) (rl4co.models.zoo.symnco.model.SymNCO parameter)

policy_kwargs (rl4co.models.zoo.am.model.AttentionModel parameter) (rl4co.models.zoo.ham.model.HeterogeneousAttentionModel parameter) (rl4co.models.zoo.mdam.model.MDAM parameter) (rl4co.models.zoo.pomo.model.POMO parameter) (rl4co.models.zoo.ppo.model.PPOModel parameter) (rl4co.models.zoo.ptrnet.model.PointerNetwork parameter) (rl4co.models.zoo.symnco.model.SymNCO parameter) policy_out (rl4co.models.rl.reinforce.reinforce.REINFORCE.calculate_loss parameter), [1] POMO (class in rl4co.models.zoo.pomo.model) POMOPolicy (class in rl4co.models.zoo.pomo.policy) post_setup_hook() (rl4co.models.rl.common.base.RL4COLitModule method) (rl4co.models.rl.reinforce.reinforce.REINFORCE method), [1] PPO (class in rl4co.models.rl.ppo.ppo), [1] PPODecoder (class in rl4co.models.zoo.ppo.decoder) PPOModel (class in rl4co.models.zoo.ppo.model) PPOPolicy (class in rl4co.models.zoo.ppo.policy) PrecomputedCache (class in rl4co.models.zoo.common.autoregressive.decoder) (class in rl4co.models.zoo.mdam.decoder) prize_required (rl4co.envs.pctsp.PCTSPEnv parameter) prize_type (rl4co.envs.op.OPEnv parameter) problem_symmetricity_loss() (in module rl4co.models.zoo.symnco.losses) progress (rl4co.tasks.eval.EvalBase parameter) progress_bar (rl4co.models.rl.reinforce.baselines.RolloutBaseline parameter), [1] projection_head (rl4co.models.zoo.symnco.policy.SymNCOPolicy parameter)

Q

q (rl4co.models.nn.flash_attention.CrossAttention.forward parameter) (rl4co.models.nn.flash_attention.FlashCrossAttention.forward parameter) (rl4co.models.zoo.ham.attention.HeterogenousMHA.forward parameter) (rl4co.models.zoo.mdam.encoder.MultiHeadAttention.forward parameter)

qkv (rl4co.models.nn.flash_attention.FlashSelfAttention.forward parameter) (rl4co.models.nn.flash_attention.SelfAttention.forward parameter) query (rl4co.models.zoo.ptrnet.decoder.SimpleAttention.forward parameter)

R

random_policy() (in module rl4co.models.nn.utils) RandomPolicy (class in rl4co.models.nn.utils) recurrence() (rl4co.models.zoo.ptrnet.decoder.Decoder method) ref (rl4co.models.zoo.ptrnet.decoder.SimpleAttention.forward parameter) REINFORCE (class in rl4co.models.rl.reinforce.reinforce), [1] REINFORCEBaseline (class in rl4co.models.rl.reinforce.baselines), [1] render() (rl4co.envs.atsp.ATSPEnv static method) (rl4co.envs.common.base.RL4COEnvBase method), [1] (rl4co.envs.cvrp.CVRPEnv static method) (rl4co.envs.dpp.DPPEnv method) (rl4co.envs.ffsp.FFSPEnv method) (rl4co.envs.mdpp.MDPPEnv method) (rl4co.envs.mtsp.MTSPEnv static method) (rl4co.envs.op.OPEnv static method) (rl4co.envs.pctsp.PCTSPEnv static method) (rl4co.envs.pdp.PDPEnv static method) (rl4co.envs.tsp.TSPEnv static method) residual (rl4co.models.nn.graph.gcn.GCNEncoder parameter) return_actions (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy.forward parameter) (rl4co.models.zoo.mdam.policy.MDAMPolicy.forward parameter) (rl4co.models.zoo.symnco.policy.SymNCOPolicy.forward parameter) return_entropy (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy.forward parameter) (rl4co.models.zoo.mdam.policy.MDAMPolicy.forward parameter) (rl4co.models.zoo.symnco.policy.SymNCOPolicy.forward parameter) reward (rl4co.models.rl.reinforce.reinforce.REINFORCE.calculate_loss parameter), [1] reward_type (rl4co.envs.mdpp.MDPPEnv parameter) rl4co.data.dataset module rl4co.data.generate_data module rl4co.data.transforms module rl4co.data.utils module rl4co.envs.atsp module rl4co.envs.common.base module, [1] rl4co.envs.cvrp module rl4co.envs.dpp module rl4co.envs.ffsp module rl4co.envs.mdpp module rl4co.envs.mtsp module rl4co.envs.op module rl4co.envs.pctsp module rl4co.envs.pdp module rl4co.envs.sdvrp module rl4co.envs.spctsp module rl4co.envs.tsp module rl4co.models.nn.attention module rl4co.models.nn.env_embeddings.context module rl4co.models.nn.env_embeddings.dynamic module rl4co.models.nn.env_embeddings.init module rl4co.models.nn.flash_attention module rl4co.models.nn.graph.attnnet module rl4co.models.nn.graph.gcn module rl4co.models.nn.graph.mpnn module rl4co.models.nn.mlp module rl4co.models.nn.ops module

rl4co.models.nn.utils module rl4co.models.rl.common.base module rl4co.models.rl.common.critic module rl4co.models.rl.ppo.ppo module, [1] rl4co.models.rl.reinforce.baselines module, [1] rl4co.models.rl.reinforce.reinforce module, [1] rl4co.models.zoo.active_search.search module rl4co.models.zoo.am.model module rl4co.models.zoo.am.policy module rl4co.models.zoo.common.autoregressive.decoder module rl4co.models.zoo.common.autoregressive.encoder module rl4co.models.zoo.common.autoregressive.policy module rl4co.models.zoo.eas.decoder module rl4co.models.zoo.eas.nn module rl4co.models.zoo.eas.search module rl4co.models.zoo.ham.attention module rl4co.models.zoo.ham.encoder module rl4co.models.zoo.ham.model module rl4co.models.zoo.ham.policy module rl4co.models.zoo.mdam.decoder module rl4co.models.zoo.mdam.encoder module rl4co.models.zoo.mdam.model module rl4co.models.zoo.mdam.policy module rl4co.models.zoo.pomo.model module rl4co.models.zoo.pomo.policy module rl4co.models.zoo.ppo.decoder module rl4co.models.zoo.ppo.model module rl4co.models.zoo.ppo.policy module rl4co.models.zoo.ptrnet.critic module rl4co.models.zoo.ptrnet.decoder module rl4co.models.zoo.ptrnet.encoder module rl4co.models.zoo.ptrnet.model module rl4co.models.zoo.ptrnet.policy module rl4co.models.zoo.symnco.losses module rl4co.models.zoo.symnco.model module rl4co.models.zoo.symnco.policy module rl4co.tasks.eval module rl4co.tasks.train module RL4COEnvBase (class in rl4co.envs.common.base), [1] RL4COLitModule (class in rl4co.models.rl.common.base) rollout() (in module rl4co.models.nn.utils) (rl4co.models.rl.reinforce.baselines.RolloutBaseline method), [1] RolloutBaseline (class in rl4co.models.rl.reinforce.baselines), [1]

S

samples (rl4co.tasks.eval.SamplingEval parameter) SamplingEval (class in rl4co.tasks.eval) save_path (rl4co.models.zoo.active_search.search.ActiveSearch parameter) (rl4co.models.zoo.eas.search.EAS parameter) SDVRPDynamicEmbedding (class in rl4co.models.nn.env_embeddings.dynamic) SDVRPEnv (class in rl4co.envs.sdvrp) seed (rl4co.envs.atsp.ATSPEnv parameter) (rl4co.envs.common.base.RL4COEnvBase parameter), [1] (rl4co.envs.pdp.PDPEnv parameter) (rl4co.envs.tsp.TSPEnv parameter) select_start_nodes_fn (rl4co.models.zoo.common.autoregressive.decoder.AutoregressiveDecoder parameter) self_loop (rl4co.models.nn.graph.gcn.GCNEncoder parameter) SelfAttention (class in rl4co.models.nn.flash_attention) set_decode_type_multistart() (rl4co.models.rl.reinforce.reinforce.REINFORCE method), [1] setup() (rl4co.models.rl.common.base.RL4COLitModule method) (rl4co.models.rl.reinforce.baselines.CriticBaseline method), [1] (rl4co.models.rl.reinforce.baselines.REINFORCEBaseline method), [1] (rl4co.models.rl.reinforce.baselines.RolloutBaseline method), [1] (rl4co.models.rl.reinforce.baselines.WarmupBaseline method), [1] (rl4co.models.zoo.active_search.search.ActiveSearch method) (rl4co.models.zoo.eas.search.EAS method) setup_loggers() (rl4co.models.rl.common.base.RL4COLitModule method) shared_step() (rl4co.models.rl.common.base.RL4COLitModule method) (rl4co.models.rl.ppo.ppo.PPO method), [1] (rl4co.models.rl.reinforce.reinforce.REINFORCE method), [1] (rl4co.models.zoo.pomo.model.POMO method) (rl4co.models.zoo.symnco.model.SymNCO method)

SharedBaseline (class in rl4co.models.rl.reinforce.baselines), [1] should contain 1 if attention is not possible (rl4co.models.zoo.ham.attention.HeterogenousMHA.forward parameter) shuffle_train_dataloader (rl4co.models.rl.common.base.RL4COLitModule parameter) SimpleAttention (class in rl4co.models.zoo.ptrnet.decoder) SkipConnection (class in rl4co.models.nn.ops) (class in rl4co.models.zoo.mdam.encoder) softmax_scale (rl4co.models.nn.flash_attention.CrossAttention parameter) (rl4co.models.nn.flash_attention.FlashCrossAttention parameter) (rl4co.models.nn.flash_attention.FlashSelfAttention parameter) (rl4co.models.nn.flash_attention.SelfAttention parameter) softmax_temp (rl4co.models.nn.attention.LogitAttention parameter) (rl4co.models.zoo.common.autoregressive.decoder.AutoregressiveDecoder.forward parameter) (rl4co.models.zoo.eas.decoder.forward_eas parameter) (rl4co.tasks.eval.SamplingEval parameter) solution_symmetricity_loss() (in module rl4co.models.zoo.symnco.losses) SPCTSPEnv (class in rl4co.envs.spctsp) start_batch_size (rl4co.tasks.eval.get_automatic_batch_size parameter) StateAugmentation (class in rl4co.data.transforms) StaticEmbedding (class in rl4co.models.nn.env_embeddings.dynamic) stochastic (rl4co.envs.pctsp.PCTSPEnv property) (rl4co.envs.spctsp.SPCTSPEnv property) symmetric_augmentation() (in module rl4co.data.transforms) symmetric_transform() (in module rl4co.data.transforms) SymNCO (class in rl4co.models.zoo.symnco.model) SymNCOPolicy (class in rl4co.models.zoo.symnco.policy)

T

tanh_clipping (rl4co.models.nn.attention.LogitAttention parameter) td (rl4co.models.nn.graph.gcn.GCNEncoder.forward parameter) (rl4co.models.rl.reinforce.reinforce.REINFORCE.calculate_loss parameter), [1] (rl4co.models.zoo.common.autoregressive.decoder.AutoregressiveDecoder.forward parameter) (rl4co.models.zoo.common.autoregressive.encoder.GraphAttentionEncoder.forward parameter) (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy.forward parameter) (rl4co.models.zoo.eas.decoder.forward_eas parameter) (rl4co.models.zoo.mdam.policy.MDAMPolicy.forward parameter) (rl4co.models.zoo.ppo.decoder.PPODecoder.evaluate_action parameter) (rl4co.models.zoo.ppo.policy.PPOPolicy.evaluate_action parameter) (rl4co.models.zoo.symnco.policy.SymNCOPolicy.forward parameter) td_params (rl4co.envs.atsp.ATSPEnv parameter) (rl4co.envs.cvrp.CVRPEnv parameter) (rl4co.envs.dpp.DPPEnv parameter) (rl4co.envs.mdpp.MDPPEnv parameter) (rl4co.envs.mtsp.MTSPEnv parameter) (rl4co.envs.op.OPEnv parameter) (rl4co.envs.pctsp.PCTSPEnv parameter) (rl4co.envs.pdp.PDPEnv parameter) (rl4co.envs.sdvrp.SDVRPEnv parameter) (rl4co.envs.tsp.TSPEnv parameter)

tensordict_collate_fn() (in module rl4co.data.dataset) TensorDictDataset (class in rl4co.data.dataset) test_batch_size (rl4co.models.rl.common.base.RL4COLitModule parameter) test_data_size (rl4co.models.rl.common.base.RL4COLitModule parameter) test_dataloader() (rl4co.models.rl.common.base.RL4COLitModule method) test_decode_type (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy parameter) test_file (rl4co.envs.common.base.RL4COEnvBase parameter), [1] test_step() (rl4co.models.rl.common.base.RL4COLitModule method) train() (in module rl4co.tasks.train) train_data_size (rl4co.models.rl.common.base.RL4COLitModule parameter) train_dataloader() (rl4co.models.rl.common.base.RL4COLitModule method) train_decode_type (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy parameter) train_file (rl4co.envs.common.base.RL4COEnvBase parameter), [1] training (rl4co.envs.common.base.RL4COEnvBase attribute) training_step() (rl4co.models.rl.common.base.RL4COLitModule method) (rl4co.models.zoo.active_search.search.ActiveSearch method) (rl4co.models.zoo.eas.search.EAS method) transform() (rl4co.envs.common.base.RL4COEnvBase method), [1] TSPContext (class in rl4co.models.nn.env_embeddings.context) TSPEnv (class in rl4co.envs.tsp) TSPInitEmbedding (class in rl4co.models.nn.env_embeddings.init)

U

update() (rl4co.models.nn.graph.mpnn.MessagePassingEncoder method) (rl4co.models.nn.graph.mpnn.MessagePassingLayer method) update_mask() (rl4co.models.zoo.ptrnet.decoder.Decoder method) url (rl4co.envs.dpp.DPPEnv parameter) use_dihedral_8 (rl4co.data.transforms.StateAugmentation parameter) (rl4co.models.zoo.pomo.model.POMO parameter)

use_eas_embedding (rl4co.models.zoo.eas.search.EAS parameter) use_eas_layer (rl4co.models.zoo.eas.search.EAS parameter) use_graph_context (rl4co.models.zoo.common.autoregressive.decoder.AutoregressiveDecoder parameter) (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy parameter) use_projection_head (rl4co.models.zoo.symnco.policy.SymNCOPolicy parameter)

V

val_batch_size (rl4co.models.rl.common.base.RL4COLitModule parameter) val_data_size (rl4co.models.rl.common.base.RL4COLitModule parameter) val_dataloader() (rl4co.models.rl.common.base.RL4COLitModule method) val_decode_type (rl4co.models.zoo.common.autoregressive.policy.AutoregressivePolicy parameter) val_file (rl4co.envs.common.base.RL4COEnvBase parameter), [1]

validation_step() (rl4co.models.rl.common.base.RL4COLitModule method) vehicle_capacity (rl4co.envs.cvrp.CVRPEnv parameter) (rl4co.envs.sdvrp.SDVRPEnv parameter) verbose (rl4co.models.zoo.eas.search.EAS parameter) VRPContext (class in rl4co.models.nn.env_embeddings.context) VRPInitEmbedding (class in rl4co.models.nn.env_embeddings.init)

W

warmup_exp_beta (rl4co.models.rl.reinforce.baselines.WarmupBaseline parameter), [1] WarmupBaseline (class in rl4co.models.rl.reinforce.baselines), [1] wrap_dataset() (rl4co.models.rl.common.base.RL4COLitModule method) (rl4co.models.rl.reinforce.baselines.REINFORCEBaseline method), [1] (rl4co.models.rl.reinforce.baselines.RolloutBaseline method), [1] (rl4co.models.rl.reinforce.baselines.WarmupBaseline method), [1] (rl4co.models.rl.reinforce.reinforce.REINFORCE method), [1]

X

x (rl4co.data.transforms.symmetric_transform parameter) (rl4co.models.nn.flash_attention.MHA.forward parameter) (rl4co.models.nn.graph.attnnet.GraphAttentionNetwork.forward parameter) (rl4co.models.rl.common.critic.CriticNetwork.forward parameter)

x_kv (rl4co.models.nn.flash_attention.MHA.forward parameter) xy (rl4co.data.transforms.dihedral_8_augmentation parameter) (rl4co.data.transforms.symmetric_augmentation parameter)

Y

y (rl4co.data.transforms.symmetric_transform parameter)