Upload folder using huggingface_hub

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1a0933e801f1aaee819ecac249065d9c2a4140c2fc784ab097d15f9cf90d7ce
+size 802136150

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_10.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e140a7d585a9cc9182dc21abbd45a045e9793b5dd78f9750b33fb3feb5a75e58
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_12.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e538e152f7889d0f7a5ddffefec71356b5df6fbe093511566e4c6e9778f7696
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_14.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b5f93c7919d0305ed43c56844182005070a2e64f53a5e8af0bcab3eed94976f
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_16.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:673a1c8a1e299a2a4c16cc56fe8197aa968047db1242abee4589c129caa5dcdd
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_18.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25d419d33c9ed1ccc5aa46acfd105d2e4a433b94e6e88ddbc537261119b940fc
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_2.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da7c0de678b04a93b7bc02500231eb1f63ce997bf109e6b0eacac283fef35ff4
+size 802136150

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_20.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4d6d77be2ccaf780cfcddcd266150a076be490e257e85edd07a629a16e4f1ff
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_22.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd9eef7d95f62bcdae9328de27b8006672352c51df603dcde248f64f4da282a4
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_24.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14c52bf36aec8bd2fe0faf0a0b9a223cfdae18b548241441634299943b9c0eba
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_26.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64f511d2ddb48ee1d3725618a32233822bb25a423f90c25996746fcf9ddc3481
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_28.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e48764698092df0557e7494fe67b1140738132762483483dc732a3ab060d0f1
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_30.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30c36affb872dbcb5d0ea67c62783605ca936dbe173a010c1a685671e904764b
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_32.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:013840c458e7c70d523ab623b3c73d029ddd90ce349e7cd9f47432ac54030d20
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_34.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b6acb831ed6e85ed29e70d6ff89fc37e7fa666e2cdc002a9b0d8d0e4369c4c0
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_36.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:211dfe843d4af38a9023a988fbe32d98b9bbff03b412f173ffe48172822b44c5
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_38.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dece199c69b9a1b148bab59d9e27a38c6b3912b24ee09eb6a2771f420dd7e3cd
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_4.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6bc3827a9b486f6b1cb9214beaad79de30f585541635307417575bccd84a0d10
+size 802136150

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_40.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9b9bb9f5b8d8504a081e6da19bdd26ddd7877fae238f63b5d108bd3eddefaa5
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_42.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36ffdf814aef1f7705e6b6807c2d5f6a8ffed62a6c5845efa6211b5ad49620f8
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_44.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0de502c3acb6a7e4d9979aac13e534a335b349745409229e9c896a45d32aaa7b
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_46.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ff21e312a23f59bafa70373f5d96cc3deef97526b34104642089eff19034f26
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_48.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4a5718c6e298533c4fa9ac019d39d47290efed78c3e5efe0aa61d872a636418
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_50.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b71626b05b33af36a5b7e62de3b1b55d4421b00d958ae066c75ac288f8f65178
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_52.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aca97bc36778745e5bf47240b0a9b02c436ba59087663ef259bb93dc64f020a2
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_54.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5598e69a9361a65a54d025fd8dbb2a4aebac45d7746c4cb0dc919595a3ea85d
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_56.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dfba37e2aac279eabf29a84b5454379329d54a53754b4c1b578b19c44e493291
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_58.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a135c18e200226bec16175b411358e38485471f34b3b394227850b7d34b1141
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_6.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c830af594eb8961c87b6b9df6c16bcebaa6cc19e193a03799bddadb5eff04a5
+size 802136150

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_60.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06cd8149b33eb5070266aff74e1b287a53673080d514e23f0b9673779494ab84
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_62.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ebf0455982a78baae8be8233b5169ce6f6f38f877fa89cdf1222730465b127d
+size 802136514

synthetic-rl-v0/ckpts/policy_weights/policy_epoch_8.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eaec29ed668f77adb360f19923e9b0bdf9c55ccfc6d1cb5127cd9a3faae25fdf
+size 802136150

synthetic-rl-v0/config.txt

@@ -0,0 +1,106 @@
+# Parameters for Actor:
+# ==============================================================================
+Actor.activation = 'leaky_relu'
+Actor.cont_dist_kind = 'normal'
+Actor.d_hidden = 512
+Actor.dropout_p = 0.0
+Actor.gmm_modes = 5
+Actor.log_std_high = 2.0
+Actor.log_std_low = -5.0
+Actor.n_layers = 2
+
+# Parameters for Agent:
+# ==============================================================================
+Agent.fake_filter = False
+Agent.gamma = 0.999
+Agent.num_critics = 6
+Agent.num_critics_td = 2
+Agent.offline_coeff = 1.0
+Agent.online_coeff = 0.0
+Agent.popart = True
+Agent.reward_multiplier = 10.0
+Agent.tau = 0.003
+Agent.use_multigamma = True
+Agent.use_target_actor = True
+
+# Parameters for Experiment:
+# ==============================================================================
+Experiment.batches_per_update = 1
+Experiment.critic_loss_weight = 10.0
+Experiment.env_mode = 'async'
+Experiment.force_reset_train_envs_every = None
+Experiment.grad_clip = 1.0
+Experiment.has_replay_buffer_rights = True
+Experiment.l2_coeff = 0.001
+Experiment.learning_rate = 0.0001
+Experiment.local_time_optimizer = False
+Experiment.lr_warmup_steps = 500
+Experiment.mixed_precision = 'no'
+Experiment.padded_sampling = 'none'
+Experiment.save_trajs_as = 'npz'
+Experiment.stagger_traj_file_lengths = True
+Experiment.wandb_group_name = None
+
+# Parameters for FlashAttention:
+# ==============================================================================
+FlashAttention.window_size = (-1, -1)
+
+# Parameters for MetamonTstepEncoder:
+# ==============================================================================
+MetamonTstepEncoder.d_model = 160
+MetamonTstepEncoder.extra_emb_dim = 18
+MetamonTstepEncoder.n_heads = 8
+MetamonTstepEncoder.n_layers = 5
+MetamonTstepEncoder.scratch_tokens = 11
+MetamonTstepEncoder.token_mask_aug = False
+
+# Parameters for Multigammas:
+# ==============================================================================
+Multigammas.continuous = [0.1, 0.9, 0.95, 0.97, 0.99, 0.995]
+Multigammas.discrete = [0.1, 0.9, 0.95, 0.97, 0.99, 0.995]
+
+# Parameters for MultiModalEmbedding:
+# ==============================================================================
+MultiModalEmbedding.dropout = 0.05
+MultiModalEmbedding.numerical_tokens = 6
+
+# Parameters for NCritics:
+# ==============================================================================
+NCritics.activation = 'leaky_relu'
+NCritics.d_hidden = 512
+NCritics.dropout_p = 0.0
+NCritics.n_layers = 2
+
+# Parameters for PopArtLayer:
+# ==============================================================================
+PopArtLayer.beta = 0.0005
+PopArtLayer.init_nu = 100.0
+
+# Parameters for TformerTrajEncoder:
+# ==============================================================================
+TformerTrajEncoder.activation = 'leaky_relu'
+TformerTrajEncoder.causal = True
+TformerTrajEncoder.d_ff = 5120
+TformerTrajEncoder.d_model = 1280
+TformerTrajEncoder.dropout_attn = 0.0
+TformerTrajEncoder.dropout_emb = 0.05
+TformerTrajEncoder.dropout_ff = 0.05
+TformerTrajEncoder.dropout_qkv = 0.0
+TformerTrajEncoder.head_scaling = True
+TformerTrajEncoder.n_heads = 20
+TformerTrajEncoder.n_layers = 9
+TformerTrajEncoder.norm = 'layer'
+TformerTrajEncoder.normformer_norms = True
+TformerTrajEncoder.sigma_reparam = True
+
+# Parameters for TimestepTransformer:
+# ==============================================================================
+# None.
+
+# Parameters for TokenEmbedding:
+# ==============================================================================
+# None.
+
+# Parameters for TransformerTurnEmbedding:
+# ==============================================================================
+TransformerTurnEmbedding.dropout = 0.05