Upload folder using huggingface_hub

small-il-filled-actions/ckpts/policy_weights/policy_epoch_0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ddd10d0b4b3a617e5a0aa2f80b5cdfcdbaa1569f19e8e5f09cce7e3e29262d7
+size 54911318

small-il-filled-actions/ckpts/policy_weights/policy_epoch_10.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:714e6e189153b7c2544b5bf26fb268bc74d6d146cf92db49f053c18ffd2dd681
+size 54911490

small-il-filled-actions/ckpts/policy_weights/policy_epoch_12.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:054d16533bdded0111d09855fc485ce437e0ceb65e99e424d12a6dc77c9c69bc
+size 54911490

small-il-filled-actions/ckpts/policy_weights/policy_epoch_14.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74297e23886489f57ae239628c951a659753309c56667320514ee6905e0c783f
+size 54911490

small-il-filled-actions/ckpts/policy_weights/policy_epoch_16.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4fedd848eb34cb83c60557b12587af5bc384b3a9e537f3634ae2648233490b4c
+size 54911490

small-il-filled-actions/ckpts/policy_weights/policy_epoch_18.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a5278df38443052baa9d5ae80ac2f1750826c250bb13bcc960115ed67fd51bc
+size 54911490

small-il-filled-actions/ckpts/policy_weights/policy_epoch_2.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4eb9769a5c16633b770866ba11cd3566b86b2db17da34fe8532d81e64ec9c5df
+size 54911318

small-il-filled-actions/ckpts/policy_weights/policy_epoch_20.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebc1eff2981b0a73b9ba300c0851d105a612da4fd084e68d8b0ca90819bf37e9
+size 54911490

small-il-filled-actions/ckpts/policy_weights/policy_epoch_22.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9680265d089e3f958f8ce715387609a8c576b608b136f9fde93f37bd465603d0
+size 54911490

small-il-filled-actions/ckpts/policy_weights/policy_epoch_24.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dca0c4d5adb81cd77d461ef1feb2df2d695862e2a57f3c39de7dd7b23d3d092d
+size 54911490

small-il-filled-actions/ckpts/policy_weights/policy_epoch_26.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:adbb6d6112d7ab873287c9fcabbe1b382ad0116accfc74b40d25be98b71d4775
+size 54911490

small-il-filled-actions/ckpts/policy_weights/policy_epoch_28.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39ee5159c27eae9b8bc911034459847d8a35668e4178d789e354a9f028190423
+size 54911490

small-il-filled-actions/ckpts/policy_weights/policy_epoch_30.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69176e8355e96af083309815245fbe8b1eb4d4a1f64d2636234031e2e01e7fcc
+size 54911490

small-il-filled-actions/ckpts/policy_weights/policy_epoch_32.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af8a2b5b009f2db4baac75e4c41bc8d34fc9f008f1a2874d30486a2c8d75c08c
+size 54911490

small-il-filled-actions/ckpts/policy_weights/policy_epoch_34.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6df602b3a4d682c7c9e29d87f8959e40be9e416581d7a798eb78d7758eff189
+size 54911490

small-il-filled-actions/ckpts/policy_weights/policy_epoch_36.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3883ceb31d9286dbe3f54969f9444b720ad0b8c834464ea15ced44f6e4b5389b
+size 54911490

small-il-filled-actions/ckpts/policy_weights/policy_epoch_38.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05970b9539dd11dc606dfac126650eefa79f6b5bc44c6efb30adf7bdf819a981
+size 54911490

small-il-filled-actions/ckpts/policy_weights/policy_epoch_4.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4411f43bde99a5a31cecbb0329adf511e9b6df93255b13f60ac0f4aab3ce52b6
+size 54911318

small-il-filled-actions/ckpts/policy_weights/policy_epoch_40.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8969849d0033017142c7014b4d944570f48a9fa150244f6ad7f02db9e18b233
+size 54911490

small-il-filled-actions/ckpts/policy_weights/policy_epoch_42.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:653ae11b7f34c9fb404210f3ee23423c8258b335dd8a3b1b1dd97ec25cd550a4
+size 54911490

small-il-filled-actions/ckpts/policy_weights/policy_epoch_44.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:817c0973caffb95a1445fb150211d5ac1ec51de3e4e1d228a9fb3607aa365522
+size 54911490

small-il-filled-actions/ckpts/policy_weights/policy_epoch_6.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1017611b71f2af60a6524accd84967ea401e5af30bd3e84be373abf70e9d2be
+size 54911318

small-il-filled-actions/ckpts/policy_weights/policy_epoch_8.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce7b6ec476b047c67c1bf3f0951fe0ffeaae393bd1b506e890a6a4c0b96fdde6
+size 54911318

small-il-filled-actions/config.txt

@@ -0,0 +1,106 @@
+# Parameters for Actor:
+# ==============================================================================
+Actor.activation = 'leaky_relu'
+Actor.cont_dist_kind = 'normal'
+Actor.d_hidden = 300
+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 = True
+Agent.gamma = 0.999
+Agent.num_critics = 4
+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 = False
+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 = 100
+MetamonTstepEncoder.extra_emb_dim = 18
+MetamonTstepEncoder.n_heads = 5
+MetamonTstepEncoder.n_layers = 3
+MetamonTstepEncoder.scratch_tokens = 4
+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 = 300
+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 = 2048
+TformerTrajEncoder.d_model = 512
+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 = 8
+TformerTrajEncoder.n_layers = 3
+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