Upload folder using huggingface_hub

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:138e3b239c53d43b0e322af7e7250c76f3b486093d7da027baec4f1ae32f7f36
+size 802136150

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_10.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a5e263f3bfd9522b50522bfd0772132dd8568f81918032ed41e3f63610875837
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_12.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bbd2d38295eb7fc29529cf916a656c7cbf12a34754a626a330e7ee4e56f17f76
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_14.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bdcc04411bdb88da06e6b1c2a9295b7d5ccbba46eaba527a3d0c75a2b44bfe19
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_16.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5dce4d292d4ce71f6715aa308476cf5af3935f14f5e4831af50e01c5d86aae10
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_18.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e54ab381bd31bfc04864ce016798a6219077310f675d264a4d0b68895b892adb
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_2.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d18edeed09295978633c8fbd58da8af833447e076e8c33895d5cd7909e34e60
+size 802136150

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_20.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:071ab72dab5c4b0150eee2842152382ed0c62fe68f346bea8af6c3f8159c86b0
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_22.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d85424f297c157efb5420eb3465f61d2c15fe0e9634325729fab7fec9fde57b
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_24.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca924bd0d38b43deb9267a024d26834aaaf48805289f311b865143da0af5bfaa
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_26.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90aa69350f2f4c9c3ce92115b23369294991c6a3c35e6b4ff986c801b8dd7032
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_28.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3ece160d904969e1daaea639131c2c3b1d3ad735509710fd12357508fdacdf0
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_30.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6313ec949f048bece17a0ff24af17dfccc972f8eb360809410cdb0bc65059a7
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_32.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:805e68fdaa826e76de2bf4fb80ad8d830895c72a4db58298e7e93affd19d9f8f
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_34.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c394c6f0c483dbe3e1a8124ba8829d22644c9e552781e14d87078531e83a91c
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_36.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:822f024359cac5457a08e2f5c05cf59b92c05eb205bc5fab4007a28020567275
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_38.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c4bffddb1d37b2cde0a6a67866d0f05a7389e06962a522c00d1dbc09f645a1d
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_4.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e06d8b4210a9260d23c352a72a1165bda9ec6c126b79c4a164fb7acc1ed8d8a
+size 802136150

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_40.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:282038907cad0decd2e83f6c4442a64ba6e183bffd7da2cb4befb307704e2b82
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_42.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5fe3d3629ff7b659ee4b6a136185ae01d2490800348e956bf5d37a56b1470bd
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_44.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2fcbff6a9b040d0a43de88dd86415e8e678ce0359c00261cfc4205ce8e4f73ac
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_46.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9580944f43d2b3c8e0d9304ca801254afcc26da70888b24df3d78529758122c8
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_48.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14f1f0b21a687d8e6ab63dd3895a3f9fd3d32f8bd05b84dd556b2b55c3d70d89
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_50.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4317c9a98cc7b7b455d0b536f242e77947642ab5dae82656945f98228af4df8f
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_52.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a2ea534fa0dcbb5adaac6ac3425f6d9ddb836e91dc5adf0d6ca8e9e8afd78f5
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_54.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68714c587dfc9aca4959c117f7971ced78b8d33b63b4b6e1435f5d7770f80a5c
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_56.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad4ecb377c9c29e8ea25b73eea96866f24e85a80a326a053dd007f405a31f905
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_58.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c117578c1778be1c6d722c757b20713b45afcc882d949ee5d0a62cc48853526b
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_6.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:59f67d0f27dbc53a31d7d8b7018b16637f835351e88f3158e840096a15b02037
+size 802136150

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_60.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b93007ea01c3ed233545f328fedcf726d1b6372c119aa009bab53e461d201f82
+size 802136514

synthetic-rl-v1/ckpts/policy_weights/policy_epoch_8.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7d18e7ac8d7fe823525d3d15ccbfa8b22f3f482a32c89c9e1b56109f506f31ef
+size 802136150

synthetic-rl-v1/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