Upload folder using huggingface_hub

small-rl-aug/ckpts/policy_weights/policy_epoch_0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49f1c6884ef557ecd9fb38925286ad2ac6947b11acfc839ea613abe4be5b4157
+size 55821590

small-rl-aug/ckpts/policy_weights/policy_epoch_10.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb91f0b1a3ecb3254df4b383d62982e2162d3a36fef9c37fd02b448eb07a401f
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_12.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b76fc2a621f7af7cf816d2d277a6b411a7d718a57b8eb43ce7ff4c6dbfb35ba0
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_14.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00f281b769dab51294cc3a732c301b5ddd8f48cd0ce21567c2fe642eab2c6f91
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_16.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5591741ffa55a5b3102df7b76a1d8218171ddf926abe0cf499c098bf3d46f892
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_18.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d905347f35e2443dfadcc3c15425d52e705e3b4d314ea8829cc0e6b613a99372
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_2.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44b690c942a55a2ce546059b54337f9046952b52c99e6a03ff782fb1679ede01
+size 55821590

small-rl-aug/ckpts/policy_weights/policy_epoch_20.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13b718c993b3351d26bef5b328bfb67295ba28f5caa6e708a0fad6807da20c61
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_22.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4134a55cb66f343db9371661b5abfacfab27fd53c1293e0a76e6e7d6d4b2f32b
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_24.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07c1705bfef7681c5bc7478ebad756f6d7f9b03f16ee96b0be5ec7c54ce52a0d
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_26.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa0ed0b59c8c101a4b8a7c44ea52bca6748c9a1182a927fb7084d475b93863f3
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_28.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:623e2c414c7a4b53ec334aa2dd8fee3edede78e3f69a4c57da9516d7b80d51d9
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_30.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd920581b60c3b3ed41a31caf77b49f2c751a4c3819004cb78c8bb00dc87c707
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_32.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c9d8c216fb70c4db9ba2595292fde8bd335fe3879e014447b06b9532a779fa6
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_34.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:012f5eba11f2ffc0e01ec72d513d7a1ca46c9ed758a40bf678f9b314aa824565
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_36.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3938b0e4a26698484481e4e405d400a17d1f9c5e849ff0ef1c2d7080d8a5daf1
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_38.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c36f47420efbbb70583f41cd01d42138d84e5bfee842f0fda527a7d581866ddf
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_4.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9d6b44f67749e55e0fa4a00547e3c18f41f22d906db22215f954a9804d08e93
+size 55821590

small-rl-aug/ckpts/policy_weights/policy_epoch_40.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da7c8b8fecb1a74eb81948fe4f92cba0f823f7f0b3ec7d12c0d217f071476df6
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_42.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0e754c1e3199e00a5b2e4a763afceb58be047541c2e0ff2157ebe7fee847260
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_44.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:31d9505cb363932ffe5d051a6908dc5d81974d1b3d4971c3e3fc6ae6acb4d20d
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_46.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:434b33b830628132ac33b5019bb02bd7f12443ea266ce344809c3065b28a3409
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_48.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:742db071f1aef103cfadd416624b5ce81bd4f151993df152f7f33fc7004bdf56
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_50.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4f4d7ce37afd96e0fd0f8c32bdb1421f45a0c5440a4483ed3021623908d6661
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_52.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:23fc68b6f4e59b49d089e9a61ef803a34162eb41c6522bdd08011f0fed041237
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_54.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:895a4bb797fb6e4c0d4cea2eb4948a17bb1d8483f9ec93dbbeebb189f8b4d5bb
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_56.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4698d378be4d1617951903ec53df506d42d96a4d2f1f60541c549045c6fa2e35
+size 55821762

small-rl-aug/ckpts/policy_weights/policy_epoch_6.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9bbf5c18251fc3b1f6e69168f9f7ada92950369d254c82601a46e6c600a5c787
+size 55821590

small-rl-aug/ckpts/policy_weights/policy_epoch_8.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32b456332dbc24390de6de1509ff90989b0af8ea648b4f4e77cdd4fc5c829203
+size 55821590

small-rl-aug/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 = False
+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 = 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 = 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