Upload folder using huggingface_hub

medium-il/ckpts/policy_weights/policy_epoch_0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70576a52076da688bdd2ba22d77b1e52496379b601bbafb4a2f776a0fab6d3d1
+size 201117242

medium-il/ckpts/policy_weights/policy_epoch_10.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be0ff830a116013ad9fb30abc762a31265c839a4d031fb0910a94c6d9dbafbfe
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_12.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ce6f9a8d4f7e8c854327fe547d2dc269c2a7319dd600f2d39120c7c730bd64e
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_14.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:101d99d162bb64e8facf12d6414b856424793db9b50177aa1ddfa2cb88c128c2
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_16.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7055ab7c1ea0889090e78f818610fbf90216c837c24c4b92ed1ea546fbb62dc
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_18.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29c91eb711d7123c16091887d11865f04c9088bb16382b4026a3f3063128804b
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_2.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0329496197ae8c7c20971abd182aa055fd8edb24066369bd1f00196e5fccc0cf
+size 201117242

medium-il/ckpts/policy_weights/policy_epoch_20.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7eb0ab858577083e9c6ed414bfb6897d94fe151a66f8e22053789b1e8db41766
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_22.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76ffb222a882e42a2f5568222df45896402cfa6bb907a8869550a03e7c518020
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_24.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c806ea150e0317d3525287c0a198c9163a0f48904f75cfda802a874bef9a693
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_26.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9b16e4f259170eab3386e58e3895637cef9f1c7ad140b55f73994b11b4cd4ea
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_28.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a1d6c971d8d44ae1e5f9f93a252c080e4e670131b28d9bcc45e840b0022246a
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_30.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:450e13edbbeb827b7db14ddea9d401b05db7689d273c6df917c18d2e88d77944
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_32.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e582b2b2395d96bdacd2ceb2df3410667bbe1b72ce70f62cbc04bcf343246ac6
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_34.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:acecb233503f20b01ec85bff53f858a371a73762885eb9071eea8bb265437ab2
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_36.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b0490724f720dd5987daeeec8f4b49c28263d27db1e3c7ab478492e50126275
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_38.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:861a5ca4b700d9c008abfc85c87fb408cf6a1fa345faf3c21359b1eca7386565
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_4.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:08ef306cace5579f7ae4dcd186d89834895ada19669c50e9896a60bae6920808
+size 201117242

medium-il/ckpts/policy_weights/policy_epoch_40.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0560a056a5aa8e36870bde0e17f1d3d3a2f15052447a2b3e91813fca6b258a0
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_42.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e5fa6c00fc92d6e78957669250264c1232f69be928bfd3943e27f84b2a45908
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_44.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44a1a87c60f2bda8a6476b976c096134ebb9170ab28251fdc9cec2c867a8bfe3
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_46.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19d19b2005dba8bbe1c052bad8c83deede32f52f4a2cdf5fd91d3b54543bbeb3
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_48.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a32fd435bfceefa8f130226d00cb9e72041efa2fcc6076d187341f690b8b896
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_50.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44c9d03bcbdbaccc9ce40dbcc30f698dae74794cbb8e0acc5ac0793f7cf93f04
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_52.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8bdc37ddde109d4d2fa7e46f97c3b0701fb64d58094e6081dbf776d2d70ecb89
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_54.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:837e168a51c81757481376e4a2110be41b4f222783339d6e8c4df0ed33508fb9
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_56.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d567efac556ba724cf95630d5fb673f5d5d062dd136bf524c45379a5e6fb3ad
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_58.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b9de23df043ac0e9dabead2becf27c329959805f6cba99bb8d5bd9be100fb24
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_6.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:243b429a612c353abf7be06b26168cb4292e27bcb990a681f7404e4a4344bfda
+size 201117242

medium-il/ckpts/policy_weights/policy_epoch_60.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:143d9e0c2cf7e51dbd568ecda3b83d9807c0db08551ce496cc2a838f63dde307
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_62.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06784eb313ed45d1a6431e5524d910249808def3dddf31a7724f152912946886
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_64.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1399e89299ea87252468e4963bdea46c83c85e28d0f65719c02ec8942ab4b055
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_66.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:59bf282ed618392b55a6bb6e30d59157993f34eda148f07bd65523b5bd76eb27
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_68.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b222e25720894a0f90fbb74c4876700b1f4ce0be906055fc355c0f87cf900134
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_70.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a56fcdb4725394493f4283691dafcabe82c4e4773417e207c04958367b10cd39
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_72.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:121194e62a77c129d7cf69dfbb9e0a81059648574033f22752a8e1c6613c1c04
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_74.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d011869915159ff361c61e1cb7cab54f00f5e9a065831879074f67ba1d49ae2
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_76.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38fec8a14dfc3efa542095955d9f5b10c5fe2b3f51ca5585766cb8b0f7b0146f
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_78.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3e416e6e82dd872e47e1ebdd315d5ee52f6e19bf89d09d8cefdc2bdc55c14fb
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_8.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3219667ba938ff46892edd8d70f74862eae7e3c0c1a279d4b91dcd5ec01c01c9
+size 201117242

medium-il/ckpts/policy_weights/policy_epoch_80.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:424d82ea0c8372b0dbea1b1e5d6962a0354395bc448fffcad441be19c9a75841
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_82.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:24c2645ef4b4b830e4cb7cbb1e9ec1246bd11fa4e50a6668adf697aa669bd81d
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_84.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60204e561f49c7693e989fc20b6396c7e756142c055359599e562c7c3c934ad6
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_86.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:870764db307a65a8772af73632386c4ccf123bd9c6ef216f63b80fb943c285f9
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_88.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d05f3ca619e2f742ab2d2f6b3f11facc6c479edf7c83f57b4169401042a2b2e
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_90.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f98e34c211d1a0480dc2ae800081bca7113772696e8ef29b47d18c1372b6084a
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_92.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb314c53f61646690eae4b96c27aa55428d1accefbab0d1e8a6e268d94a2c2bc
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_94.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96c2ff9a9aa4378820cb3733138217a457331cab348b91382b6aa5d77cc73fa7
+size 201117498

medium-il/ckpts/policy_weights/policy_epoch_96.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1cc142f4e1f7b7a2b093a2b3401c293d151ba56781f543b6ba9a5393a12976b
+size 201117498

medium-il/config.txt

@@ -0,0 +1,106 @@
+# Parameters for Actor:
+# ==============================================================================
+Actor.activation = 'leaky_relu'
+Actor.cont_dist_kind = 'normal'
+Actor.d_hidden = 400
+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 = 6
+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 = 400
+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 = 3072
+TformerTrajEncoder.d_model = 768
+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 = 6
+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