jakegrigsby commited on
Commit
567af47
·
verified ·
1 Parent(s): e2733c7

Upload folder using huggingface_hub

Browse files
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_0.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae374aa7833188118e494a1f63b134b4ea148add51e577964ba645e1a90aca9f
3
+ size 55821590
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_10.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b4aed278834b0d7736e475ae16dfcfb7a72db43108edea75012b9cbd7b65d73
3
+ size 55821762
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_12.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:729d12ecbf43452292ebdeff36b909fde4d423f57113a64de5acd4e1fba5ff16
3
+ size 55821762
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_14.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0202a5ddc1a1f1bd6528244b68ab2fd757cbddb0c21cda483492837db996cbc7
3
+ size 55821762
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_16.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fed2fb9f27e0f985eac1248fc2bc289d8a7350ea8507751284e5b82c6c524e16
3
+ size 55821762
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_18.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1cddaced1dd41d8e1c97cf8db6989e3cec0450d437c469860236d460431374f
3
+ size 55821762
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_2.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:feaf4d61e9c2bb342debd21d8b144219a8b8404a1a7096cef2e6c2062e55aa1c
3
+ size 55821590
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_20.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f3668edf28ac26cc36f631ffb43d9cadc410e0462f78ea16cd17118d7b819a0a
3
+ size 55821762
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_22.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99223296574c3912aaedc34595eeb281b0628332d9a2c13185d05bc43ec44ae8
3
+ size 55821762
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_24.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fbb1c5a9acf4abf8a72b5b910ae147677ef23946facecbcb42c312cda1d32fc1
3
+ size 55821762
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_26.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:483815d838d47408ab494ee86800c61f105b7c1e0d31b8d11c11023a85ea9f5f
3
+ size 55821762
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_28.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b792fef72c2859caec062891ce8c3397fc9dd77e28e7ce4826ad89a0815f97dd
3
+ size 55821762
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_30.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4128e38d0f70c4d5c90bb0c57d702cd4764520ad3774b3bc75cfd9b14857cb2d
3
+ size 55821762
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_32.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd69088d67c90ab4a2a8ef133a85d65ba4afef66a88cbb47eadd7ad05b885c7b
3
+ size 55821762
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_34.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02065de6be523d5b84d70c3bc738281868a2cd14570ccbf84a6df81d7ae0da0d
3
+ size 55821762
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_36.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1e435ddbe1f8442eacecf215a5bf8bdb27d9a1f8b35dc8d293048473504bbe1
3
+ size 55821762
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_38.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:edb5f5814e6545c235f216fc5a87a07acb25d535c5b1a7e15919165f32c22111
3
+ size 55821762
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_4.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:17403c568b2f7fca0858bf3d9ae49d56b0eaf3aedbf1c01768c330c523ae19d3
3
+ size 55821590
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_6.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dcf764e48574167d67508df6533a74d410a8b93487cb1df2e940d8f06f2ea97c
3
+ size 55821590
small-rl-exp-extreme/ckpts/policy_weights/policy_epoch_8.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c6631bf669cfc1a0a19fa2e5cec02780b5a042b7c1ae819583abccdec055ebf
3
+ size 55821590
small-rl-exp-extreme/config.txt ADDED
@@ -0,0 +1,106 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Parameters for Actor:
2
+ # ==============================================================================
3
+ Actor.activation = 'leaky_relu'
4
+ Actor.cont_dist_kind = 'normal'
5
+ Actor.d_hidden = 300
6
+ Actor.dropout_p = 0.0
7
+ Actor.gmm_modes = 5
8
+ Actor.log_std_high = 2.0
9
+ Actor.log_std_low = -5.0
10
+ Actor.n_layers = 2
11
+
12
+ # Parameters for Agent:
13
+ # ==============================================================================
14
+ Agent.fake_filter = False
15
+ Agent.gamma = 0.999
16
+ Agent.num_critics = 4
17
+ Agent.num_critics_td = 2
18
+ Agent.offline_coeff = 1.0
19
+ Agent.online_coeff = 0.0
20
+ Agent.popart = True
21
+ Agent.reward_multiplier = 10.0
22
+ Agent.tau = 0.003
23
+ Agent.use_multigamma = True
24
+ Agent.use_target_actor = True
25
+
26
+ # Parameters for Experiment:
27
+ # ==============================================================================
28
+ Experiment.batches_per_update = 1
29
+ Experiment.critic_loss_weight = 10.0
30
+ Experiment.env_mode = 'async'
31
+ Experiment.force_reset_train_envs_every = None
32
+ Experiment.grad_clip = 1.0
33
+ Experiment.has_replay_buffer_rights = True
34
+ Experiment.l2_coeff = 0.001
35
+ Experiment.learning_rate = 0.0001
36
+ Experiment.local_time_optimizer = False
37
+ Experiment.lr_warmup_steps = 500
38
+ Experiment.mixed_precision = 'no'
39
+ Experiment.padded_sampling = 'none'
40
+ Experiment.save_trajs_as = 'npz'
41
+ Experiment.stagger_traj_file_lengths = True
42
+ Experiment.wandb_group_name = None
43
+
44
+ # Parameters for FlashAttention:
45
+ # ==============================================================================
46
+ FlashAttention.window_size = (-1, -1)
47
+
48
+ # Parameters for MetamonTstepEncoder:
49
+ # ==============================================================================
50
+ MetamonTstepEncoder.d_model = 100
51
+ MetamonTstepEncoder.extra_emb_dim = 18
52
+ MetamonTstepEncoder.n_heads = 5
53
+ MetamonTstepEncoder.n_layers = 3
54
+ MetamonTstepEncoder.scratch_tokens = 4
55
+ MetamonTstepEncoder.token_mask_aug = False
56
+
57
+ # Parameters for Multigammas:
58
+ # ==============================================================================
59
+ Multigammas.continuous = [0.1, 0.9, 0.95, 0.97, 0.99, 0.995]
60
+ Multigammas.discrete = [0.1, 0.9, 0.95, 0.97, 0.99, 0.995]
61
+
62
+ # Parameters for MultiModalEmbedding:
63
+ # ==============================================================================
64
+ MultiModalEmbedding.dropout = 0.05
65
+ MultiModalEmbedding.numerical_tokens = 6
66
+
67
+ # Parameters for NCritics:
68
+ # ==============================================================================
69
+ NCritics.activation = 'leaky_relu'
70
+ NCritics.d_hidden = 300
71
+ NCritics.dropout_p = 0.0
72
+ NCritics.n_layers = 2
73
+
74
+ # Parameters for PopArtLayer:
75
+ # ==============================================================================
76
+ PopArtLayer.beta = 0.0005
77
+ PopArtLayer.init_nu = 100.0
78
+
79
+ # Parameters for TformerTrajEncoder:
80
+ # ==============================================================================
81
+ TformerTrajEncoder.activation = 'leaky_relu'
82
+ TformerTrajEncoder.causal = True
83
+ TformerTrajEncoder.d_ff = 2048
84
+ TformerTrajEncoder.d_model = 512
85
+ TformerTrajEncoder.dropout_attn = 0.0
86
+ TformerTrajEncoder.dropout_emb = 0.05
87
+ TformerTrajEncoder.dropout_ff = 0.05
88
+ TformerTrajEncoder.dropout_qkv = 0.0
89
+ TformerTrajEncoder.head_scaling = True
90
+ TformerTrajEncoder.n_heads = 8
91
+ TformerTrajEncoder.n_layers = 3
92
+ TformerTrajEncoder.norm = 'layer'
93
+ TformerTrajEncoder.normformer_norms = True
94
+ TformerTrajEncoder.sigma_reparam = True
95
+
96
+ # Parameters for TimestepTransformer:
97
+ # ==============================================================================
98
+ # None.
99
+
100
+ # Parameters for TokenEmbedding:
101
+ # ==============================================================================
102
+ # None.
103
+
104
+ # Parameters for TransformerTurnEmbedding:
105
+ # ==============================================================================
106
+ TransformerTurnEmbedding.dropout = 0.05