jakegrigsby commited on
Commit
c5d1cb9
·
1 Parent(s): fe8861f
minikazam/ckpts/config.txt ADDED
@@ -0,0 +1,104 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import amago.agent
2
+ import amago.experiment as amago2
3
+ import amago.nets.actor_critic as amago3
4
+ import amago.nets.traj_encoders as amago4
5
+
6
+ # Parameters for Actor:
7
+ # ==============================================================================
8
+ # None.
9
+
10
+ # Parameters for Agent:
11
+ # ==============================================================================
12
+ Agent.actor_type = @MetamonMaskedActor
13
+ Agent.critic_type = @actor_critic.NCritics
14
+ Agent.fake_filter = False
15
+ Agent.fbc_filter_func = @agent.binary_filter
16
+ Agent.gamma = 0.999
17
+ Agent.num_actions_for_value_in_actor_loss = 1
18
+ Agent.num_actions_for_value_in_critic_loss = 1
19
+ Agent.num_critics = 4
20
+ Agent.num_critics_td = 2
21
+ Agent.offline_coeff = 1.0
22
+ Agent.online_coeff = 0.0
23
+ Agent.pass_obs_keys_to_actor = ['illegal_actions']
24
+ Agent.popart = True
25
+ Agent.reward_multiplier = 10.0
26
+ Agent.tau = 0.004
27
+ Agent.use_multigamma = True
28
+ Agent.use_target_actor = True
29
+
30
+ # Parameters for Discrete:
31
+ # ==============================================================================
32
+ Discrete.clip_prob_high = 0.99
33
+ Discrete.clip_prob_low = 0.001
34
+
35
+ # Parameters for Experiment:
36
+ # ==============================================================================
37
+ Experiment.always_load_latest = False
38
+ Experiment.always_save_latest = True
39
+ Experiment.has_dset_edit_rights = True
40
+ Experiment.save_trajs_as = 'npz'
41
+ Experiment.stagger_traj_file_lengths = True
42
+ Experiment.traj_save_len = 10000000000.0
43
+ Experiment.wandb_group_name = None
44
+
45
+ # Parameters for GRUTrajEncoder:
46
+ # ==============================================================================
47
+ GRUTrajEncoder.d_hidden = 400
48
+ GRUTrajEncoder.d_output = 300
49
+ GRUTrajEncoder.n_layers = 2
50
+ GRUTrajEncoder.norm = 'layer'
51
+
52
+ # Parameters for MetamonAMAGOExperiment:
53
+ # ==============================================================================
54
+ MetamonAMAGOExperiment.agent_type = @agent.Agent
55
+ MetamonAMAGOExperiment.critic_loss_weight = 10.0
56
+ MetamonAMAGOExperiment.grad_clip = 1.5
57
+ MetamonAMAGOExperiment.l2_coeff = 0.0001
58
+ MetamonAMAGOExperiment.learning_rate = 0.00015
59
+ MetamonAMAGOExperiment.lr_warmup_steps = 1000
60
+ MetamonAMAGOExperiment.max_seq_len = 64
61
+ MetamonAMAGOExperiment.traj_encoder_type = @traj_encoders.GRUTrajEncoder
62
+ MetamonAMAGOExperiment.tstep_encoder_type = @MetamonPerceiverTstepEncoder
63
+
64
+ # Parameters for MetamonMaskedActor:
65
+ # ==============================================================================
66
+ MetamonMaskedActor.activation = 'leaky_relu'
67
+ MetamonMaskedActor.continuous_dist_type = None
68
+ MetamonMaskedActor.d_hidden = 256
69
+ MetamonMaskedActor.dropout_p = 0.0
70
+ MetamonMaskedActor.mask_illegal_actions = True
71
+ MetamonMaskedActor.n_layers = 2
72
+
73
+ # Parameters for MetamonPerceiverTstepEncoder:
74
+ # ==============================================================================
75
+ MetamonPerceiverTstepEncoder.d_model = 64
76
+ MetamonPerceiverTstepEncoder.dropout = 0.05
77
+ MetamonPerceiverTstepEncoder.extra_emb_dim = 12
78
+ MetamonPerceiverTstepEncoder.latent_tokens = 5
79
+ MetamonPerceiverTstepEncoder.max_tokens_per_turn = 64
80
+ MetamonPerceiverTstepEncoder.n_heads = 4
81
+ MetamonPerceiverTstepEncoder.n_layers = 3
82
+ MetamonPerceiverTstepEncoder.numerical_tokens = 3
83
+ MetamonPerceiverTstepEncoder.token_mask_aug = False
84
+
85
+ # Parameters for Multigammas:
86
+ # ==============================================================================
87
+ Multigammas.continuous = [0.1, 0.9, 0.95, 0.97, 0.99, 0.995]
88
+ Multigammas.discrete = [0.1, 0.9, 0.95, 0.97, 0.99, 0.995]
89
+
90
+ # Parameters for NCritics:
91
+ # ==============================================================================
92
+ NCritics.activation = 'leaky_relu'
93
+ NCritics.d_hidden = 256
94
+ NCritics.dropout_p = 0.0
95
+ NCritics.n_layers = 2
96
+
97
+ # Parameters for PerceiverTurnEmbedding:
98
+ # ==============================================================================
99
+ # None.
100
+
101
+ # Parameters for PopArtLayer:
102
+ # ==============================================================================
103
+ PopArtLayer.beta = 0.0005
104
+ PopArtLayer.init_nu = 100.0
minikazam/ckpts/policy_weights/policy_epoch_0.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c718dcbfae9f382cbfd623b6e2cffe9d6e8cbf91a42ceafa932cefd7acb14bb0
3
+ size 19100644
minikazam/ckpts/policy_weights/policy_epoch_10.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:69252909102f9734080994657a77886bcff4b9f928f756b707adcdbc92df5621
3
+ size 19100777
minikazam/ckpts/policy_weights/policy_epoch_12.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:be462ca3e276b1da1f4063eaca4243265ffd4f4ca05a9ebd2131057fc2029416
3
+ size 19100777
minikazam/ckpts/policy_weights/policy_epoch_14.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:43f8889dbd5bbafaad39f55dd6ea1195a3c1306d004da8f0d384d73dfafdc358
3
+ size 19100777
minikazam/ckpts/policy_weights/policy_epoch_16.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ffd75dd32ca1b52976f6d968e9229925beb4c55545957203b62ac946cf8bdc29
3
+ size 19100777
minikazam/ckpts/policy_weights/policy_epoch_18.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a3235469579f618860ecf0f2c1c2549f97624a6a8cba7d8bd1da121c2e16b55a
3
+ size 19100777
minikazam/ckpts/policy_weights/policy_epoch_2.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ad085199917d9813409541b3b9bf2da6809c64ca3c27cb7134d4608c6ed4b1a
3
+ size 19100644
minikazam/ckpts/policy_weights/policy_epoch_20.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89c8b1e8ceea451876bb0723117fc0c2f8b3fcc56c84c85ae77ffb183bd4ec0a
3
+ size 19100777
minikazam/ckpts/policy_weights/policy_epoch_22.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3f1c3975b8d35c4afc59087253a6bf71568710d4aec8e5361ce1925f2e33cc0
3
+ size 19100777
minikazam/ckpts/policy_weights/policy_epoch_24.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7507ee7fbe0937411fd0d6e4f6d51462d14e71ba4b11c9afe05d39cce75bd7b2
3
+ size 19100777
minikazam/ckpts/policy_weights/policy_epoch_26.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fc32e803ec1a5626fb0f40bc826bf0208d77047b3485b283c76158efdb6166b7
3
+ size 19100777
minikazam/ckpts/policy_weights/policy_epoch_28.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d9ed7de81291de49a726f87e6405a6ddcadcd261fe5d3179b70078c94bf3f70f
3
+ size 19100777
minikazam/ckpts/policy_weights/policy_epoch_30.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3505d1196d57964a5c80c1052bd91680900b045c97a38b4a0a89a028c225db3
3
+ size 19100777
minikazam/ckpts/policy_weights/policy_epoch_32.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:800b3f43edcf4c4eb9b14fa8caf66f05562db0afb5b192bf7f7c2a4cc1464dec
3
+ size 19100777
minikazam/ckpts/policy_weights/policy_epoch_34.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71abdfd88f5c3b069d86001cff786be9b40171eabd8239db3224eb8d75074e26
3
+ size 19100777
minikazam/ckpts/policy_weights/policy_epoch_36.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:caad79021cd2384ea7517377e2869f98b02eba7191c451806168d265fd594649
3
+ size 19100777
minikazam/ckpts/policy_weights/policy_epoch_38.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:19c8c1d2cf18c8d5a4052daa4a613ea392fe84a667385c7a6331e6346373f48b
3
+ size 19100777
minikazam/ckpts/policy_weights/policy_epoch_4.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bde56224cba13cbc1fd39dad7b9cbaddd42ca10b4d75cb219616d8eb1eebedf7
3
+ size 19100644
minikazam/ckpts/policy_weights/policy_epoch_40.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dda9cf7752d68343992987319a96db0a4c87f011ae6fedbb0521e46f0385cc88
3
+ size 19100777
minikazam/ckpts/policy_weights/policy_epoch_42.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7516d2c6f4b668b835ad5ebf95c2a89c00c67989512312c494b640be3be9365b
3
+ size 19100777
minikazam/ckpts/policy_weights/policy_epoch_44.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c3eada9e0110a39a28b2ca871fa19bb2843c21aba898e4a6a1e0b6f23a5a6007
3
+ size 19100777
minikazam/ckpts/policy_weights/policy_epoch_46.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7bccf96b8159438e92b95b2f518488ba3519b27575e640ae3508587bb71031e
3
+ size 19100777
minikazam/ckpts/policy_weights/policy_epoch_6.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:854dd96d9054fb6f3024489202cd5f0452a8a0f6bc05eedda782057831814e22
3
+ size 19100644
minikazam/ckpts/policy_weights/policy_epoch_8.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b34f13fa265b9f7046b555663956ab8b59b54f92f3784e052358c583c0b039dc
3
+ size 19100644