garvitsachdeva commited on
Commit
901dc66
Β·
verified Β·
1 Parent(s): dd6cf6a

Add trained SpindleFlow RL policy

Browse files
README.md ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ tags:
4
+ - reinforcement-learning
5
+ - stable-baselines3
6
+ - sb3-contrib
7
+ - gymnasium
8
+ - multi-agent
9
+ - openenv
10
+ library_name: stable-baselines3
11
+ ---
12
+
13
+ # SpindleFlow RL β€” Delegation Policy
14
+
15
+ LSTM PPO agent trained on SpindleFlow-v0 (OpenEnv).
16
+
17
+ ## Training summary
18
+ | Metric | Value |
19
+ |---|---|
20
+ | Algorithm | RecurrentPPO (SB3 + sb3-contrib) |
21
+ | Total timesteps | 30,000 |
22
+ | Episodes completed | 13526 |
23
+ | First-5 mean reward | 1.2053 |
24
+ | Last-5 mean reward | 2.2038 |
25
+ | Improvement | +0.9984 |
26
+ | Device | cuda |
27
+
28
+ ![Reward Curve](reward_curve.png)
29
+
30
+ ## Load
31
+ ```python
32
+ from sb3_contrib import RecurrentPPO
33
+ from huggingface_hub import hf_hub_download
34
+ model = RecurrentPPO.load(hf_hub_download("garvitsachdeva/spindleflow-rl", "spindleflow_model.zip"))
35
+ ```
data/resolution_memory.jsonl ADDED
@@ -0,0 +1,384 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": -0.0330449640750885, "episode_idx": 2}
2
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.06518733501434326, "episode_idx": 3}
3
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.04933221638202667, "episode_idx": 7}
4
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.04493609070777893, "episode_idx": 21}
5
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04224050045013428, "episode_idx": 58}
6
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.1007842868566513, "episode_idx": 76}
7
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.1007842868566513, "episode_idx": 76}
8
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": 0.08712831139564514, "episode_idx": 101}
9
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.010232031345367432, "episode_idx": 124}
10
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07675498723983765, "episode_idx": 132}
11
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04261964559555054, "episode_idx": 170}
12
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07675498723983765, "episode_idx": 173}
13
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.045468464493751526, "episode_idx": 180}
14
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.045468464493751526, "episode_idx": 180}
15
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.05466374754905701, "episode_idx": 183}
16
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.08782553672790527, "episode_idx": 190}
17
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.08782553672790527, "episode_idx": 190}
18
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.011156141757965088, "episode_idx": 191}
19
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.008337244391441345, "episode_idx": 192}
20
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.00812441110610962, "episode_idx": 198}
21
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": 0.00812441110610962, "episode_idx": 198}
22
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.024517789483070374, "episode_idx": 200}
23
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.04386478662490845, "episode_idx": 203}
24
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.018974751234054565, "episode_idx": 209}
25
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.005004033446311951, "episode_idx": 210}
26
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.01077599823474884, "episode_idx": 223}
27
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.010795056819915771, "episode_idx": 227}
28
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07926848530769348, "episode_idx": 233}
29
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07926848530769348, "episode_idx": 233}
30
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07023906707763672, "episode_idx": 235}
31
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04532673954963684, "episode_idx": 238}
32
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.005424603819847107, "episode_idx": 240}
33
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.0005497634410858154, "episode_idx": 241}
34
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.028721779584884644, "episode_idx": 246}
35
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.001269906759262085, "episode_idx": 269}
36
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.015913724899291992, "episode_idx": 291}
37
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.05068022012710571, "episode_idx": 299}
38
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": 0.0419897735118866, "episode_idx": 314}
39
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.023089230060577393, "episode_idx": 330}
40
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.023089230060577393, "episode_idx": 333}
41
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.023089230060577393, "episode_idx": 333}
42
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.05451443791389465, "episode_idx": 345}
43
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.05451443791389465, "episode_idx": 345}
44
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": 0.06659451127052307, "episode_idx": 356}
45
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": -0.012906238436698914, "episode_idx": 365}
46
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.041640281677246094, "episode_idx": 383}
47
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.02827012538909912, "episode_idx": 436}
48
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.027964696288108826, "episode_idx": 448}
49
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04387013614177704, "episode_idx": 454}
50
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.019650816917419434, "episode_idx": 462}
51
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.01874762773513794, "episode_idx": 466}
52
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.045781493186950684, "episode_idx": 475}
53
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.024114668369293213, "episode_idx": 481}
54
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": 0.026415586471557617, "episode_idx": 484}
55
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": 0.026415586471557617, "episode_idx": 486}
56
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": 0.010844096541404724, "episode_idx": 490}
57
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.009040668606758118, "episode_idx": 496}
58
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.023326635360717773, "episode_idx": 498}
59
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": -0.0019979923963546753, "episode_idx": 505}
60
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.01874762773513794, "episode_idx": 510}
61
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.002065315842628479, "episode_idx": 527}
62
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.026185855269432068, "episode_idx": 537}
63
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": 0.03301416337490082, "episode_idx": 540}
64
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.022716403007507324, "episode_idx": 541}
65
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": 0.008568033576011658, "episode_idx": 546}
66
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": 0.05161993205547333, "episode_idx": 553}
67
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": 0.05161993205547333, "episode_idx": 560}
68
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": -0.05892130732536316, "episode_idx": 576}
69
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": -0.021849453449249268, "episode_idx": 579}
70
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.01755937933921814, "episode_idx": 588}
71
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.02775372564792633, "episode_idx": 598}
72
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.05161993205547333, "episode_idx": 601}
73
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.026043027639389038, "episode_idx": 610}
74
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.026043027639389038, "episode_idx": 610}
75
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.05161993205547333, "episode_idx": 618}
76
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.06213347613811493, "episode_idx": 642}
77
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.009781628847122192, "episode_idx": 657}
78
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.015997350215911865, "episode_idx": 680}
79
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.015997350215911865, "episode_idx": 680}
80
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.048536524176597595, "episode_idx": 689}
81
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.014931261539459229, "episode_idx": 701}
82
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.014931261539459229, "episode_idx": 701}
83
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.05244259536266327, "episode_idx": 723}
84
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.03290490806102753, "episode_idx": 760}
85
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.03043729066848755, "episode_idx": 765}
86
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.03043729066848755, "episode_idx": 765}
87
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.04515865445137024, "episode_idx": 778}
88
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.03043729066848755, "episode_idx": 781}
89
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": -0.02089989185333252, "episode_idx": 783}
90
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": -0.02089989185333252, "episode_idx": 783}
91
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.03863123059272766, "episode_idx": 784}
92
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.016744285821914673, "episode_idx": 789}
93
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.09203149378299713, "episode_idx": 799}
94
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.09203149378299713, "episode_idx": 799}
95
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.05973999202251434, "episode_idx": 807}
96
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": -0.06391112506389618, "episode_idx": 815}
97
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.03427058458328247, "episode_idx": 821}
98
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.055653005838394165, "episode_idx": 850}
99
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.1099025160074234, "episode_idx": 868}
100
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.1099025160074234, "episode_idx": 868}
101
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.03887948393821716, "episode_idx": 885}
102
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.0798504501581192, "episode_idx": 899}
103
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.0798504501581192, "episode_idx": 901}
104
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.004693865776062012, "episode_idx": 908}
105
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.003975778818130493, "episode_idx": 921}
106
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.04933221638202667, "episode_idx": 923}
107
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.060086339712142944, "episode_idx": 947}
108
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.0829857587814331, "episode_idx": 948}
109
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.04815760254859924, "episode_idx": 950}
110
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0031991302967071533, "episode_idx": 954}
111
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.08787737786769867, "episode_idx": 966}
112
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.08787737786769867, "episode_idx": 966}
113
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": -0.040795326232910156, "episode_idx": 973}
114
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.021380990743637085, "episode_idx": 989}
115
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0030046552419662476, "episode_idx": 1026}
116
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.015435397624969482, "episode_idx": 1029}
117
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.012760654091835022, "episode_idx": 1045}
118
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.01062484085559845, "episode_idx": 1049}
119
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.005004033446311951, "episode_idx": 1066}
120
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.02788183093070984, "episode_idx": 1068}
121
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.005004033446311951, "episode_idx": 1080}
122
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0030046552419662476, "episode_idx": 1092}
123
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.012906238436698914, "episode_idx": 1155}
124
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.005424603819847107, "episode_idx": 1157}
125
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": 0.03271615505218506, "episode_idx": 1168}
126
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.010918542742729187, "episode_idx": 1176}
127
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.03942258656024933, "episode_idx": 1184}
128
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": -0.01755937933921814, "episode_idx": 1189}
129
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.00951869785785675, "episode_idx": 1193}
130
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0027276575565338135, "episode_idx": 1204}
131
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.042350634932518005, "episode_idx": 1206}
132
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.002044960856437683, "episode_idx": 1208}
133
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.002044960856437683, "episode_idx": 1208}
134
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.01770801842212677, "episode_idx": 1214}
135
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04573869705200195, "episode_idx": 1220}
136
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": -0.0671481043100357, "episode_idx": 1221}
137
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.017973914742469788, "episode_idx": 1225}
138
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.03043729066848755, "episode_idx": 1226}
139
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.052227914333343506, "episode_idx": 1227}
140
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0029807239770889282, "episode_idx": 1229}
141
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.005533337593078613, "episode_idx": 1234}
142
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.004112556576728821, "episode_idx": 1235}
143
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.01738116145133972, "episode_idx": 1241}
144
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.01738116145133972, "episode_idx": 1241}
145
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.038884758949279785, "episode_idx": 1245}
146
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.0055619776248931885, "episode_idx": 1246}
147
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.042350634932518005, "episode_idx": 1254}
148
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.0007271915674209595, "episode_idx": 1268}
149
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.01770801842212677, "episode_idx": 1278}
150
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.012396752834320068, "episode_idx": 1279}
151
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": -0.012396752834320068, "episode_idx": 1279}
152
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0325581431388855, "episode_idx": 1280}
153
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": 0.04948052763938904, "episode_idx": 1283}
154
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04948052763938904, "episode_idx": 1283}
155
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.005004033446311951, "episode_idx": 1287}
156
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.002768293023109436, "episode_idx": 1288}
157
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.007666334509849548, "episode_idx": 1293}
158
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.03485400974750519, "episode_idx": 1294}
159
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0038469135761260986, "episode_idx": 1296}
160
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": -0.0330449640750885, "episode_idx": 1303}
161
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.0007271915674209595, "episode_idx": 1305}
162
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.03550150990486145, "episode_idx": 1310}
163
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.047868117690086365, "episode_idx": 1311}
164
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.03817342221736908, "episode_idx": 1313}
165
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.04515865445137024, "episode_idx": 1314}
166
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.018898412585258484, "episode_idx": 1320}
167
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": 0.016658708453178406, "episode_idx": 1324}
168
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0021373331546783447, "episode_idx": 1337}
169
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0021373331546783447, "episode_idx": 1337}
170
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.021833211183547974, "episode_idx": 1343}
171
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.03043729066848755, "episode_idx": 1345}
172
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0029807239770889282, "episode_idx": 1349}
173
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.012175574898719788, "episode_idx": 1350}
174
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.01755937933921814, "episode_idx": 1353}
175
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.01755937933921814, "episode_idx": 1353}
176
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.00657692551612854, "episode_idx": 1360}
177
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.00657692551612854, "episode_idx": 1360}
178
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0029807239770889282, "episode_idx": 1361}
179
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.010705456137657166, "episode_idx": 1371}
180
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.02782997488975525, "episode_idx": 1374}
181
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04573869705200195, "episode_idx": 1376}
182
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.014289215207099915, "episode_idx": 1379}
183
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.03330673277378082, "episode_idx": 1380}
184
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.0032100528478622437, "episode_idx": 1383}
185
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.0032100528478622437, "episode_idx": 1383}
186
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.08787737786769867, "episode_idx": 1384}
187
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.08787737786769867, "episode_idx": 1384}
188
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04599034786224365, "episode_idx": 1387}
189
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04599034786224365, "episode_idx": 1387}
190
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.047923505306243896, "episode_idx": 1389}
191
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.08787737786769867, "episode_idx": 1393}
192
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.08787737786769867, "episode_idx": 1393}
193
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0013818293809890747, "episode_idx": 1400}
194
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.08787737786769867, "episode_idx": 1401}
195
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.08787737786769867, "episode_idx": 1401}
196
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.03187800943851471, "episode_idx": 1426}
197
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.004609793424606323, "episode_idx": 1435}
198
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.009926363825798035, "episode_idx": 1440}
199
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.03377266228199005, "episode_idx": 1450}
200
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.0045613497495651245, "episode_idx": 1451}
201
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.0045613497495651245, "episode_idx": 1451}
202
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07993735373020172, "episode_idx": 1483}
203
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07993735373020172, "episode_idx": 1483}
204
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0565386563539505, "episode_idx": 1484}
205
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04532673954963684, "episode_idx": 1493}
206
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.016683846712112427, "episode_idx": 1511}
207
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.053027570247650146, "episode_idx": 1527}
208
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.030305370688438416, "episode_idx": 1556}
209
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.03187800943851471, "episode_idx": 1568}
210
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0017482191324234009, "episode_idx": 1573}
211
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.013293549418449402, "episode_idx": 1579}
212
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.0045613497495651245, "episode_idx": 1588}
213
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.0045613497495651245, "episode_idx": 1588}
214
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.003232702612876892, "episode_idx": 1590}
215
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.002853095531463623, "episode_idx": 1591}
216
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.08665834367275238, "episode_idx": 1593}
217
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.08665834367275238, "episode_idx": 1593}
218
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.0606907457113266, "episode_idx": 1596}
219
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.0606907457113266, "episode_idx": 1596}
220
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.005647405982017517, "episode_idx": 1603}
221
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.030605092644691467, "episode_idx": 1607}
222
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.08787737786769867, "episode_idx": 1613}
223
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.08787737786769867, "episode_idx": 1613}
224
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.03245049715042114, "episode_idx": 1619}
225
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.011172682046890259, "episode_idx": 1651}
226
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.008613958954811096, "episode_idx": 1653}
227
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.009040668606758118, "episode_idx": 1656}
228
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.04343084990978241, "episode_idx": 1663}
229
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.009532496333122253, "episode_idx": 1685}
230
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.008568033576011658, "episode_idx": 1690}
231
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.022988364100456238, "episode_idx": 1698}
232
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.004928573966026306, "episode_idx": 1699}
233
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.08787737786769867, "episode_idx": 1700}
234
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.08787737786769867, "episode_idx": 1700}
235
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0038469135761260986, "episode_idx": 1702}
236
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.019435837864875793, "episode_idx": 1715}
237
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.008380219340324402, "episode_idx": 1723}
238
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.004928573966026306, "episode_idx": 1725}
239
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.060612455010414124, "episode_idx": 1751}
240
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07153545320034027, "episode_idx": 1762}
241
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07153545320034027, "episode_idx": 1762}
242
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0817023366689682, "episode_idx": 1770}
243
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04573869705200195, "episode_idx": 1777}
244
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0027276575565338135, "episode_idx": 1781}
245
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.03524799644947052, "episode_idx": 1789}
246
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04261964559555054, "episode_idx": 1799}
247
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0021393001079559326, "episode_idx": 1800}
248
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.019460976123809814, "episode_idx": 1850}
249
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.06014864146709442, "episode_idx": 1851}
250
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.08271047472953796, "episode_idx": 1861}
251
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.035781651735305786, "episode_idx": 1862}
252
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.016605108976364136, "episode_idx": 1867}
253
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.05556735396385193, "episode_idx": 1871}
254
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.03194144368171692, "episode_idx": 1876}
255
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04224050045013428, "episode_idx": 1883}
256
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.012643590569496155, "episode_idx": 1884}
257
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.015619456768035889, "episode_idx": 1886}
258
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.03318503499031067, "episode_idx": 1896}
259
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04573869705200195, "episode_idx": 1902}
260
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04261964559555054, "episode_idx": 1903}
261
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.002924486994743347, "episode_idx": 1911}
262
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.01874762773513794, "episode_idx": 1912}
263
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": 0.029606997966766357, "episode_idx": 1924}
264
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04224050045013428, "episode_idx": 1926}
265
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.12702801078557968, "episode_idx": 1929}
266
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.02262955904006958, "episode_idx": 1944}
267
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.02931562066078186, "episode_idx": 1956}
268
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04224050045013428, "episode_idx": 1964}
269
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.055653005838394165, "episode_idx": 1990}
270
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.07692860066890717, "episode_idx": 2004}
271
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.009959131479263306, "episode_idx": 2026}
272
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.023197829723358154, "episode_idx": 2032}
273
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.056450843811035156, "episode_idx": 2042}
274
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.011301696300506592, "episode_idx": 2084}
275
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.06656001508235931, "episode_idx": 2112}
276
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.03825581073760986, "episode_idx": 2113}
277
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.07103295624256134, "episode_idx": 2121}
278
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04273587465286255, "episode_idx": 2138}
279
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.05179959535598755, "episode_idx": 2176}
280
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.025090783834457397, "episode_idx": 2182}
281
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.01585310697555542, "episode_idx": 2185}
282
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.014728844165802002, "episode_idx": 2193}
283
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04224050045013428, "episode_idx": 2198}
284
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.01770801842212677, "episode_idx": 2203}
285
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.023754030466079712, "episode_idx": 2214}
286
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04261964559555054, "episode_idx": 2224}
287
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.055334001779556274, "episode_idx": 2227}
288
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.006604120135307312, "episode_idx": 2246}
289
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.012360900640487671, "episode_idx": 2256}
290
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": 0.03648601472377777, "episode_idx": 2277}
291
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.014289215207099915, "episode_idx": 2290}
292
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.005004033446311951, "episode_idx": 2293}
293
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.0182933509349823, "episode_idx": 2294}
294
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.01770801842212677, "episode_idx": 2305}
295
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.029606997966766357, "episode_idx": 2306}
296
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.07600346207618713, "episode_idx": 2307}
297
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.005004033446311951, "episode_idx": 2319}
298
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04261964559555054, "episode_idx": 2321}
299
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.01770801842212677, "episode_idx": 2331}
300
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.010918542742729187, "episode_idx": 2339}
301
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.03142789006233215, "episode_idx": 2384}
302
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.048536524176597595, "episode_idx": 2439}
303
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.03975306451320648, "episode_idx": 2447}
304
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.05312521755695343, "episode_idx": 2457}
305
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07501533627510071, "episode_idx": 2495}
306
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07501533627510071, "episode_idx": 2495}
307
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.03880752623081207, "episode_idx": 2517}
308
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.0419897735118866, "episode_idx": 2535}
309
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.04957498610019684, "episode_idx": 2546}
310
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07501533627510071, "episode_idx": 2576}
311
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07501533627510071, "episode_idx": 2576}
312
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.01563422381877899, "episode_idx": 2723}
313
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.060256898403167725, "episode_idx": 2789}
314
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.01874762773513794, "episode_idx": 2790}
315
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07501533627510071, "episode_idx": 2791}
316
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07501533627510071, "episode_idx": 2791}
317
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07501533627510071, "episode_idx": 2793}
318
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07501533627510071, "episode_idx": 2793}
319
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07023906707763672, "episode_idx": 2810}
320
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.03155544400215149, "episode_idx": 2862}
321
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.05966341495513916, "episode_idx": 2887}
322
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": -0.03858259320259094, "episode_idx": 2889}
323
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.05312521755695343, "episode_idx": 2899}
324
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07023906707763672, "episode_idx": 2910}
325
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07023906707763672, "episode_idx": 2917}
326
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.03627529740333557, "episode_idx": 2944}
327
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.05476678907871246, "episode_idx": 2972}
328
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07023906707763672, "episode_idx": 3078}
329
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04940195381641388, "episode_idx": 3091}
330
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.08250468969345093, "episode_idx": 3145}
331
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.00293925404548645, "episode_idx": 3183}
332
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.014289215207099915, "episode_idx": 3222}
333
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": 0.0688471794128418, "episode_idx": 3266}
334
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.003232702612876892, "episode_idx": 3609}
335
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0030046552419662476, "episode_idx": 3669}
336
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.038337573409080505, "episode_idx": 3715}
337
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": -0.0798504501581192, "episode_idx": 3801}
338
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.014278769493103027, "episode_idx": 3861}
339
+ {"conflict_type": "technical", "template_key": "standard", "quality_delta": -0.058873072266578674, "episode_idx": 3925}
340
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04532673954963684, "episode_idx": 3929}
341
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.07065899670124054, "episode_idx": 3940}
342
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.0058425962924957275, "episode_idx": 4007}
343
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.017700672149658203, "episode_idx": 4017}
344
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.017700672149658203, "episode_idx": 4140}
345
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.002397477626800537, "episode_idx": 4319}
346
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.01770801842212677, "episode_idx": 4390}
347
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0018948465585708618, "episode_idx": 4463}
348
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07993735373020172, "episode_idx": 4552}
349
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07993735373020172, "episode_idx": 4552}
350
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04573869705200195, "episode_idx": 4622}
351
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.04573869705200195, "episode_idx": 4632}
352
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.0666174590587616, "episode_idx": 4705}
353
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.05178782343864441, "episode_idx": 4968}
354
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.035536155104637146, "episode_idx": 4988}
355
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.009939327836036682, "episode_idx": 5097}
356
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.03212524950504303, "episode_idx": 5166}
357
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.000596463680267334, "episode_idx": 5273}
358
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.000596463680267334, "episode_idx": 5275}
359
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.014289215207099915, "episode_idx": 5479}
360
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.009817525744438171, "episode_idx": 5530}
361
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.0844281017780304, "episode_idx": 5806}
362
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.0844281017780304, "episode_idx": 5806}
363
+ {"conflict_type": "technical", "template_key": "synthesise", "quality_delta": 0.0419897735118866, "episode_idx": 5918}
364
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.045914486050605774, "episode_idx": 6035}
365
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.027218639850616455, "episode_idx": 6740}
366
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.010060101747512817, "episode_idx": 6835}
367
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -3.872811794281006e-05, "episode_idx": 7253}
368
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.05179959535598755, "episode_idx": 7320}
369
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.042350634932518005, "episode_idx": 7433}
370
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.009939327836036682, "episode_idx": 7996}
371
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.0012574940919876099, "episode_idx": 8185}
372
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.014289215207099915, "episode_idx": 8282}
373
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07501533627510071, "episode_idx": 8724}
374
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.07501533627510071, "episode_idx": 8724}
375
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.042350634932518005, "episode_idx": 9224}
376
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.009939327836036682, "episode_idx": 9322}
377
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.05161993205547333, "episode_idx": 9431}
378
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.009939327836036682, "episode_idx": 9439}
379
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": -0.010060101747512817, "episode_idx": 10312}
380
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.042350634932518005, "episode_idx": 10369}
381
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.014278769493103027, "episode_idx": 10665}
382
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.008568033576011658, "episode_idx": 11248}
383
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.014278769493103027, "episode_idx": 12020}
384
+ {"conflict_type": "technical", "template_key": "defer_to_a", "quality_delta": 0.03880752623081207, "episode_idx": 13306}
data/specialist_memory.json ADDED
The diff for this file is too large to render. See raw diff
 
reward_curve.json CHANGED
@@ -1 +1 @@
1
- {"episodes": [0, 67, 134, 201, 268, 335, 402, 469, 536, 603, 670, 737, 804, 871, 938, 1005, 1072, 1139, 1206, 1273, 1340, 1407, 1474, 1541, 1608, 1675, 1742, 1809, 1876, 1943, 2010, 2077, 2144, 2211, 2278, 2345, 2412, 2479, 2546, 2613, 2680, 2747, 2814, 2881, 2948, 3015, 3082, 3149, 3216, 3283, 3350, 3417, 3484, 3551, 3618, 3685, 3752, 3819, 3886, 3953, 4020, 4087, 4154, 4221, 4288, 4355, 4422, 4489, 4556, 4623, 4690, 4757, 4824, 4891, 4958, 5025, 5092, 5159, 5226, 5293, 5360, 5427, 5494, 5561, 5628, 5695, 5762, 5829, 5896, 5963, 6030, 6097, 6164, 6231, 6298, 6365, 6432, 6499, 6566, 6633, 6700, 6767, 6834, 6901, 6968, 7035, 7102, 7169, 7236, 7303, 7370, 7437, 7504, 7571, 7638, 7705, 7772, 7839, 7906, 7973, 8040, 8107, 8174, 8241, 8308, 8375, 8442, 8509, 8576, 8643, 8710, 8777, 8844, 8911, 8978, 9045, 9112, 9179, 9246, 9313, 9380, 9447, 9514, 9581, 9648, 9715, 9782, 9849, 9916, 9983, 10050, 10117, 10184, 10251, 10318, 10385, 10452, 10519, 10586, 10653, 10720, 10787, 10854, 10921, 10988, 11055, 11122, 11189, 11256, 11323, 11390, 11457, 11524, 11591, 11658, 11725, 11792, 11859, 11926, 11993, 12060, 12127, 12194, 12261, 12328, 12395, 12462, 12529, 12596, 12663, 12730, 12797, 12864, 12931, 12998, 13065, 13132, 13199, 13266, 13333, 13400], "mean_rewards": [7.869385480880737, -0.4680378864902784, -0.48984615507501145, -0.5551410619040379, -0.4882915579094153, -0.4724334484614831, -0.4706858817982504, -0.49536512678172046, -0.488181437265121, -0.49218673040220723, -0.4815726703055059, -0.47754327278345865, -0.43784233579173726, -0.3844110888298712, -0.3559347061479154, -0.30412048705953304, -0.266127535926422, -0.1874622176040984, -0.14371091075008735, -0.02650309386219652, 0.03699350765202239, 0.11568798260192298, 0.1813001428686437, 0.234962336935236, 0.29998476978480104, 0.3494067042688368, 0.4188010052682893, 0.4982103691668488, 0.5493174741537327, 0.5519611505521989, 0.5843701104372269, 0.6528626296241695, 0.6805431383403456, 0.7341929864990648, 0.7834733703783795, 0.8403997419894993, 0.8835701591176855, 0.9107975078377638, 0.9910712995973492, 1.0697907954830996, 1.1305848434251078, 1.1324089907647075, 1.172262050650997, 1.2235095935128422, 1.2517143468368366, 1.2724046870820789, 1.3005815128812432, 1.3231664394510776, 1.3306386821592848, 1.3325763804993283, 1.3586343029331016, 1.40881654574436, 1.4413863711607908, 1.4488289435588133, 1.4570241300250582, 1.4681397577771262, 1.499341548685888, 1.5262605464494776, 1.5235170595468455, 1.513340130324477, 1.5307294156028504, 1.5475655648487667, 1.5575386175265262, 1.5641262251109376, 1.5792971486511564, 1.5843108974196574, 1.6019751663429598, 1.5715943494819948, 1.5931392893901157, 1.6346125779210805, 1.6342209757898392, 1.617359425582728, 1.6516810876938788, 1.6577659129061755, 1.7004437399976031, 1.7311268771364081, 1.739649844424992, 1.7626731381482645, 1.7597383664972204, 1.7712974552394458, 1.7726923163304273, 1.7791328078125108, 1.7655685633567155, 1.7530806338330702, 1.7465425723497727, 1.7384169118655943, 1.7192243574163946, 1.7400900246355464, 1.7620760225301118, 1.7782208430467326, 1.7714304287994007, 1.763625342698084, 1.7601928197433543, 1.7576644544521427, 1.7743939839113496, 1.769043635923286, 1.810667814309779, 1.8216552026374302, 1.8478464421820198, 1.8269981258918702, 1.8744575704903192, 1.8827178977458279, 1.866869720512893, 1.8740134621361253, 1.8397641184360727, 1.8410966690533401, 1.8426638447435835, 1.8283637543299847, 1.8131030868182625, 1.8027979614466207, 1.7552007339833966, 1.7917554398061926, 1.809832562660328, 1.833266973746071, 1.8741590295668997, 1.880611007449434, 1.8882735193283657, 1.8773927914049378, 1.865357443877396, 1.8789978755369139, 1.8915645213065935, 1.8952674455164622, 1.9155843285400243, 1.9085990070265584, 1.90080054640393, 1.9256198932583044, 1.9366444012710071, 1.9493767221422222, 1.9839627611148725, 1.9764564716675697, 2.002629030457077, 1.9820295862487651, 1.9856562615653577, 1.9859819096593474, 1.9698493608136363, 1.947345323234913, 1.9319047969357122, 1.9212369738192396, 1.915823275121511, 1.9219170818084552, 1.9338193936557564, 1.9390760462537227, 1.9549254322946799, 1.9771323770544638, 1.968752693505264, 1.9706327813298576, 1.979237121560172, 1.994255026763767, 1.9968812663644582, 2.0071281646386785, 1.9723110195435158, 1.9356860965843745, 1.9251148387156718, 1.9225462930269368, 1.9381719721641275, 1.9472862571208334, 1.9549440001736955, 1.943592506697141, 1.9607074977289554, 1.966358552660538, 1.9941627379199158, 2.017277606701966, 2.0028778048013227, 1.9798156342786526, 1.9463988085238573, 1.961345897422039, 1.9355557291343022, 1.9354226221486377, 1.9281403603033271, 1.9492224078499671, 1.9400939026771604, 1.957794354431826, 1.9947251642922874, 2.016812087167117, 2.0479436611309385, 2.025213644837225, 2.0251720844453738, 1.9995202234105298, 1.9965415502759112, 1.974851788492829, 1.9879056021842796, 1.9780626381897102, 1.9584395736040148, 1.9372385981398457, 1.9209191222658515, 1.9350956156822143, 1.960611907474231, 1.995446871384047, 1.9792517610698095, 1.9596557471280296, 1.9577739956190012, 1.9561633419888538, 1.967589806283026, 1.9925558521534272, 2.0040653944348117, 2.0273019113921604, 2.0106458478063964, 2.0137654255044515, 2.0376576196867973, 2.0654571994618025, 2.042709940440199], "raw_rewards": [7.869385480880737, -1.4724462032318115, -0.9430539831519127, -2.300199344754219, -0.8089542239904404, -1.4609763622283936, 1.1505420207977295, -1.4753345251083374, -1.0061301589012146, 1.4091460406780243, -1.7102899551391602, 0.8560748100280762, -1.2121876776218414, 0.8951746374368668, -0.3283916711807251, 1.4731740355491638, -1.5742558240890503, 1.3293038997799158, -1.9147660434246063, 2.114831119775772, 1.0735967457294464, 1.6181618869304657, 0.5473835095763206, -0.971550777554512, 2.4433728456497192, -0.5766929388046265, 0.6217614114284515, 0.1080636978149414, -1.6272716522216797, 0.801971822977066, 1.0068292915821075, 3.228931248188019, 0.9021086767315865, -0.1995646357536316, -1.6238511800765991, -1.621373176574707, 2.2451024651527405, 2.389804482460022, 0.174017071723938, -0.6571964621543884, 2.4821357131004333, 0.0889243632555008, -1.7433454543352127, 3.609505534172058, 2.2524542808532715, 2.5349577367305756, 1.3504652380943298, 2.934185355901718, 2.9824480414390564, -0.17685255408287048, 2.9204043447971344, 2.178996592760086, 1.329479992389679, 0.6736606955528259, 0.8348989896476269, 3.0606049299240112, 1.8705639839172363, 2.923104405403137, 1.4114362634718418, -0.1116487979888916, 2.0751985609531403, -1.1050611436367035, 0.5921751260757446, 0.9946736395359039, 1.8352702856063843, 0.796254113316536, 1.7449798882007599, 0.7606222331523895, 0.12062180042266846, 1.7189838290214539, 3.541672945022583, 1.5339390933513641, 2.767375946044922, 2.2297381162643433, -0.27031809091567993, 1.6640383005142212, 0.8177437596023083, 1.9806551933288574, 1.2750858068466187, 1.5373672842979431, 1.3882066011428833, 0.6665303558111191, 1.0681654214859009, 1.0687440931797028, 2.2572388648986816, 2.4432771801948547, 1.3739816546440125, 2.555147707462311, 1.5532113909721375, 1.5525203347206116, 1.0045947134494781, 2.4867050647735596, 2.0954394340515137, 1.6426711678504944, 2.5995242595672607, 0.4808852504938841, 1.495392918586731, 0.6842523217201233, 3.7635045051574707, 2.212641790509224, 3.5520986318588257, 1.2874490916728973, 1.5225331783294678, 2.5105870962142944, 2.647844910621643, 0.5050036013126373, 2.310435175895691, 2.4384663701057434, 2.8605822324752808, 2.404734432697296, 0.4821899086236954, 1.2641501389443874, 1.6644573211669922, 1.6970094442367554, 1.0590541064739227, -0.8185135126113892, 2.923989415168762, 0.46643319725990295, 1.3798049688339233, 1.27825129032135, 3.6883862614631653, 1.7710249423980713, 3.0534738898277283, 2.207607924938202, 1.0182976424694061, 2.965312957763672, 2.836131751537323, -1.1759849786758423, 0.8884927332401276, 1.3794639706611633, 1.0612505674362183, 1.510947585105896, 1.0037859827280045, 2.1822192072868347, 2.5573002099990845, 0.9873855113983154, 0.9950265288352966, 1.686583399772644, 2.587091416120529, 1.0622715055942535, 2.9554337859153748, 1.7556167542934418, 2.477279305458069, 1.6855851411819458, 1.325159728527069, 3.6202566027641296, 0.9852837026119232, 2.3055906891822815, 1.8989630937576294, 3.8421117663383484, 3.716716766357422, 3.286076307296753, 2.095924496650696, 1.8744736313819885, 1.2755839824676514, 2.1855685114860535, 2.951150119304657, -0.03335070610046387, 2.9794004559516907, 1.974601686000824, 1.5195317268371582, 2.607335329055786, 2.901742935180664, 2.9537888169288635, 2.556188404560089, 1.6433136463165283, 1.981132984161377, 0.7490701808128506, 2.8644319772720337, 1.9312551617622375, 3.218564510345459, 1.717534363269806, 1.7507363855838776, 0.8022308275103569, 0.9781382381916046, 0.9786528944969177, 2.802109658718109, 2.290560722351074, 0.803216639906168, 3.095746695995331, 3.292268693447113, 2.4694343209266663, 1.7080549597740173, 2.6128552556037903, 1.0149105489253998, 2.6496033668518066, 1.5982393622398376, 3.1078230142593384, 2.3775070905685425, 1.684832751750946, 2.6528014540672302, 2.9457955360412598, 2.1817689538002014, 1.707306683063507, 2.385865569114685, 2.928332030773163, 1.6798464059829712, 3.398195207118988, 1.6161009706556797, 4.029117822647095, 1.5490476191043854], "step": 30000}
 
1
+ {"episodes": [0, 67, 134, 201, 268, 335, 402, 469, 536, 603, 670, 737, 804, 871, 938, 1005, 1072, 1139, 1206, 1273, 1340, 1407, 1474, 1541, 1608, 1675, 1742, 1809, 1876, 1943, 2010, 2077, 2144, 2211, 2278, 2345, 2412, 2479, 2546, 2613, 2680, 2747, 2814, 2881, 2948, 3015, 3082, 3149, 3216, 3283, 3350, 3417, 3484, 3551, 3618, 3685, 3752, 3819, 3886, 3953, 4020, 4087, 4154, 4221, 4288, 4355, 4422, 4489, 4556, 4623, 4690, 4757, 4824, 4891, 4958, 5025, 5092, 5159, 5226, 5293, 5360, 5427, 5494, 5561, 5628, 5695, 5762, 5829, 5896, 5963, 6030, 6097, 6164, 6231, 6298, 6365, 6432, 6499, 6566, 6633, 6700, 6767, 6834, 6901, 6968, 7035, 7102, 7169, 7236, 7303, 7370, 7437, 7504, 7571, 7638, 7705, 7772, 7839, 7906, 7973, 8040, 8107, 8174, 8241, 8308, 8375, 8442, 8509, 8576, 8643, 8710, 8777, 8844, 8911, 8978, 9045, 9112, 9179, 9246, 9313, 9380, 9447, 9514, 9581, 9648, 9715, 9782, 9849, 9916, 9983, 10050, 10117, 10184, 10251, 10318, 10385, 10452, 10519, 10586, 10653, 10720, 10787, 10854, 10921, 10988, 11055, 11122, 11189, 11256, 11323, 11390, 11457, 11524, 11591, 11658, 11725, 11792, 11859, 11926, 11993, 12060, 12127, 12194, 12261, 12328, 12395, 12462, 12529, 12596, 12663, 12730, 12797, 12864, 12931, 12998, 13065, 13132, 13199, 13266, 13333, 13400, 13467], "mean_rewards": [7.869385480880737, -0.4680378864902784, -0.48984615507501145, -0.5551410619040379, -0.4882915579094153, -0.4724334484614831, -0.4706858817982504, -0.49536512678172046, -0.488181437265121, -0.49218673040220723, -0.4815726703055059, -0.4820088524676782, -0.43784926100558375, -0.38558788626885354, -0.3533458443251239, -0.3040934353599124, -0.26831979214033574, -0.19140849363449494, -0.14203320211143491, -0.03154456203850192, 0.036356829438342037, 0.10923927782606911, 0.18537376434205202, 0.23818426830707837, 0.2971860210863843, 0.351879875832877, 0.41131269029737827, 0.49316205722733814, 0.54418244560461, 0.5529096998612051, 0.5803396979899199, 0.647491346514733, 0.6767948179212062, 0.7278867457176142, 0.7801977058292352, 0.8324114113203805, 0.8846536156650867, 0.9135522807000074, 0.9906498014824725, 1.0747783365352257, 1.118064828834938, 1.1254296454959651, 1.1729068560170468, 1.223296715608557, 1.250830467714188, 1.2658592661535932, 1.3026064717935804, 1.3210587293828564, 1.3291228487317284, 1.3327255037611838, 1.3617869722608757, 1.4155327298567029, 1.43804301345443, 1.4533835403824795, 1.4490587244572413, 1.4650314738802017, 1.4970641783053438, 1.5264675998406834, 1.5249301680553555, 1.5104940346199014, 1.5259575240262075, 1.5392910738888663, 1.5554828937112133, 1.5631932486042976, 1.577678209906162, 1.578992944977183, 1.6025471957099515, 1.5807723147838604, 1.5952618680505644, 1.6292361341937072, 1.6326987860754143, 1.6245606879480392, 1.6551997169478658, 1.6609219776152042, 1.6956165906986145, 1.7281981333328353, 1.7382102399385817, 1.7609851391969418, 1.7657828451398392, 1.7754888206320893, 1.7645542004733432, 1.7736322304512135, 1.7687026667687216, 1.7580264367949372, 1.7511381573787308, 1.738839277673473, 1.722614758535346, 1.736745614141461, 1.7516905580270015, 1.7799245542025521, 1.7700678259391585, 1.7663207697615921, 1.7562163755505256, 1.7587610971747853, 1.7777937432416644, 1.7704596316888854, 1.8123356010955924, 1.8279217910185117, 1.8466502024516873, 1.8298634645036476, 1.864942445150443, 1.8738156217091924, 1.866000112849446, 1.8781902273140842, 1.8416169193525835, 1.8387443213458785, 1.8485462165543736, 1.8278536795119362, 1.8172942939169872, 1.8037021071630313, 1.7578254819032049, 1.789927663865958, 1.8083126330250383, 1.8363163345167517, 1.8767271982310745, 1.8823451488214322, 1.8882187248528846, 1.879882797724712, 1.8682739835334294, 1.8768755783047664, 1.8853264416580362, 1.8954450143514918, 1.9106359317722665, 1.9006636210578318, 1.8993877949828675, 1.9237823593464614, 1.9344475316007566, 1.9503340135305085, 1.9860419661475228, 1.9742594653096717, 2.001614794778903, 1.9832609987142256, 1.9856477235292567, 1.9926783719954344, 1.9750121485619383, 1.9528438657518075, 1.9334705030714798, 1.9176346147756205, 1.9197328341457625, 1.929320812971638, 1.935611099781306, 1.9301533763026026, 1.9524929260035577, 1.9774023529251636, 1.9670411776322891, 1.9650732083037827, 1.9756527485480824, 1.9960569345849135, 1.9994612800762286, 2.0023671949464004, 1.968471810134468, 1.9351308788716286, 1.9281577662479867, 1.9180797577689113, 1.941868303399712, 1.9551983473480623, 1.9582474042363658, 1.9446796976185323, 1.9599162018053136, 1.9677243051538726, 1.9905415299682818, 2.0126337819571773, 2.0055850803634727, 1.9728688634663787, 1.942376174849831, 1.961291219857542, 1.9382520422395841, 1.9331532626002175, 1.9315321168749136, 1.9510658689675868, 1.93959757961361, 1.9583952448716349, 1.9957628746242215, 2.0171623558352496, 2.045212839087822, 2.027753655145466, 2.023227911412426, 2.0044681133162108, 1.9950296741726627, 1.9736256399910699, 1.9867524237810816, 1.9778640382239483, 1.9617218394162603, 1.9373436287414156, 1.9262697115193839, 1.9367491794142053, 1.9615602505914702, 1.9952850935889883, 1.9757408594996002, 1.9619378888624635, 1.9616936761638262, 1.9569105818823662, 1.966847003359166, 1.995661675253198, 2.011519161810991, 2.030814020499653, 2.015335946653363, 2.010846167103975, 2.036694835887966, 2.063149935127053, 2.0468815370260613, 2.0691288812374875]}
reward_curve.png CHANGED
spindleflow_model.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8f1c845438f02b1fb1d229221a2b378411ddafb0186627e63497ec18fc5e0948
3
+ size 143819555
training_log.txt CHANGED
@@ -566,3 +566,13 @@
566
  [08:39:04] Ep 13375 | reward +0.413 | Phase 3/3 | Rolling mean: 2.023 / β€” | Episodes in phase: 11150
567
  [08:39:05] Ep 13400 | reward +3.291 | Phase 3/3 | Rolling mean: 1.999 / β€” | Episodes in phase: 11175
568
  [08:39:19] Periodic save at step 30,000 ...
 
 
 
 
 
 
 
 
 
 
 
566
  [08:39:04] Ep 13375 | reward +0.413 | Phase 3/3 | Rolling mean: 2.023 / β€” | Episodes in phase: 11150
567
  [08:39:05] Ep 13400 | reward +3.291 | Phase 3/3 | Rolling mean: 1.999 / β€” | Episodes in phase: 11175
568
  [08:39:19] Periodic save at step 30,000 ...
569
+ [08:39:22] Periodic push done β€” 5 files at step 30,000
570
+ [08:39:22] Ep 13425 | reward +1.596 | Phase 3/3 | Rolling mean: 1.962 / β€” | Episodes in phase: 11200
571
+ [08:39:23] Ep 13450 | reward +3.219 | Phase 3/3 | Rolling mean: 1.964 / β€” | Episodes in phase: 11225
572
+ [08:39:25] Ep 13475 | reward +3.019 | Phase 3/3 | Rolling mean: 2.014 / β€” | Episodes in phase: 11250
573
+ [08:39:26] Ep 13500 | reward +0.549 | Phase 3/3 | Rolling mean: 2.003 / β€” | Episodes in phase: 11275
574
+ [08:39:40] Ep 13525 | reward +3.228 | Phase 3/3 | Rolling mean: 2.011 / β€” | Episodes in phase: 11300
575
+ [08:39:42] Model saved β€” 13526 episodes completed.
576
+ [08:39:42] Final curriculum: Phase 3/3 | Rolling mean: 2.007 / β€” | Episodes in phase: 11301
577
+ [08:39:43] Reward curve saved.
578
+ [08:39:43] Pushing to https://huggingface.co/garvitsachdeva/spindleflow-rl ...
vec_normalize.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a0580efba59194f97764849fae501e3e85a59ecb4a5edfad2385a3962b50464
3
+ size 166596