[2024-12-16 13:49:10,579][mode.evaluation.multistep_sequences][INFO] - Start generating evaluation sequences.
[2024-12-16 13:49:26,551][mode.evaluation.multistep_sequences][INFO] - Done generating evaluation sequences.
  0%|                                                                                                                                 | 0/1000 [00:00<?, ?it/s]




























































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































1/5 : 95.9% | 2/5 : 88.6% | 3/5 : 81.2% | 4/5 : 72.7% | 5/5 : 65.2% | Average: 4.0 |: 100%|██████████████████████████████| 1000/1000 [2:54:28<00:00, 10.47s/it]
Results for Epoch 0:
Average successful sequence length: 4.036
Success rates for i instructions in a row:
1: 95.9%
2: 88.6%
3: 81.2%
4: 72.7%
5: 65.2%
rotate_blue_block_right: 71 / 74 |  SR: 95.9%
move_slider_right: 282 / 282 |  SR: 100.0%
turn_off_led: 162 / 164 |  SR: 98.8%
push_into_drawer: 98 / 121 |  SR: 81.0%
lift_blue_block_drawer: 17 / 19 |  SR: 89.5%
place_in_slider: 281 / 341 |  SR: 82.4%
close_drawer: 204 / 204 |  SR: 100.0%
lift_pink_block_slider: 128 / 142 |  SR: 90.1%
open_drawer: 342 / 342 |  SR: 100.0%
rotate_red_block_right: 72 / 73 |  SR: 98.6%
lift_red_block_table: 169 / 175 |  SR: 96.6%
lift_pink_block_table: 150 / 164 |  SR: 91.5%
turn_on_lightbulb: 160 / 183 |  SR: 87.4%
rotate_blue_block_left: 65 / 65 |  SR: 100.0%
turn_on_led: 169 / 172 |  SR: 98.3%
stack_block: 136 / 187 |  SR: 72.7%
push_red_block_left: 74 / 77 |  SR: 96.1%
lift_blue_block_table: 178 / 184 |  SR: 96.7%
place_in_drawer: 173 / 175 |  SR: 98.9%
turn_off_lightbulb: 125 / 136 |  SR: 91.9%
move_slider_left: 239 / 241 |  SR: 99.2%
rotate_red_block_left: 60 / 60 |  SR: 100.0%
lift_red_block_slider: 117 / 134 |  SR: 87.3%
push_pink_block_left: 68 / 75 |  SR: 90.7%
lift_blue_block_slider: 111 / 131 |  SR: 84.7%
rotate_pink_block_right: 66 / 68 |  SR: 97.1%
unstack_block: 51 / 52 |  SR: 98.1%
push_blue_block_right: 44 / 63 |  SR: 69.8%
push_red_block_right: 46 / 67 |  SR: 68.7%
rotate_pink_block_left: 54 / 54 |  SR: 100.0%
push_blue_block_left: 62 / 66 |  SR: 93.9%
lift_pink_block_drawer: 13 / 14 |  SR: 92.9%
push_pink_block_right: 32 / 62 |  SR: 51.6%
lift_red_block_drawer: 17 / 17 |  SR: 100.0%
Best model: epoch 0 with average sequences length of 4.036