RMBench-VLM / validation_metrics.json
jellerode's picture
Add files using upload-large-folder tool
6a28b6f verified
Raw History Blame Contribute Delete
16.9 kB
{
"precision": 0.9406,
"recall": 0.9965,
"f1": 0.9677,
"segment": {
"1": {
"hit": 1.0,
"acc": 1.0
},
"2": {
"hit": 0.9821,
"acc": 0.9464
},
"3": {
"hit": 1.0,
"acc": 0.9032
},
"4": {
"hit": 1.0,
"acc": 0.9583
},
"5": {
"hit": 1.0,
"acc": 1.0
},
"6": {
"hit": 1.0,
"acc": 1.0
},
"7": {
"hit": 1.0,
"acc": 0.9
},
"8": {
"hit": 1.0,
"acc": 1.0
},
"9": {
"hit": 1.0,
"acc": 1.0
},
"10": {
"hit": 1.0,
"acc": 1.0
},
"11": {
"hit": 1.0,
"acc": 1.0
},
"12": {
"hit": 1.0,
"acc": 1.0
},
"overall": {
"hit": 0.9965,
"acc": 0.972
}
},
"segment_acc": 0.972,
"non_key_acc": 0.8889,
"overall_score": 0.9305,
"exact_rate": 0.972,
"segment_positive": 0.9965,
"non_key_frame_acc": 0.9993,
"n_episodes": 90,
"n_windows": 286,
"n_segments": 286,
"n_nonkey_episodes": 90,
"missed": 1,
"excess_windows": 7,
"spurious_total": 11,
"spurious_per_ep": 0.122,
"total_yes_pred": 303,
"perfect_episodes": 74,
"key_interval_hit_rates": [
1.0,
0.9821,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0
],
"key_interval_acc_rates": [
1.0,
0.9464,
0.9032,
0.9583,
1.0,
1.0,
0.9,
1.0,
1.0,
1.0,
1.0,
1.0
],
"key_interval_hit_counts": [
90,
55,
31,
24,
22,
21,
10,
10,
10,
7,
3,
2
],
"key_interval_acc_counts": [
90,
53,
28,
23,
22,
21,
9,
10,
10,
7,
3,
2
],
"key_interval_episode_counts": [
90,
56,
31,
24,
22,
21,
10,
10,
10,
7,
3,
2
],
"trajectory_count": 90,
"overall": {
"precision": 0.9406,
"recall": 0.9965,
"f1": 0.9677,
"segment": {
"1": {
"hit": 1.0,
"acc": 1.0
},
"2": {
"hit": 0.9821,
"acc": 0.9464
},
"3": {
"hit": 1.0,
"acc": 0.9032
},
"4": {
"hit": 1.0,
"acc": 0.9583
},
"5": {
"hit": 1.0,
"acc": 1.0
},
"6": {
"hit": 1.0,
"acc": 1.0
},
"7": {
"hit": 1.0,
"acc": 0.9
},
"8": {
"hit": 1.0,
"acc": 1.0
},
"9": {
"hit": 1.0,
"acc": 1.0
},
"10": {
"hit": 1.0,
"acc": 1.0
},
"11": {
"hit": 1.0,
"acc": 1.0
},
"12": {
"hit": 1.0,
"acc": 1.0
},
"overall": {
"hit": 0.9965,
"acc": 0.972
}
},
"segment_acc": 0.972,
"non_key_acc": 0.8889,
"overall_score": 0.9305,
"exact_rate": 0.972,
"segment_positive": 0.9965,
"non_key_frame_acc": 0.9993,
"n_episodes": 90,
"n_windows": 286,
"n_segments": 286,
"n_nonkey_episodes": 90,
"missed": 1,
"excess_windows": 7,
"spurious_total": 11,
"spurious_per_ep": 0.122,
"total_yes_pred": 303,
"perfect_episodes": 74,
"key_interval_hit_rates": [
1.0,
0.9821,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0
],
"key_interval_acc_rates": [
1.0,
0.9464,
0.9032,
0.9583,
1.0,
1.0,
0.9,
1.0,
1.0,
1.0,
1.0,
1.0
],
"key_interval_hit_counts": [
90,
55,
31,
24,
22,
21,
10,
10,
10,
7,
3,
2
],
"key_interval_acc_counts": [
90,
53,
28,
23,
22,
21,
9,
10,
10,
7,
3,
2
],
"key_interval_episode_counts": [
90,
56,
31,
24,
22,
21,
10,
10,
10,
7,
3,
2
],
"trajectory_count": 90
},
"by_task": {
"battery_try": {
"precision": 0.9524,
"recall": 1.0,
"f1": 0.9756,
"segment": {
"1": {
"hit": 1.0,
"acc": 1.0
},
"2": {
"hit": 1.0,
"acc": 1.0
},
"3": {
"hit": 1.0,
"acc": 1.0
},
"overall": {
"hit": 1.0,
"acc": 1.0
}
},
"segment_acc": 1.0,
"non_key_acc": 0.9,
"overall_score": 0.95,
"exact_rate": 1.0,
"segment_positive": 1.0,
"non_key_frame_acc": 0.9995,
"n_episodes": 10,
"n_windows": 20,
"n_segments": 20,
"n_nonkey_episodes": 10,
"missed": 0,
"excess_windows": 0,
"spurious_total": 1,
"spurious_per_ep": 0.1,
"total_yes_pred": 21,
"perfect_episodes": 9,
"key_interval_hit_rates": [
1.0,
1.0,
1.0
],
"key_interval_acc_rates": [
1.0,
1.0,
1.0
],
"key_interval_hit_counts": [
10,
6,
4
],
"key_interval_acc_counts": [
10,
6,
4
],
"key_interval_episode_counts": [
10,
6,
4
],
"trajectory_count": 10
},
"blocks_ranking_try": {
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"segment": {
"1": {
"hit": 1.0,
"acc": 1.0
},
"2": {
"hit": 1.0,
"acc": 1.0
},
"3": {
"hit": 1.0,
"acc": 1.0
},
"4": {
"hit": 1.0,
"acc": 1.0
},
"5": {
"hit": 1.0,
"acc": 1.0
},
"6": {
"hit": 1.0,
"acc": 1.0
},
"overall": {
"hit": 1.0,
"acc": 1.0
}
},
"segment_acc": 1.0,
"non_key_acc": 1.0,
"overall_score": 1.0,
"exact_rate": 1.0,
"segment_positive": 1.0,
"non_key_frame_acc": 1.0,
"n_episodes": 10,
"n_windows": 34,
"n_segments": 34,
"n_nonkey_episodes": 10,
"missed": 0,
"excess_windows": 0,
"spurious_total": 0,
"spurious_per_ep": 0.0,
"total_yes_pred": 34,
"perfect_episodes": 10,
"key_interval_hit_rates": [
1.0,
1.0,
1.0,
1.0,
1.0,
1.0
],
"key_interval_acc_rates": [
1.0,
1.0,
1.0,
1.0,
1.0,
1.0
],
"key_interval_hit_counts": [
10,
10,
7,
4,
2,
1
],
"key_interval_acc_counts": [
10,
10,
7,
4,
2,
1
],
"key_interval_episode_counts": [
10,
10,
7,
4,
2,
1
],
"trajectory_count": 10
},
"cover_blocks": {
"precision": 0.8824,
"recall": 1.0,
"f1": 0.9375,
"segment": {
"1": {
"hit": 1.0,
"acc": 1.0
},
"2": {
"hit": 1.0,
"acc": 1.0
},
"3": {
"hit": 1.0,
"acc": 0.7
},
"4": {
"hit": 1.0,
"acc": 0.9
},
"5": {
"hit": 1.0,
"acc": 1.0
},
"6": {
"hit": 1.0,
"acc": 1.0
},
"overall": {
"hit": 1.0,
"acc": 0.9333
}
},
"segment_acc": 0.9333,
"non_key_acc": 0.7,
"overall_score": 0.8167,
"exact_rate": 0.9333,
"segment_positive": 1.0,
"non_key_frame_acc": 0.9987,
"n_episodes": 10,
"n_windows": 60,
"n_segments": 60,
"n_nonkey_episodes": 10,
"missed": 0,
"excess_windows": 4,
"spurious_total": 4,
"spurious_per_ep": 0.4,
"total_yes_pred": 68,
"perfect_episodes": 4,
"key_interval_hit_rates": [
1.0,
1.0,
1.0,
1.0,
1.0,
1.0
],
"key_interval_acc_rates": [
1.0,
1.0,
0.7,
0.9,
1.0,
1.0
],
"key_interval_hit_counts": [
10,
10,
10,
10,
10,
10
],
"key_interval_acc_counts": [
10,
10,
7,
9,
10,
10
],
"key_interval_episode_counts": [
10,
10,
10,
10,
10,
10
],
"trajectory_count": 10
},
"observe_and_pickup": {
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"segment": {
"1": {
"hit": 1.0,
"acc": 1.0
},
"overall": {
"hit": 1.0,
"acc": 1.0
}
},
"segment_acc": 1.0,
"non_key_acc": 1.0,
"overall_score": 1.0,
"exact_rate": 1.0,
"segment_positive": 1.0,
"non_key_frame_acc": 1.0,
"n_episodes": 10,
"n_windows": 10,
"n_segments": 10,
"n_nonkey_episodes": 10,
"missed": 0,
"excess_windows": 0,
"spurious_total": 0,
"spurious_per_ep": 0.0,
"total_yes_pred": 10,
"perfect_episodes": 10,
"key_interval_hit_rates": [
1.0
],
"key_interval_acc_rates": [
1.0
],
"key_interval_hit_counts": [
10
],
"key_interval_acc_counts": [
10
],
"key_interval_episode_counts": [
10
],
"trajectory_count": 10
},
"press_button": {
"precision": 0.9903,
"recall": 1.0,
"f1": 0.9951,
"segment": {
"1": {
"hit": 1.0,
"acc": 1.0
},
"2": {
"hit": 1.0,
"acc": 1.0
},
"3": {
"hit": 1.0,
"acc": 1.0
},
"4": {
"hit": 1.0,
"acc": 1.0
},
"5": {
"hit": 1.0,
"acc": 1.0
},
"6": {
"hit": 1.0,
"acc": 1.0
},
"7": {
"hit": 1.0,
"acc": 0.9
},
"8": {
"hit": 1.0,
"acc": 1.0
},
"9": {
"hit": 1.0,
"acc": 1.0
},
"10": {
"hit": 1.0,
"acc": 1.0
},
"11": {
"hit": 1.0,
"acc": 1.0
},
"12": {
"hit": 1.0,
"acc": 1.0
},
"overall": {
"hit": 1.0,
"acc": 0.9902
}
},
"segment_acc": 0.9902,
"non_key_acc": 1.0,
"overall_score": 0.9951,
"exact_rate": 0.9902,
"segment_positive": 1.0,
"non_key_frame_acc": 1.0,
"n_episodes": 10,
"n_windows": 102,
"n_segments": 102,
"n_nonkey_episodes": 10,
"missed": 0,
"excess_windows": 1,
"spurious_total": 0,
"spurious_per_ep": 0.0,
"total_yes_pred": 103,
"perfect_episodes": 9,
"key_interval_hit_rates": [
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0
],
"key_interval_acc_rates": [
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
0.9,
1.0,
1.0,
1.0,
1.0,
1.0
],
"key_interval_hit_counts": [
10,
10,
10,
10,
10,
10,
10,
10,
10,
7,
3,
2
],
"key_interval_acc_counts": [
10,
10,
10,
10,
10,
10,
9,
10,
10,
7,
3,
2
],
"key_interval_episode_counts": [
10,
10,
10,
10,
10,
10,
10,
10,
10,
7,
3,
2
],
"trajectory_count": 10
},
"put_back_block": {
"precision": 0.9091,
"recall": 1.0,
"f1": 0.9524,
"segment": {
"1": {
"hit": 1.0,
"acc": 1.0
},
"2": {
"hit": 1.0,
"acc": 1.0
},
"overall": {
"hit": 1.0,
"acc": 1.0
}
},
"segment_acc": 1.0,
"non_key_acc": 0.8,
"overall_score": 0.9,
"exact_rate": 1.0,
"segment_positive": 1.0,
"non_key_frame_acc": 0.9979,
"n_episodes": 10,
"n_windows": 20,
"n_segments": 20,
"n_nonkey_episodes": 10,
"missed": 0,
"excess_windows": 0,
"spurious_total": 2,
"spurious_per_ep": 0.2,
"total_yes_pred": 22,
"perfect_episodes": 8,
"key_interval_hit_rates": [
1.0,
1.0
],
"key_interval_acc_rates": [
1.0,
1.0
],
"key_interval_hit_counts": [
10,
10
],
"key_interval_acc_counts": [
10,
10
],
"key_interval_episode_counts": [
10,
10
],
"trajectory_count": 10
},
"rearrange_blocks": {
"precision": 0.76,
"recall": 0.95,
"f1": 0.8444,
"segment": {
"1": {
"hit": 1.0,
"acc": 1.0
},
"2": {
"hit": 0.9,
"acc": 0.7
},
"overall": {
"hit": 0.95,
"acc": 0.85
}
},
"segment_acc": 0.85,
"non_key_acc": 0.6,
"overall_score": 0.725,
"exact_rate": 0.85,
"segment_positive": 0.95,
"non_key_frame_acc": 0.9966,
"n_episodes": 10,
"n_windows": 20,
"n_segments": 20,
"n_nonkey_episodes": 10,
"missed": 1,
"excess_windows": 2,
"spurious_total": 4,
"spurious_per_ep": 0.4,
"total_yes_pred": 25,
"perfect_episodes": 4,
"key_interval_hit_rates": [
1.0,
0.9
],
"key_interval_acc_rates": [
1.0,
0.7
],
"key_interval_hit_counts": [
10,
9
],
"key_interval_acc_counts": [
10,
7
],
"key_interval_episode_counts": [
10,
10
],
"trajectory_count": 10
},
"swap_T": {
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"segment": {
"1": {
"hit": 1.0,
"acc": 1.0
},
"overall": {
"hit": 1.0,
"acc": 1.0
}
},
"segment_acc": 1.0,
"non_key_acc": 1.0,
"overall_score": 1.0,
"exact_rate": 1.0,
"segment_positive": 1.0,
"non_key_frame_acc": 1.0,
"n_episodes": 10,
"n_windows": 10,
"n_segments": 10,
"n_nonkey_episodes": 10,
"missed": 0,
"excess_windows": 0,
"spurious_total": 0,
"spurious_per_ep": 0.0,
"total_yes_pred": 10,
"perfect_episodes": 10,
"key_interval_hit_rates": [
1.0
],
"key_interval_acc_rates": [
1.0
],
"key_interval_hit_counts": [
10
],
"key_interval_acc_counts": [
10
],
"key_interval_episode_counts": [
10
],
"trajectory_count": 10
},
"swap_blocks": {
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"segment": {
"1": {
"hit": 1.0,
"acc": 1.0
},
"overall": {
"hit": 1.0,
"acc": 1.0
}
},
"segment_acc": 1.0,
"non_key_acc": 1.0,
"overall_score": 1.0,
"exact_rate": 1.0,
"segment_positive": 1.0,
"non_key_frame_acc": 1.0,
"n_episodes": 10,
"n_windows": 10,
"n_segments": 10,
"n_nonkey_episodes": 10,
"missed": 0,
"excess_windows": 0,
"spurious_total": 0,
"spurious_per_ep": 0.0,
"total_yes_pred": 10,
"perfect_episodes": 10,
"key_interval_hit_rates": [
1.0
],
"key_interval_acc_rates": [
1.0
],
"key_interval_hit_counts": [
10
],
"key_interval_acc_counts": [
10
],
"key_interval_episode_counts": [
10
],
"trajectory_count": 10
}
},
"trajectory_counts": {
"overall": 90,
"by_task": {
"battery_try": 10,
"blocks_ranking_try": 10,
"cover_blocks": 10,
"observe_and_pickup": 10,
"press_button": 10,
"put_back_block": 10,
"rearrange_blocks": 10,
"swap_T": 10,
"swap_blocks": 10
}
},
"step": 18000,
"threshold": 0.5,
"split": "val",
"split_ratio": 0.2,
"train_split_ratio": 0.8,
"validation_split_ratio": 0.2,
"protocol": "autoregressive"
}