eval_pi0_so101 / evaluation_summary.json
SGPatil's picture
Upload folder using huggingface_hub
f7ab77b verified
{
"config": {
"config_name": "pi0_so101_lora",
"checkpoint_path": "/workspace/so101_table_cleanup_checkpoints/so101_finetune_table_cleanup/10000",
"dataset_repo": "SGPatil/so101_table_cleanup_2",
"num_trajectories": 10,
"max_steps_per_traj": 150,
"action_horizon": 16,
"plot": true,
"save_plots": true,
"output_dir": "./so101_table_cleanup_eval_results",
"skip_norm_stats": false,
"prediction_visualization": true,
"show_prediction_horizon": true
},
"dataset_info": {
"total_samples": 21342,
"total_episodes": 10,
"trajectory_lengths": [
1991,
2395,
2950,
1,
2363,
1926,
2799,
2098,
2167,
2652
]
},
"evaluation_summary": {
"trajectories_evaluated": 10,
"total_steps": 21342,
"total_trajectory_steps": 21342,
"completion_rate": 1.0,
"overall_mse": 92.66323264624474,
"overall_mae": 5.69791082210956
},
"all_trajectory_metrics": [
{
"overall_mse": 117.41691214191862,
"mse_per_dim": "[ 33.75217355 141.78902481 194.81987301 108.21602774 146.44095877\n 79.48341496]",
"mae": 6.651712886891315,
"max_error": 85.6978912108335,
"total_inferences": 125,
"mean_inference_time_ms": 394.4953923262656,
"total_inference_time_ms": 49311.9240407832,
"inference_frequency_hz": 2.5348838527699575,
"action_frequency_hz": 40.55814164431932,
"steps_per_inference": 15.928,
"real_time_factor": 0.7430224616893502,
"trajectory_duration_s": 66.36666666666666,
"total_compute_time_s": 49.3119240407832,
"trajectory_id": 0,
"steps_evaluated": 1991,
"trajectory_length": 1991,
"prediction_points": 125,
"trajectory_completion": 1.0
},
{
"overall_mse": 89.31719109969124,
"mse_per_dim": "[ 24.28022703 81.84848464 112.88838129 71.10057971 135.54888171\n 110.23659221]",
"mae": 5.575415097766607,
"max_error": 70.38384317054106,
"total_inferences": 150,
"mean_inference_time_ms": 133.6694772914052,
"total_inference_time_ms": 20050.42159371078,
"inference_frequency_hz": 7.481139451304632,
"action_frequency_hz": 119.6982312208741,
"steps_per_inference": 15.966666666666667,
"real_time_factor": 0.2511535063930369,
"trajectory_duration_s": 79.83333333333333,
"total_compute_time_s": 20.05042159371078,
"trajectory_id": 1,
"steps_evaluated": 2395,
"trajectory_length": 2395,
"prediction_points": 150,
"trajectory_completion": 1.0
},
{
"overall_mse": 108.89362054318462,
"mse_per_dim": "[ 30.74533793 98.48555634 158.93579591 98.96829197 121.80158088\n 144.42516024]",
"mae": 6.204135471773346,
"max_error": 112.77167540158348,
"total_inferences": 185,
"mean_inference_time_ms": 135.1896471243251,
"total_inference_time_ms": 25010.084718000144,
"inference_frequency_hz": 7.3970161271326145,
"action_frequency_hz": 118.35225803412183,
"steps_per_inference": 15.945945945945946,
"real_time_factor": 0.254339844589832,
"trajectory_duration_s": 98.33333333333333,
"total_compute_time_s": 25.010084718000144,
"trajectory_id": 2,
"steps_evaluated": 2950,
"trajectory_length": 2950,
"prediction_points": 185,
"trajectory_completion": 1.0
},
{
"overall_mse": 2.334123749218712,
"mse_per_dim": "[0.49545456 2.78371481 7.12946492 2.53734238 0.23363216 0.82513367]",
"mae": 1.3378442490310085,
"max_error": 2.670105788430064,
"total_inferences": 1,
"mean_inference_time_ms": 117.1392579562962,
"total_inference_time_ms": 117.1392579562962,
"inference_frequency_hz": 8.536847658477507,
"action_frequency_hz": 136.5895625356401,
"steps_per_inference": 1.0,
"real_time_factor": 3.5141777386888857,
"trajectory_duration_s": 0.03333333333333333,
"total_compute_time_s": 0.1171392579562962,
"trajectory_id": 3,
"steps_evaluated": 1,
"trajectory_length": 1,
"prediction_points": 1,
"trajectory_completion": 1.0
},
{
"overall_mse": 81.79151826026786,
"mse_per_dim": "[ 30.7786456 92.66727207 133.3681253 81.32807201 96.8656252\n 55.74136938]",
"mae": 5.479987127517952,
"max_error": 56.392993099490894,
"total_inferences": 148,
"mean_inference_time_ms": 135.38044110582388,
"total_inference_time_ms": 20036.30528366193,
"inference_frequency_hz": 7.386591385223234,
"action_frequency_hz": 118.18546216357174,
"steps_per_inference": 15.966216216216216,
"real_time_factor": 0.25437543737192464,
"trajectory_duration_s": 78.76666666666667,
"total_compute_time_s": 20.03630528366193,
"trajectory_id": 4,
"steps_evaluated": 2363,
"trajectory_length": 2363,
"prediction_points": 148,
"trajectory_completion": 1.0
},
{
"overall_mse": 90.6695644210243,
"mse_per_dim": "[ 30.46134065 103.41117919 145.47142887 115.31919318 92.15842852\n 57.19581611]",
"mae": 6.0825723685003155,
"max_error": 70.35922241966495,
"total_inferences": 121,
"mean_inference_time_ms": 133.50109953033038,
"total_inference_time_ms": 16153.633043169975,
"inference_frequency_hz": 7.490575010378907,
"action_frequency_hz": 119.84920016606252,
"steps_per_inference": 15.917355371900827,
"real_time_factor": 0.25161422185623017,
"trajectory_duration_s": 64.2,
"total_compute_time_s": 16.153633043169975,
"trajectory_id": 5,
"steps_evaluated": 1926,
"trajectory_length": 1926,
"prediction_points": 121,
"trajectory_completion": 1.0
},
{
"overall_mse": 98.7208116175219,
"mse_per_dim": "[ 32.30608855 88.79200271 143.06268672 126.74578835 109.82338451\n 91.59491886]",
"mae": 6.331176612206364,
"max_error": 73.07399605597364,
"total_inferences": 175,
"mean_inference_time_ms": 132.40290249032634,
"total_inference_time_ms": 23170.50793580711,
"inference_frequency_hz": 7.552704519246187,
"action_frequency_hz": 120.84327230793899,
"steps_per_inference": 15.994285714285715,
"real_time_factor": 0.24834413650382753,
"trajectory_duration_s": 93.3,
"total_compute_time_s": 23.17050793580711,
"trajectory_id": 6,
"steps_evaluated": 2799,
"trajectory_length": 2799,
"prediction_points": 175,
"trajectory_completion": 1.0
},
{
"overall_mse": 93.91889705028609,
"mse_per_dim": "[ 37.30934062 98.67510082 166.47588759 97.36617788 113.7578624\n 49.92901299]",
"mae": 6.083543485402971,
"max_error": 61.6112560290487,
"total_inferences": 132,
"mean_inference_time_ms": 132.78021972457117,
"total_inference_time_ms": 17526.989003643394,
"inference_frequency_hz": 7.531242244321641,
"action_frequency_hz": 120.49987590914625,
"steps_per_inference": 15.893939393939394,
"real_time_factor": 0.25062424695390934,
"trajectory_duration_s": 69.93333333333334,
"total_compute_time_s": 17.526989003643394,
"trajectory_id": 7,
"steps_evaluated": 2098,
"trajectory_length": 2098,
"prediction_points": 132,
"trajectory_completion": 1.0
},
{
"overall_mse": 101.98447416633441,
"mse_per_dim": "[ 48.03679853 89.23403209 161.4384558 96.04179859 174.42870696\n 42.72705302]",
"mae": 6.094967712272187,
"max_error": 71.77420839794762,
"total_inferences": 136,
"mean_inference_time_ms": 135.32244616105933,
"total_inference_time_ms": 18403.85267790407,
"inference_frequency_hz": 7.389757045995242,
"action_frequency_hz": 118.23611273592387,
"steps_per_inference": 15.933823529411764,
"real_time_factor": 0.254783378097426,
"trajectory_duration_s": 72.23333333333333,
"total_compute_time_s": 18.40385267790407,
"trajectory_id": 8,
"steps_evaluated": 2167,
"trajectory_length": 2167,
"prediction_points": 136,
"trajectory_completion": 1.0
},
{
"overall_mse": 141.58521341299965,
"mse_per_dim": "[ 68.79387229 90.17015123 139.25531606 109.19437184 384.12434219\n 57.97322687]",
"mae": 7.137753209733537,
"max_error": 122.03772037589201,
"total_inferences": 166,
"mean_inference_time_ms": 132.95422438868738,
"total_inference_time_ms": 22070.401248522103,
"inference_frequency_hz": 7.521385684418212,
"action_frequency_hz": 120.3421709506914,
"steps_per_inference": 15.975903614457831,
"real_time_factor": 0.2496651724945939,
"trajectory_duration_s": 88.4,
"total_compute_time_s": 22.070401248522103,
"trajectory_id": 9,
"steps_evaluated": 2652,
"trajectory_length": 2652,
"prediction_points": 166,
"trajectory_completion": 1.0
}
]
}