Tao-toolkit:5.0.0-tf1.15.5 Detectnet_v2 training is reaching more than 1.0 Mean Average Precision

Accelerated Computing Intelligent Video Analytics TAO Toolkit
1

My Issue is I am simply training a person dataset from Roboflow, however the training set is currently mean average precision is shooting up to 4.0952 which to my understanding is 400%.

Or am I missing something, is it just 4% now? because from the tensorboard when i saw the inference there are lots of false detections and loss of detections as well.

• Hardware (RTX3060)
• Network Type (Detectnet_v2)
• TLT Version (tao-toolkit:5.0.0-tf1.15.5)
• Training spec file(If have, please share here)
model_config {
arch: “resnet”
all_projections: True
num_layers: 18
use_pooling: True
use_batch_norm: True
dropout_rate: 0.0
objective_set: {
cov {}
bbox {
scale: 35.0
offset: 0.5
}
}
}

bbox_rasterizer_config {
target_class_config {
key: “person”
value: {
cov_center_x: 0.5
cov_center_y: 0.5
cov_radius_x: 0.4
cov_radius_y: 0.4
bbox_min_radius: 1.0
}
}
deadzone_radius: 0.67
}

training_config {
batch_size_per_gpu: 16
num_epochs: 100
learning_rate {
soft_start_annealing_schedule {
min_learning_rate: 5e-6
max_learning_rate: 5e-4
soft_start: 0.1
annealing: 0.7
}
}
regularizer {
type: L1
weight: 3e-9
}
optimizer {
adam {
epsilon: 1e-08
beta1: 0.9
beta2: 0.999
}
}
cost_scaling {
enabled: true
initial_exponent: 20.0
increment: 0.005
decrement: 1.0
}
visualizer {
enabled: true
num_images: 3
scalar_logging_frequency: 10
infrequent_logging_frequency: 1
target_class_config {
key: “person”
value: {
coverage_threshold: 0.005
}
}
}
}

postprocessing_config {
target_class_config {
key: “person”
value: {
clustering_config {
coverage_threshold: 0.005
dbscan_eps: 0.15
dbscan_min_samples: 1
minimum_bounding_box_height: 20
}
}
}
}

cost_function_config {
target_classes {
name: “person”
class_weight: 1.0
coverage_foreground_weight: 0.05
objectives {
name: “cov”
initial_weight: 1.0
weight_target: 1.0
}
objectives {
name: “bbox”
initial_weight: 10.0
weight_target: 10.0
}
}
enable_autoweighting: True
max_objective_weight: 0.9999
min_objective_weight: 0.0001
}

augmentation_config {
preprocessing {
output_image_width: 960
output_image_height: 544
output_image_channel: 3
min_bbox_width: 1.0
min_bbox_height: 1.0
}
spatial_augmentation {

hflip_probability: 0.5
vflip_probability: 0.0
zoom_min: 1.0
zoom_max: 1.0
translate_max_x: 8.0
translate_max_y: 8.0

}
color_augmentation {
color_shift_stddev: 0.0
hue_rotation_max: 25.0
saturation_shift_max: 0.2
contrast_scale_max: 0.1
contrast_center: 0.5
}
}

evaluation_config {
average_precision_mode: INTEGRATE
validation_period_during_training: 10
first_validation_epoch: 1
minimum_detection_ground_truth_overlap {
key: “person”
value: 0.5
}
evaluation_box_config {
key: “person”
value {
minimum_height: 4
maximum_height: 9999
minimum_width: 4
maximum_width: 9999
}
}
}

dataset_config {
data_sources: {
tfrecords_path: “/workspace/tao-experiments/head_coco_dataset-fold-*”
image_directory_path: “/workspace/tao-experiments/head_coco”
}
image_extension: “jpg”

target_class_mapping {
key: “person”
value: “person”
}
validation_data_source: {
tfrecords_path: “/workspace/tao-experiments/head_coco_dataset-fold-*”
image_directory_path: “/workspace/tao-experiments/head_coco”
}

}

Status.json
{“date”: “9/30/2025”, “time”: “3:19:58”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Evaluation metrics generated.”, “categorical”: {“average_precision”: {“person”: 0.0}}, “graphical”: {“loss”: 0.404430627822876, “learning_rate”: “4.9999994e-06”}, “kpi”: {“validation cost”: 0.24861654, “mean average precision”: 0.0}}
{“epoch”: 1, “max_epoch”: 100, “time_per_epoch”: “0:04:15.080275”, “eta”: “7:00:52.947254”, “date”: “9/30/2025”, “time”: “3:20:1”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0}}, “graphical”: {“loss”: 0.20976409316062927, “learning_rate”: “7.924464e-06”}, “kpi”: {“validation cost”: 0.24861654, “mean average precision”: 0.0}}
{“epoch”: 2, “max_epoch”: 100, “time_per_epoch”: “0:02:14.991050”, “eta”: “3:40:29.122854”, “date”: “9/30/2025”, “time”: “3:22:14”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0}}, “graphical”: {“loss”: 0.19671112298965454, “learning_rate”: “1.2559428e-05”}, “kpi”: {“validation cost”: 0.24861654, “mean average precision”: 0.0}}
{“date”: “9/30/2025”, “time”: “3:23:47”, “status”: “STARTED”, “verbosity”: “INFO”, “message”: “Starting DetectNet_v2 Training job”}
{“date”: “9/30/2025”, “time”: “3:23:47”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Training gridbox model.”}
{“date”: “9/30/2025”, “time”: “3:23:49”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Building DetectNet V2 model”}
{“date”: “9/30/2025”, “time”: “3:23:50”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “DetectNet V2 model built.”}
{“date”: “9/30/2025”, “time”: “3:23:50”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Building rasterizer.”}
{“date”: “9/30/2025”, “time”: “3:23:50”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Rasterizers built.”}
{“date”: “9/30/2025”, “time”: “3:23:50”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Building training graph.”}
{“date”: “9/30/2025”, “time”: “3:23:53”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Rasterizing tensors.”}
{“date”: “9/30/2025”, “time”: “3:23:53”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Tensors rasterized.”}
{“date”: “9/30/2025”, “time”: “3:23:55”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Training graph built.”}
{“date”: “9/30/2025”, “time”: “3:23:55”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Building validation graph.”}
{“date”: “9/30/2025”, “time”: “3:23:55”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Rasterizing tensors.”}
{“date”: “9/30/2025”, “time”: “3:23:55”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Tensors rasterized.”}
{“date”: “9/30/2025”, “time”: “3:23:55”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Validation graph built.”}
{“date”: “9/30/2025”, “time”: “3:23:56”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Running training loop.”}
{“epoch”: 0, “max_epoch”: 100, “time_per_epoch”: “0:00:00”, “eta”: “0:00:00”, “date”: “9/30/2025”, “time”: “3:24:39”, “status”: “RUNNING”, “verbosity”: “INFO”, “graphical”: {“loss”: 0.4044340252876282, “learning_rate”: “4.9999994e-06”}}
{“date”: “9/30/2025”, “time”: “3:28:22”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Evaluation metrics generated.”, “categorical”: {“average_precision”: {“person”: 0.0}}, “graphical”: {“loss”: 0.4044340252876282, “learning_rate”: “4.9999994e-06”}, “kpi”: {“validation cost”: 0.24905401, “mean average precision”: 0.0}}
{“epoch”: 1, “max_epoch”: 100, “time_per_epoch”: “0:04:10.378934”, “eta”: “6:53:07.514433”, “date”: “9/30/2025”, “time”: “3:28:24”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0}}, “graphical”: {“loss”: 0.2099231630563736, “learning_rate”: “7.924464e-06”}, “kpi”: {“validation cost”: 0.24905401, “mean average precision”: 0.0}}
{“epoch”: 2, “max_epoch”: 100, “time_per_epoch”: “0:02:12.502018”, “eta”: “3:36:25.197808”, “date”: “9/30/2025”, “time”: “3:30:35”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0}}, “graphical”: {“loss”: 0.1956145167350769, “learning_rate”: “1.2559428e-05”}, “kpi”: {“validation cost”: 0.24905401, “mean average precision”: 0.0}}
{“epoch”: 3, “max_epoch”: 100, “time_per_epoch”: “0:02:12.429793”, “eta”: “3:34:05.689886”, “date”: “9/30/2025”, “time”: “3:32:47”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0}}, “graphical”: {“loss”: 0.17567414045333862, “learning_rate”: “1.990535e-05”}, “kpi”: {“validation cost”: 0.24905401, “mean average precision”: 0.0}}
{“epoch”: 4, “max_epoch”: 100, “time_per_epoch”: “0:02:11.806518”, “eta”: “3:30:53.425690”, “date”: “9/30/2025”, “time”: “3:34:59”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0}}, “graphical”: {“loss”: 0.18284818530082703, “learning_rate”: “3.154785e-05”}, “kpi”: {“validation cost”: 0.24905401, “mean average precision”: 0.0}}
{“epoch”: 5, “max_epoch”: 100, “time_per_epoch”: “0:02:11.481423”, “eta”: “3:28:10.735153”, “date”: “9/30/2025”, “time”: “3:37:11”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0}}, “graphical”: {“loss”: 0.1571374386548996, “learning_rate”: “4.999997e-05”}, “kpi”: {“validation cost”: 0.24905401, “mean average precision”: 0.0}}
{“epoch”: 6, “max_epoch”: 100, “time_per_epoch”: “0:02:12.037913”, “eta”: “3:26:51.563785”, “date”: “9/30/2025”, “time”: “3:39:23”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0}}, “graphical”: {“loss”: 0.1412360668182373, “learning_rate”: “7.924468e-05”}, “kpi”: {“validation cost”: 0.24905401, “mean average precision”: 0.0}}
{“epoch”: 7, “max_epoch”: 100, “time_per_epoch”: “0:02:11.484504”, “eta”: “3:23:48.058915”, “date”: “9/30/2025”, “time”: “3:41:34”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0}}, “graphical”: {“loss”: 0.12434294819831848, “learning_rate”: “0.00012559434”}, “kpi”: {“validation cost”: 0.24905401, “mean average precision”: 0.0}}
{“epoch”: 8, “max_epoch”: 100, “time_per_epoch”: “0:02:11.370558”, “eta”: “3:21:26.091294”, “date”: “9/30/2025”, “time”: “3:43:46”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0}}, “graphical”: {“loss”: 0.1364600658416748, “learning_rate”: “0.0001990536”}, “kpi”: {“validation cost”: 0.24905401, “mean average precision”: 0.0}}
{“epoch”: 9, “max_epoch”: 100, “time_per_epoch”: “0:02:11.525963”, “eta”: “3:19:28.862639”, “date”: “9/30/2025”, “time”: “3:45:57”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0}}, “graphical”: {“loss”: 0.1269717812538147, “learning_rate”: “0.00031547868”}, “kpi”: {“validation cost”: 0.24905401, “mean average precision”: 0.0}}
{“epoch”: 10, “max_epoch”: 100, “time_per_epoch”: “0:02:12.885961”, “eta”: “3:19:19.736452”, “date”: “9/30/2025”, “time”: “3:48:10”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0}}, “graphical”: {“loss”: 0.1143883764743805, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.24905401, “mean average precision”: 0.0}}
{“date”: “9/30/2025”, “time”: “3:51:27”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Evaluation metrics generated.”, “categorical”: {“average_precision”: {“person”: 0.035}}, “graphical”: {“loss”: 0.1143883764743805, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.13437318, “mean average precision”: 0.035}}
{“epoch”: 11, “max_epoch”: 100, “time_per_epoch”: “0:03:17.874772”, “eta”: “4:53:30.854714”, “date”: “9/30/2025”, “time”: “3:51:28”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.035}}, “graphical”: {“loss”: 0.12403935939073563, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.13437318, “mean average precision”: 0.035}}
{“epoch”: 12, “max_epoch”: 100, “time_per_epoch”: “0:02:14.014062”, “eta”: “3:16:33.237492”, “date”: “9/30/2025”, “time”: “3:53:42”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.035}}, “graphical”: {“loss”: 0.1133691668510437, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.13437318, “mean average precision”: 0.035}}
{“epoch”: 13, “max_epoch”: 100, “time_per_epoch”: “0:02:12.168556”, “eta”: “3:11:38.664411”, “date”: “9/30/2025”, “time”: “3:55:54”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.035}}, “graphical”: {“loss”: 0.10995160788297653, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.13437318, “mean average precision”: 0.035}}
{“epoch”: 14, “max_epoch”: 100, “time_per_epoch”: “0:02:13.514683”, “eta”: “3:11:22.262780”, “date”: “9/30/2025”, “time”: “3:58:7”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.035}}, “graphical”: {“loss”: 0.10176346451044083, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.13437318, “mean average precision”: 0.035}}
{“epoch”: 15, “max_epoch”: 100, “time_per_epoch”: “0:02:13.899862”, “eta”: “3:09:41.488254”, “date”: “9/30/2025”, “time”: “4:0:21”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.035}}, “graphical”: {“loss”: 0.10129997134208679, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.13437318, “mean average precision”: 0.035}}
{“epoch”: 16, “max_epoch”: 100, “time_per_epoch”: “0:02:13.472258”, “eta”: “3:06:51.669700”, “date”: “9/30/2025”, “time”: “4:2:35”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.035}}, “graphical”: {“loss”: 0.09512515366077423, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.13437318, “mean average precision”: 0.035}}
{“epoch”: 17, “max_epoch”: 100, “time_per_epoch”: “0:02:11.886167”, “eta”: “3:02:26.551865”, “date”: “9/30/2025”, “time”: “4:4:47”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.035}}, “graphical”: {“loss”: 0.10050839185714722, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.13437318, “mean average precision”: 0.035}}
{“epoch”: 18, “max_epoch”: 100, “time_per_epoch”: “0:02:09.674380”, “eta”: “2:57:13.299185”, “date”: “9/30/2025”, “time”: “4:6:56”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.035}}, “graphical”: {“loss”: 0.09863967448472977, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.13437318, “mean average precision”: 0.035}}
{“epoch”: 19, “max_epoch”: 100, “time_per_epoch”: “0:02:09.570286”, “eta”: “2:54:55.193130”, “date”: “9/30/2025”, “time”: “4:9:6”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.035}}, “graphical”: {“loss”: 0.11210843920707703, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.13437318, “mean average precision”: 0.035}}
{“epoch”: 20, “max_epoch”: 100, “time_per_epoch”: “0:02:10.634506”, “eta”: “2:54:10.760441”, “date”: “9/30/2025”, “time”: “4:11:17”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.035}}, “graphical”: {“loss”: 0.09492345154285431, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.13437318, “mean average precision”: 0.035}}
{“date”: “9/30/2025”, “time”: “4:14:29”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Evaluation metrics generated.”, “categorical”: {“average_precision”: {“person”: 0.0771}}, “graphical”: {“loss”: 0.09492345154285431, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.11266375, “mean average precision”: 0.0771}}
{“epoch”: 21, “max_epoch”: 100, “time_per_epoch”: “0:03:13.016746”, “eta”: “4:14:08.322937”, “date”: “9/30/2025”, “time”: “4:14:30”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0771}}, “graphical”: {“loss”: 0.09396826475858688, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.11266375, “mean average precision”: 0.0771}}
{“epoch”: 22, “max_epoch”: 100, “time_per_epoch”: “0:02:09.551654”, “eta”: “2:48:25.029038”, “date”: “9/30/2025”, “time”: “4:16:39”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0771}}, “graphical”: {“loss”: 0.0998244360089302, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.11266375, “mean average precision”: 0.0771}}
{“epoch”: 23, “max_epoch”: 100, “time_per_epoch”: “0:02:09.602973”, “eta”: “2:46:19.428883”, “date”: “9/30/2025”, “time”: “4:18:49”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0771}}, “graphical”: {“loss”: 0.0872088372707367, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.11266375, “mean average precision”: 0.0771}}
{“epoch”: 24, “max_epoch”: 100, “time_per_epoch”: “0:02:09.669931”, “eta”: “2:44:14.914769”, “date”: “9/30/2025”, “time”: “4:20:58”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0771}}, “graphical”: {“loss”: 0.08118899911642075, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.11266375, “mean average precision”: 0.0771}}
{“epoch”: 25, “max_epoch”: 100, “time_per_epoch”: “0:02:09.569647”, “eta”: “2:41:57.723513”, “date”: “9/30/2025”, “time”: “4:23:8”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0771}}, “graphical”: {“loss”: 0.0845017209649086, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.11266375, “mean average precision”: 0.0771}}
{“epoch”: 26, “max_epoch”: 100, “time_per_epoch”: “0:02:09.585279”, “eta”: “2:39:49.310645”, “date”: “9/30/2025”, “time”: “4:25:18”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0771}}, “graphical”: {“loss”: 0.08165459334850311, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.11266375, “mean average precision”: 0.0771}}
{“epoch”: 27, “max_epoch”: 100, “time_per_epoch”: “0:02:09.911366”, “eta”: “2:38:03.529682”, “date”: “9/30/2025”, “time”: “4:27:28”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0771}}, “graphical”: {“loss”: 0.08103282004594803, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.11266375, “mean average precision”: 0.0771}}
{“epoch”: 28, “max_epoch”: 100, “time_per_epoch”: “0:02:09.569090”, “eta”: “2:35:28.974506”, “date”: “9/30/2025”, “time”: “4:29:37”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0771}}, “graphical”: {“loss”: 0.09038716554641724, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.11266375, “mean average precision”: 0.0771}}
{“epoch”: 29, “max_epoch”: 100, “time_per_epoch”: “0:02:09.480799”, “eta”: “2:33:13.136709”, “date”: “9/30/2025”, “time”: “4:31:47”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0771}}, “graphical”: {“loss”: 0.08249232172966003, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.11266375, “mean average precision”: 0.0771}}
{“epoch”: 30, “max_epoch”: 100, “time_per_epoch”: “0:02:10.444663”, “eta”: “2:32:11.126397”, “date”: “9/30/2025”, “time”: “4:33:57”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.0771}}, “graphical”: {“loss”: 0.07102286070585251, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.11266375, “mean average precision”: 0.0771}}
{“date”: “9/30/2025”, “time”: “4:37:10”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Evaluation metrics generated.”, “categorical”: {“average_precision”: {“person”: 0.4242}}, “graphical”: {“loss”: 0.07102286070585251, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.08068494, “mean average precision”: 0.4242}}
{“epoch”: 31, “max_epoch”: 100, “time_per_epoch”: “0:03:13.743084”, “eta”: “3:42:48.272826”, “date”: “9/30/2025”, “time”: “4:37:11”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.4242}}, “graphical”: {“loss”: 0.08345421403646469, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.08068494, “mean average precision”: 0.4242}}
{“epoch”: 32, “max_epoch”: 100, “time_per_epoch”: “0:02:09.545792”, “eta”: “2:26:49.113831”, “date”: “9/30/2025”, “time”: “4:39:20”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.4242}}, “graphical”: {“loss”: 0.0773557648062706, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.08068494, “mean average precision”: 0.4242}}
{“epoch”: 33, “max_epoch”: 100, “time_per_epoch”: “0:02:09.618582”, “eta”: “2:24:44.445027”, “date”: “9/30/2025”, “time”: “4:41:30”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.4242}}, “graphical”: {“loss”: 0.07543889433145523, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.08068494, “mean average precision”: 0.4242}}
{“epoch”: 34, “max_epoch”: 100, “time_per_epoch”: “0:02:09.681843”, “eta”: “2:22:39.001657”, “date”: “9/30/2025”, “time”: “4:43:40”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.4242}}, “graphical”: {“loss”: 0.08766428381204605, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.08068494, “mean average precision”: 0.4242}}
{“epoch”: 35, “max_epoch”: 100, “time_per_epoch”: “0:02:09.811540”, “eta”: “2:20:37.750077”, “date”: “9/30/2025”, “time”: “4:45:49”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.4242}}, “graphical”: {“loss”: 0.0708388239145279, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.08068494, “mean average precision”: 0.4242}}
{“epoch”: 36, “max_epoch”: 100, “time_per_epoch”: “0:02:09.751746”, “eta”: “2:18:24.111771”, “date”: “9/30/2025”, “time”: “4:47:59”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.4242}}, “graphical”: {“loss”: 0.08282347768545151, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.08068494, “mean average precision”: 0.4242}}
{“epoch”: 37, “max_epoch”: 100, “time_per_epoch”: “0:02:09.714375”, “eta”: “2:16:12.005641”, “date”: “9/30/2025”, “time”: “4:50:9”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.4242}}, “graphical”: {“loss”: 0.06885235756635666, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.08068494, “mean average precision”: 0.4242}}
{“epoch”: 38, “max_epoch”: 100, “time_per_epoch”: “0:02:09.714933”, “eta”: “2:14:02.325856”, “date”: “9/30/2025”, “time”: “4:52:19”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.4242}}, “graphical”: {“loss”: 0.06692330539226532, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.08068494, “mean average precision”: 0.4242}}
{“epoch”: 39, “max_epoch”: 100, “time_per_epoch”: “0:02:09.826357”, “eta”: “2:11:59.407770”, “date”: “9/30/2025”, “time”: “4:54:28”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.4242}}, “graphical”: {“loss”: 0.06867922097444534, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.08068494, “mean average precision”: 0.4242}}
{“epoch”: 40, “max_epoch”: 100, “time_per_epoch”: “0:02:10.488129”, “eta”: “2:10:29.287763”, “date”: “9/30/2025”, “time”: “4:56:39”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.4242}}, “graphical”: {“loss”: 0.06795738637447357, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.08068494, “mean average precision”: 0.4242}}
{“date”: “9/30/2025”, “time”: “4:59:53”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Evaluation metrics generated.”, “categorical”: {“average_precision”: {“person”: 0.8251}}, “graphical”: {“loss”: 0.06795738637447357, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.06829815, “mean average precision”: 0.8251}}
{“epoch”: 41, “max_epoch”: 100, “time_per_epoch”: “0:03:14.843125”, “eta”: “3:11:35.744367”, “date”: “9/30/2025”, “time”: “4:59:54”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8251}}, “graphical”: {“loss”: 0.07392945140600204, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.06829815, “mean average precision”: 0.8251}}
{“epoch”: 42, “max_epoch”: 100, “time_per_epoch”: “0:02:10.662194”, “eta”: “2:06:18.407239”, “date”: “9/30/2025”, “time”: “5:2:4”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8251}}, “graphical”: {“loss”: 0.061782531440258026, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.06829815, “mean average precision”: 0.8251}}
{“epoch”: 43, “max_epoch”: 100, “time_per_epoch”: “0:02:11.300570”, “eta”: “2:04:44.132491”, “date”: “9/30/2025”, “time”: “5:4:16”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8251}}, “graphical”: {“loss”: 0.06186389550566673, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.06829815, “mean average precision”: 0.8251}}
{“epoch”: 44, “max_epoch”: 100, “time_per_epoch”: “0:02:09.878021”, “eta”: “2:01:13.169163”, “date”: “9/30/2025”, “time”: “5:6:26”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8251}}, “graphical”: {“loss”: 0.07462797313928604, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.06829815, “mean average precision”: 0.8251}}
{“epoch”: 45, “max_epoch”: 100, “time_per_epoch”: “0:02:09.671085”, “eta”: “1:58:51.909668”, “date”: “9/30/2025”, “time”: “5:8:35”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8251}}, “graphical”: {“loss”: 0.05940079316496849, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.06829815, “mean average precision”: 0.8251}}
{“epoch”: 46, “max_epoch”: 100, “time_per_epoch”: “0:02:09.789669”, “eta”: “1:56:48.642141”, “date”: “9/30/2025”, “time”: “5:10:45”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8251}}, “graphical”: {“loss”: 0.07930578291416168, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.06829815, “mean average precision”: 0.8251}}
{“epoch”: 47, “max_epoch”: 100, “time_per_epoch”: “0:02:09.921391”, “eta”: “1:54:45.833698”, “date”: “9/30/2025”, “time”: “5:12:55”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8251}}, “graphical”: {“loss”: 0.06578066945075989, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.06829815, “mean average precision”: 0.8251}}
{“epoch”: 48, “max_epoch”: 100, “time_per_epoch”: “0:02:09.592834”, “eta”: “1:52:18.827368”, “date”: “9/30/2025”, “time”: “5:15:5”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8251}}, “graphical”: {“loss”: 0.06182970106601715, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.06829815, “mean average precision”: 0.8251}}
{“epoch”: 49, “max_epoch”: 100, “time_per_epoch”: “0:02:09.790911”, “eta”: “1:50:19.336471”, “date”: “9/30/2025”, “time”: “5:17:14”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8251}}, “graphical”: {“loss”: 0.06735049188137054, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.06829815, “mean average precision”: 0.8251}}
{“epoch”: 50, “max_epoch”: 100, “time_per_epoch”: “0:02:10.866373”, “eta”: “1:49:03.318629”, “date”: “9/30/2025”, “time”: “5:19:25”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8251}}, “graphical”: {“loss”: 0.056647732853889465, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.06829815, “mean average precision”: 0.8251}}
{“date”: “9/30/2025”, “time”: “5:22:39”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Evaluation metrics generated.”, “categorical”: {“average_precision”: {“person”: 0.8148}}, “graphical”: {“loss”: 0.056647732853889465, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.07057582, “mean average precision”: 0.8148}}
{“epoch”: 51, “max_epoch”: 100, “time_per_epoch”: “0:03:13.827450”, “eta”: “2:38:17.545028”, “date”: “9/30/2025”, “time”: “5:22:39”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8148}}, “graphical”: {“loss”: 0.0622844323515892, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.07057582, “mean average precision”: 0.8148}}
{“epoch”: 52, “max_epoch”: 100, “time_per_epoch”: “0:02:09.983496”, “eta”: “1:43:59.207829”, “date”: “9/30/2025”, “time”: “5:24:49”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8148}}, “graphical”: {“loss”: 0.06122328341007233, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.07057582, “mean average precision”: 0.8148}}
{“epoch”: 53, “max_epoch”: 100, “time_per_epoch”: “0:02:08.735008”, “eta”: “1:40:50.545376”, “date”: “9/30/2025”, “time”: “5:26:58”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8148}}, “graphical”: {“loss”: 0.05108253285288811, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.07057582, “mean average precision”: 0.8148}}
{“epoch”: 54, “max_epoch”: 100, “time_per_epoch”: “0:02:09.974349”, “eta”: “1:39:38.820066”, “date”: “9/30/2025”, “time”: “5:29:8”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8148}}, “graphical”: {“loss”: 0.06568079441785812, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.07057582, “mean average precision”: 0.8148}}
{“epoch”: 55, “max_epoch”: 100, “time_per_epoch”: “0:02:09.820657”, “eta”: “1:37:21.929544”, “date”: “9/30/2025”, “time”: “5:31:18”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8148}}, “graphical”: {“loss”: 0.05669553950428963, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.07057582, “mean average precision”: 0.8148}}
{“epoch”: 56, “max_epoch”: 100, “time_per_epoch”: “0:02:10.073460”, “eta”: “1:35:23.232224”, “date”: “9/30/2025”, “time”: “5:33:28”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8148}}, “graphical”: {“loss”: 0.05038151517510414, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.07057582, “mean average precision”: 0.8148}}
{“epoch”: 57, “max_epoch”: 100, “time_per_epoch”: “0:02:09.863137”, “eta”: “1:33:04.114881”, “date”: “9/30/2025”, “time”: “5:35:38”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8148}}, “graphical”: {“loss”: 0.05699966102838516, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.07057582, “mean average precision”: 0.8148}}
{“epoch”: 58, “max_epoch”: 100, “time_per_epoch”: “0:02:10.016340”, “eta”: “1:31:00.686271”, “date”: “9/30/2025”, “time”: “5:37:48”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8148}}, “graphical”: {“loss”: 0.05744409188628197, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.07057582, “mean average precision”: 0.8148}}
{“epoch”: 59, “max_epoch”: 100, “time_per_epoch”: “0:02:09.960656”, “eta”: “1:28:48.386903”, “date”: “9/30/2025”, “time”: “5:39:57”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8148}}, “graphical”: {“loss”: 0.05008644238114357, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.07057582, “mean average precision”: 0.8148}}
{“epoch”: 60, “max_epoch”: 100, “time_per_epoch”: “0:02:11.079690”, “eta”: “1:27:23.187599”, “date”: “9/30/2025”, “time”: “5:42:9”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 0.8148}}, “graphical”: {“loss”: 0.05636274814605713, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.07057582, “mean average precision”: 0.8148}}
{“date”: “9/30/2025”, “time”: “5:45:22”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Evaluation metrics generated.”, “categorical”: {“average_precision”: {“person”: 1.3909}}, “graphical”: {“loss”: 0.05636274814605713, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.05145665, “mean average precision”: 1.3909}}
{“epoch”: 61, “max_epoch”: 100, “time_per_epoch”: “0:03:13.940118”, “eta”: “2:06:03.664586”, “date”: “9/30/2025”, “time”: “5:45:23”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.3909}}, “graphical”: {“loss”: 0.055028483271598816, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.05145665, “mean average precision”: 1.3909}}
{“epoch”: 62, “max_epoch”: 100, “time_per_epoch”: “0:02:09.410680”, “eta”: “1:21:57.605824”, “date”: “9/30/2025”, “time”: “5:47:32”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.3909}}, “graphical”: {“loss”: 0.06432587653398514, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.05145665, “mean average precision”: 1.3909}}
{“epoch”: 63, “max_epoch”: 100, “time_per_epoch”: “0:02:09.974692”, “eta”: “1:20:09.063599”, “date”: “9/30/2025”, “time”: “5:49:42”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.3909}}, “graphical”: {“loss”: 0.055571895092725754, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.05145665, “mean average precision”: 1.3909}}
{“epoch”: 64, “max_epoch”: 100, “time_per_epoch”: “0:02:09.963825”, “eta”: “1:17:58.697708”, “date”: “9/30/2025”, “time”: “5:51:52”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.3909}}, “graphical”: {“loss”: 0.04877893999218941, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.05145665, “mean average precision”: 1.3909}}
{“epoch”: 65, “max_epoch”: 100, “time_per_epoch”: “0:02:09.663894”, “eta”: “1:15:38.236296”, “date”: “9/30/2025”, “time”: “5:54:2”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.3909}}, “graphical”: {“loss”: 0.05168722942471504, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.05145665, “mean average precision”: 1.3909}}
{“epoch”: 66, “max_epoch”: 100, “time_per_epoch”: “0:02:09.924422”, “eta”: “1:13:37.430333”, “date”: “9/30/2025”, “time”: “5:56:11”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.3909}}, “graphical”: {“loss”: 0.053638607263565063, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.05145665, “mean average precision”: 1.3909}}
{“epoch”: 67, “max_epoch”: 100, “time_per_epoch”: “0:02:09.893441”, “eta”: “1:11:26.483552”, “date”: “9/30/2025”, “time”: “5:58:21”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.3909}}, “graphical”: {“loss”: 0.05979212373495102, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.05145665, “mean average precision”: 1.3909}}
{“epoch”: 68, “max_epoch”: 100, “time_per_epoch”: “0:02:09.818014”, “eta”: “1:09:14.176445”, “date”: “9/30/2025”, “time”: “6:0:31”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.3909}}, “graphical”: {“loss”: 0.05115517973899841, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.05145665, “mean average precision”: 1.3909}}
{“epoch”: 69, “max_epoch”: 100, “time_per_epoch”: “0:02:09.830847”, “eta”: “1:07:04.756243”, “date”: “9/30/2025”, “time”: “6:2:41”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.3909}}, “graphical”: {“loss”: 0.06259104609489441, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.05145665, “mean average precision”: 1.3909}}
{“epoch”: 70, “max_epoch”: 100, “time_per_epoch”: “0:02:10.790578”, “eta”: “1:05:23.717351”, “date”: “9/30/2025”, “time”: “6:4:52”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.3909}}, “graphical”: {“loss”: 0.059487227350473404, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.05145665, “mean average precision”: 1.3909}}
{“date”: “9/30/2025”, “time”: “6:8:5”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Evaluation metrics generated.”, “categorical”: {“average_precision”: {“person”: 1.1293}}, “graphical”: {“loss”: 0.059487227350473404, “learning_rate”: “0.00049999997”}, “kpi”: {“validation cost”: 0.05468823, “mean average precision”: 1.1293}}
{“epoch”: 71, “max_epoch”: 100, “time_per_epoch”: “0:03:14.106615”, “eta”: “1:33:49.091830”, “date”: “9/30/2025”, “time”: “6:8:6”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.1293}}, “graphical”: {“loss”: 0.059290021657943726, “learning_rate”: “0.00042884806”}, “kpi”: {“validation cost”: 0.05468823, “mean average precision”: 1.1293}}
{“epoch”: 72, “max_epoch”: 100, “time_per_epoch”: “0:02:09.636771”, “eta”: “1:00:29.829580”, “date”: “9/30/2025”, “time”: “6:10:16”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.1293}}, “graphical”: {“loss”: 0.046074651181697845, “learning_rate”: “0.0003678212”}, “kpi”: {“validation cost”: 0.05468823, “mean average precision”: 1.1293}}
{“epoch”: 73, “max_epoch”: 100, “time_per_epoch”: “0:02:09.877580”, “eta”: “0:58:26.694664”, “date”: “9/30/2025”, “time”: “6:12:25”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.1293}}, “graphical”: {“loss”: 0.04747997969388962, “learning_rate”: “0.00031547897”}, “kpi”: {“validation cost”: 0.05468823, “mean average precision”: 1.1293}}
{“epoch”: 74, “max_epoch”: 100, “time_per_epoch”: “0:02:09.899544”, “eta”: “0:56:17.388156”, “date”: “9/30/2025”, “time”: “6:14:35”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.1293}}, “graphical”: {“loss”: 0.045363109558820724, “learning_rate”: “0.0002705846”}, “kpi”: {“validation cost”: 0.05468823, “mean average precision”: 1.1293}}
{“epoch”: 75, “max_epoch”: 100, “time_per_epoch”: “0:02:09.693707”, “eta”: “0:54:02.342675”, “date”: “9/30/2025”, “time”: “6:16:45”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.1293}}, “graphical”: {“loss”: 0.03633156046271324, “learning_rate”: “0.00023207937”}, “kpi”: {“validation cost”: 0.05468823, “mean average precision”: 1.1293}}
{“epoch”: 76, “max_epoch”: 100, “time_per_epoch”: “0:02:10.028047”, “eta”: “0:52:00.673124”, “date”: “9/30/2025”, “time”: “6:18:55”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.1293}}, “graphical”: {“loss”: 0.04404548555612564, “learning_rate”: “0.0001990536”}, “kpi”: {“validation cost”: 0.05468823, “mean average precision”: 1.1293}}
{“epoch”: 77, “max_epoch”: 100, “time_per_epoch”: “0:02:09.716488”, “eta”: “0:49:43.479216”, “date”: “9/30/2025”, “time”: “6:21:5”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.1293}}, “graphical”: {“loss”: 0.03986875340342522, “learning_rate”: “0.00017072751”}, “kpi”: {“validation cost”: 0.05468823, “mean average precision”: 1.1293}}
{“epoch”: 78, “max_epoch”: 100, “time_per_epoch”: “0:02:10.016083”, “eta”: “0:47:40.353826”, “date”: “9/30/2025”, “time”: “6:23:15”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.1293}}, “graphical”: {“loss”: 0.03458838909864426, “learning_rate”: “0.00014643234”}, “kpi”: {“validation cost”: 0.05468823, “mean average precision”: 1.1293}}
{“epoch”: 79, “max_epoch”: 100, “time_per_epoch”: “0:02:09.884801”, “eta”: “0:45:27.580824”, “date”: “9/30/2025”, “time”: “6:25:25”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.1293}}, “graphical”: {“loss”: 0.040294330567121506, “learning_rate”: “0.00012559434”}, “kpi”: {“validation cost”: 0.05468823, “mean average precision”: 1.1293}}
{“epoch”: 80, “max_epoch”: 100, “time_per_epoch”: “0:02:10.686687”, “eta”: “0:43:33.733745”, “date”: “9/30/2025”, “time”: “6:27:35”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 1.1293}}, “graphical”: {“loss”: 0.03988732770085335, “learning_rate”: “0.00010772168”}, “kpi”: {“validation cost”: 0.05468823, “mean average precision”: 1.1293}}
{“date”: “9/30/2025”, “time”: “6:30:50”, “status”: “RUNNING”, “verbosity”: “INFO”, “message”: “Evaluation metrics generated.”, “categorical”: {“average_precision”: {“person”: 4.0952}}, “graphical”: {“loss”: 0.03988732770085335, “learning_rate”: “0.00010772168”}, “kpi”: {“validation cost”: 0.03309437, “mean average precision”: 4.0952}}
{“epoch”: 81, “max_epoch”: 100, “time_per_epoch”: “0:03:14.714617”, “eta”: “1:01:39.577719”, “date”: “9/30/2025”, “time”: “6:30:50”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 4.0952}}, “graphical”: {“loss”: 0.040571920573711395, “learning_rate”: “9.239248e-05”}, “kpi”: {“validation cost”: 0.03309437, “mean average precision”: 4.0952}}
{“epoch”: 82, “max_epoch”: 100, “time_per_epoch”: “0:02:09.591179”, “eta”: “0:38:52.641220”, “date”: “9/30/2025”, “time”: “6:33:0”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 4.0952}}, “graphical”: {“loss”: 0.036677874624729156, “learning_rate”: “7.924468e-05”}, “kpi”: {“validation cost”: 0.03309437, “mean average precision”: 4.0952}}
{“epoch”: 83, “max_epoch”: 100, “time_per_epoch”: “0:02:09.995054”, “eta”: “0:36:49.915914”, “date”: “9/30/2025”, “time”: “6:35:10”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 4.0952}}, “graphical”: {“loss”: 0.03213275223970413, “learning_rate”: “6.7967856e-05”}, “kpi”: {“validation cost”: 0.03309437, “mean average precision”: 4.0952}}
{“epoch”: 84, “max_epoch”: 100, “time_per_epoch”: “0:02:09.870754”, “eta”: “0:34:37.932056”, “date”: “9/30/2025”, “time”: “6:37:20”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 4.0952}}, “graphical”: {“loss”: 0.040249571204185486, “learning_rate”: “5.829572e-05”}, “kpi”: {“validation cost”: 0.03309437, “mean average precision”: 4.0952}}
{“epoch”: 85, “max_epoch”: 100, “time_per_epoch”: “0:02:09.903138”, “eta”: “0:32:28.547065”, “date”: “9/30/2025”, “time”: “6:39:29”, “status”: “RUNNING”, “verbosity”: “INFO”, “categorical”: {“average_precision”: {“person”: 4.0952}}, “graphical”: {“loss”: 0.041655246168375015, “learning_rate”: “5.000002e-05”}, “kpi”: {“validation cost”: 0.03309437, “mean average precision”: 4.0952}}

• Just trained from roboflow dataset pulled as coco and converted with tao itself.

2

Yes, it is just 4% now.

3


neeraj.sj

1m

Thanks and are there any suggestion to get best training possible with spec file I have sent, I need it to perform like trafficcamnet however with a single class and my custom dataset. I am asking this because on converting to Tensorrt the trafficamnet version gives me perfetc QPS of 1756, which is my need.

4

I suggest you train with peoplenet pretrained model instead of trafficcamnet.
For peoplenet model, please refer to one of the trainable version: https://catalog.ngc.nvidia.com/orgs/nvidia/teams/tao/models/peoplenet/files?version=trainable_unencrypted_v2.6.

For the spec file, you may search topics in TAO forum. For example, you can find some experience shared in https://forums.developer.nvidia.com/t/training-results-are-not-good-for-peoplenet-pretrained-model/343678.

5

Ok for the trafficCam net I want only the head part so my annotations are only for head, if i take traffic cam net as pretrained would it not lose the accuracy due to domain in-adaptabilty

6

The peoplenet model is better since its most class is people.
But the most class of the trafficcamnet model is car.

7

Since face is the nearest class that is similar to head class, how can i configure the face class from pretrained model and head from dataset in spec file, and even if i succeed with this what will happen to the other two classes handbag and person.

8

If you are going to train only person class, please set only person class in the spec file. Please use peoplenet as pretrained model. The example spec file can be found in above-mentioned topic.

9

There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks.