Please provide the following information when requesting support.
• Hardware : Triton inference server
• Network Type :Yolo_v4 for People detection
• Training spec file
random_seed: 42
yolov4_config {
# big_anchor_shape: "[(114.94, 60.67), (159.06, 114.59), (297.59, 176.38)]"
# mid_anchor_shape: "[(42.99, 31.91), (79.57, 31.75), (56.80, 56.93)]"
# small_anchor_shape: "[(15.60, 13.88), (30.25, 20.25), (20.67, 49.63)]"
big_anchor_shape: "[(56.00, 136.00), (95.00, 167.00), (170.00, 212.00)]"
mid_anchor_shape: "[(51.00, 64.00), (36.00, 109.00), (53.00, 98.00)]"
small_anchor_shape: "[(21.00, 39.00), (35.00, 43.00), (30.00, 66.00)]"
box_matching_iou: 0.25
matching_neutral_box_iou: 0.5
arch: "resnet"
nlayers: 18
arch_conv_blocks: 2
loss_loc_weight: 1.0
loss_neg_obj_weights: 1.0
loss_class_weights: 1.0
label_smoothing: 0.0
big_grid_xy_extend: 0.05
mid_grid_xy_extend: 0.1
small_grid_xy_extend: 0.2
freeze_bn: false
#freeze_blocks: 0
force_relu: false
}
training_config {
batch_size_per_gpu: 8
num_epochs: 80
enable_qat: false
checkpoint_interval: 2
learning_rate {
soft_start_cosine_annealing_schedule {
min_learning_rate: 1e-7
max_learning_rate: 1e-4
soft_start: 0.3
}
}
regularizer {
type: L1
weight: 3e-5
}
optimizer {
adam {
epsilon: 1e-7
beta1: 0.9
beta2: 0.999
amsgrad: false
}
}
pretrain_model_path: "/workspace/tao-experiments/yolo_v4/pretrained_resnet18/pretrained_object_detection_vresnet18/resnet_18.hdf5"
#resume_model_path: "/workspace/tao-experiments/yolo_v4/experiment_dir_unpruned/weights/yolov4_resnet18_epoch_054.tlt"
}
eval_config {
average_precision_mode: SAMPLE
batch_size: 8
matching_iou_threshold: 0.5
}
nms_config {
confidence_threshold: 0.001
clustering_iou_threshold: 0.5
force_on_cpu: true
top_k: 200
}
augmentation_config {
hue: 0.1
saturation: 1.5
exposure:1.5
vertical_flip:0
horizontal_flip: 0.5
jitter: 0.3
#output_width: 1248
#output_height: 384
output_width: 960
output_height: 544
output_channel: 3
randomize_input_shape_period: 0
mosaic_prob: 0.5
mosaic_min_ratio:0.2
}
dataset_config {
data_sources: {
tfrecords_path: "/workspace/tao-experiments/data/training/tfrecords/train*"
image_directory_path: "/workspace/tao-experiments/data/training"
}
include_difficult_in_training: true
image_extension: "jpg"
#target_class_mapping {
# key: "car"
# value: "car"
#}
target_class_mapping {
key: "person"
value: "pedestrian"
}
# target_class_mapping {
# key: "cyclist"
# value: "cyclist"
#}
#target_class_mapping {
# key: "van"
# value: "car"
#}
#target_class_mapping {
# key: "person_sitting"
# value: "pedestrian"
#}
validation_data_sources: {
tfrecords_path: "/workspace/tao-experiments/data/val/tfrecords/val*"
image_directory_path: "/workspace/tao-experiments/data/val"
}
}
• How to reproduce the issue ? (This is for errors. Please share the command line and the detailed log here.)
I am using the Tao toolkit Jupyter Notebook for Yolo v4 for custom training object detection on our data set.
The training is sucessfully completed. However when i try executing the Evaluate Trained module section
!tao yolo_v4 evaluate -e $SPECS_DIR/yolo_v4_train_resnet18_kitti.txt \
-m $USER_EXPERIMENT_DIR/experiment_dir_unpruned/weights/yolov4_resnet18_epoch_$EPOCH.tlt \
-k $KEY
It does not show the out put sometimes
Following is the screenshot of the same.
Can you tell what could be the problem as to why it is not evaluating the trained model. Note: Sometimes it does shows me the output predictions and sometimes the container just stops
