rknn例子中ssd_mobilenet_v2修改ssd_mobilenet_v2.quantization.cfg的问题 - Toybrick

# SSD with Mobilenet v2, configured for egohands dataset.
# This file was extracted modified from 'pipeline.config' in
# http://download.tensorflow.org/models/object_detection/ssd_mobilenet_v2_coco_2018_03_29.tar.gz
model {
ssd {
num_classes: 2
box_coder {
faster_rcnn_box_coder {
y_scale: 10.0
x_scale: 10.0
height_scale: 5.0
width_scale: 5.0
}
}
matcher {
argmax_matcher {
matched_threshold: 0.5
unmatched_threshold: 0.5
ignore_thresholds: false
negatives_lower_than_unmatched: true
force_match_for_each_row: true
}
}
similarity_calculator {
iou_similarity {
}
}
anchor_generator {
ssd_anchor_generator {
num_layers: 6
min_scale: 0.05
max_scale: 0.95
aspect_ratios: 1.0
aspect_ratios: 2.0
aspect_ratios: 0.5
aspect_ratios: 3.0
aspect_ratios: 0.3333
}
}
image_resizer {
fixed_shape_resizer {
height: 300
width: 300
}
}
box_predictor {
convolutional_box_predictor {
min_depth: 0
max_depth: 0
num_layers_before_predictor: 0
use_dropout: false
dropout_keep_probability: 0.8
kernel_size: 3
box_code_size: 4
apply_sigmoid_to_scores: false
conv_hyperparams {
activation: RELU_6
regularizer {
l2_regularizer {
weight: 0.00004
}
}
initializer {
truncated_normal_initializer {
stddev: 0.03
mean: 0.0
}
}
batch_norm {
train: true
scale: true
center: true
decay: 0.9997
epsilon: 0.001
}
}
}
}
feature_extractor {
type: "ssd_mobilenet_v2"
min_depth: 16
depth_multiplier: 1.0
conv_hyperparams {
activation: RELU_6
regularizer {
l2_regularizer {
weight: 4e-05
}
}
initializer {
truncated_normal_initializer {
stddev: 0.03
mean: 0.0
}
}
batch_norm {
train: true
scale: true
center: true
decay: 0.9997
epsilon: 0.001
}
}
#batch_norm_trainable: true
use_depthwise: true
}
loss {
classification_loss {
weighted_sigmoid {
}
}
localization_loss {
weighted_smooth_l1 {
}
}
hard_example_miner {
num_hard_examples: 3000
iou_threshold: 0.99
loss_type: CLASSIFICATION
max_negatives_per_positive: 3
min_negatives_per_image: 3
}
classification_weight: 1.0
localization_weight: 1.0
}
normalize_loss_by_num_matches: true
post_processing {
batch_non_max_suppression {
score_threshold: 1e-8
iou_threshold: 0.6
max_detections_per_class: 100
max_total_detections: 100
}
score_converter: SIGMOID
}
}
}
train_config {
batch_size: 24
optimizer {
rms_prop_optimizer {
learning_rate {
exponential_decay_learning_rate {
initial_learning_rate: 0.004
decay_steps: 1000
decay_factor: 0.8
}
}
momentum_optimizer_value: 0.9
decay: 0.9
epsilon: 1.0
}
}
fine_tune_checkpoint: "ssd_mobilenet_v2_coco_2018_03_29/model.ckpt"
fine_tune_checkpoint_type: "detection"
num_steps: 20000
data_augmentation_options {
random_horizontal_flip {
}
}
data_augmentation_options {
ssd_random_crop {
}
}
}
train_input_reader {
tf_record_input_reader {
input_path: "data/pill_case.tfrecord"
}
label_map_path: "data/pill_case_label_map.txt"
}
eval_config {
num_examples: 500
max_evals: 10
use_moving_averages: false
}
eval_input_reader {
tf_record_input_reader {
input_path: "data/pill_case.tfrecord"
}
label_map_path: "data/pill_case_label_map.txt"
shuffle: false
num_readers: 1
}