Ale*_*exU 4 python machine-learning object-detection tensorflow
该模型过度拟合训练集，无法泛化到测试集。
如何将 dropout 添加到模型的特征提取器部分？（.config 文件只提供了一个键值来将 dropout 添加到框预测器中）
我可以采取哪些其他措施来最大程度地减少过度拟合？
更多详情如下：
我正在尝试在玩具动物数据集上重新训练模型检查点“ssd_mobilenet_v1_coco_11_06_2017”。有 14 个类，每个类有 400-600 个图像。网络在不到 3 万步的时间内学习训练集。张量板。在最初的训练之后，损失似乎仍然相当不稳定，尽管我没有足够的经验来评估这一点。
我正在通过将导出的图形应用于图像并手动检查结果来测试模型。（我只是没有时间正确实施验证）。该模型在与训练集非常相似的条件下拍摄的照片上效果很好。这些不好的测试集图像是从训练集随机删除的，训练集是通过连续拍摄许多图像并稍微改变相机角度而获得的。训练集还包括各种光照条件、背景、扭曲和相机位置。我估计它在来自坏测试集的大约 95% 的图像中获得了正确的类和位置。由此我得出结论，该模型非常适合训练集，并且可以进行一些概括。
然而，该模型在不同时间用不同相机单独拍摄的图片上表现非常差（即该测试集和训练集之间的相关性应该小得多）。我估计在这个好的测试集中的性能大约是 25%。由此我得出结论，该模型过度拟合且无法泛化。
我尝试在 .config 文件中进行一些更改。
将特征提取器和框预测器的 l2_regularizer 权重从 0.00004 增加到 0.0001。
盒子预测设置use_dropout到true以使20％辍学。
我正在使用 Tensorflow 1.4 pip install 和大约 3 周前从 github 克隆的模型。
我使用以下参数调用 object_detection 中的 train.py：
python train.py --logtostderr --train_dir=/home/X/TrainDir/Process --pipeline_config_path=/home/X/ssd_mobilenet_v1_coco.config
我的配置文件如下：
# SSD with Mobilenet v1 configuration for MSCOCO Dataset.
# Users should configure the fine_tune_checkpoint field in the train config as
# well as the label_map_path and input_path fields in the train_input_reader and
# eval_input_reader. Search for "PATH_TO_BE_CONFIGURED" to find the fields that
# should be configured.
model {
ssd {
num_classes: 14
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: 0.2
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: true
dropout_keep_probability: 0.8
kernel_size: 1
box_code_size: 4
apply_sigmoid_to_scores: false
conv_hyperparams {
activation: RELU_6,
regularizer {
l2_regularizer {
weight: 0.0001
}
}
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_v1'
min_depth: 16
depth_multiplier: 1.0
conv_hyperparams {
activation: RELU_6,
regularizer {
l2_regularizer {
weight: 0.0001
}
}
initializer {
truncated_normal_initializer {
stddev: 0.03
mean: 0.0
}
}
batch_norm {
train: true,
scale: true,
center: true,
decay: 0.9997,
epsilon: 0.001,
}
}
}
loss {
classification_loss {
weighted_sigmoid {
anchorwise_output: true
}
}
localization_loss {
weighted_smooth_l1 {
anchorwise_output: true
}
}
hard_example_miner {
num_hard_examples: 3000
iou_threshold: 0.99
loss_type: CLASSIFICATION
max_negatives_per_positive: 3
min_negatives_per_image: 0
}
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: 8
optimizer {
rms_prop_optimizer: {
learning_rate: {
exponential_decay_learning_rate {
initial_learning_rate: 0.004
decay_steps: 800720
decay_factor: 0.95
}
}
momentum_optimizer_value: 0.9
decay: 0.9
epsilon: 1.0
}
}
fine_tune_checkpoint: "/home/X/tensorflow/models/research/object_detection/ssd_mobilenet_v1_coco_11_06_2017/model.ckpt"
from_detection_checkpoint: true
# Note: The below line limits the training process to 200K steps, which we
# empirically found to be sufficient enough to train the pets dataset. This
# effectively bypasses the learning rate schedule (the learning rate will
# never decay). Remove the below line to train indefinitely.
num_steps: 200000
data_augmentation_options {
random_horizontal_flip {
}
}
data_augmentation_options {
ssd_random_crop {
}
}
}
train_input_reader: {
tf_record_input_reader {
input_path: "/home/X/TrainDir/train.record"
}
label_map_path: "/home/X/TrainDir/data_label_map.pbtxt"
}
eval_config: {
num_examples: 1200
# Note: The below line limits the evaluation process to 10 evaluations.
# Remove the below line to evaluate indefinitely.
max_evals: 30
}
eval_input_reader: {
tf_record_input_reader {
input_path: "/home/X/TrainDir/test.record"
}
label_map_path: "/home/X/TrainDir/data_label_map.pbtxt"
shuffle: false
num_readers: 1
num_epochs: 1
}
经过一些技巧后，网络学习得很好，并开始在好的测试集上进行泛化。
经过训练的网络在真实世界场景中表现良好，对这些动物进行在线检测，而照片是在全新的场景和光照条件下拍摄的。
以下添加于 06/03/2020
针对评论中的要求，我挖出了我在这个项目中存储的配置文件（> 2年前）。这很可能是我最终使用的最终配置，效果很好。
# SSD with Mobilenet v1 configuration for MSCOCO Dataset.
# Users should configure the fine_tune_checkpoint field in the train config as
# well as the label_map_path and input_path fields in the train_input_reader and
# eval_input_reader. Search for "PATH_TO_BE_CONFIGURED" to find the fields that
# should be configured.
model {
ssd {
num_classes: 14
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: 0.2
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: true
dropout_keep_probability: 0.8
kernel_size: 1
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_v1'
min_depth: 16
depth_multiplier: 1.0
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,
}
}
}
loss {
classification_loss {
weighted_sigmoid {
anchorwise_output: true
}
}
localization_loss {
weighted_smooth_l1 {
anchorwise_output: true
}
}
hard_example_miner {
num_hard_examples: 3000
iou_threshold: 0.99
loss_type: CLASSIFICATION
max_negatives_per_positive: 3
min_negatives_per_image: 0
}
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: 8
optimizer {
rms_prop_optimizer: {
learning_rate: {
exponential_decay_learning_rate {
initial_learning_rate: 0.004
decay_steps: 7000
decay_factor: 0.75
}
}
momentum_optimizer_value: 0.9
decay: 0.9
epsilon: 1.0
}
}
fine_tune_checkpoint: "/home/sander/tensorflow/models/research/object_detection/ssd_mobilenet_v1_coco_11_06_2017/model.ckpt"
from_detection_checkpoint: true
# Note: The below line limits the training process to 200K steps, which we
# empirically found to be sufficient enough to train the pets dataset. This
# effectively bypasses the learning rate schedule (the learning rate will
# never decay). Remove the below line to train indefinitely.
num_steps: 200000
data_augmentation_options {
random_horizontal_flip {
}
}
data_augmentation_options {
ssd_random_crop {
}
}
}
train_input_reader: {
tf_record_input_reader {
input_path: "/home/sander/ROBOT/TrainDir/train.record"
}
label_map_path: "/home/sander/ROBOT/TrainDir/data_label_map.pbtxt"
}
eval_config: {
num_examples: 1200
# Note: The below line limits the evaluation process to 10 evaluations.
# Remove the below line to evaluate indefinitely.
max_evals: 30
}
eval_input_reader: {
tf_record_input_reader {
input_path: "/home/sander/ROBOT/TrainDir/test.record"
}
label_map_path: "/home/sander/ROBOT/TrainDir/data_label_map.pbtxt"
shuffle: false
num_readers: 1
num_epochs: 1
}
您是否尝试过设置验证集并使用部分训练的模型与训练并行自动运行评估？如果验证准确性尚未收敛，那么您可能只需要更长时间地训练模型即可。
其他需要注意的事项：“好”和“坏”测试集之间是否还有其他差异？例如分辨率/纵横比。您是否正确遵循了所有预处理步骤，与训练期间执行这些步骤的方式完全相同？例如数据标准化、使用相同算法调整大小等......
编辑：我检查了你的张量板屏幕截图，为什么你认为网络学习了你的训练集？损失似乎并没有真正收敛。你绝对应该做的另一件事是设置一个调度程序来减少你的学习除以 10，比如说每 40K 步，在学习了一些特征之后，你的梯度下降可能会难以收敛，因为你永远不会改变起始值的学习率，并且对于训练中的那个时间点来说可能太大了。
| 归档时间： |
|
| 查看次数： |
4018 次 |
| 最近记录： |