working with TF (Tensor Flow) Object Detection API.
I'm training a custom dataset with TF v1.9 and TF Models recent version, 2018/07/20.
OS: Windows 10. TF: v1.9 Only CPU. Installed via pip.
I've followed a mix of various tutorials:
https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/installation.md https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/running_pets.md https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/preparing_inputs.md https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/running_locally.md
And one very useful to get started:
However TF Models repo has changed some files, leaving some to the legacy folder. That last tutorial is only useful with a specific old commit.
I've tried to adapt to the current version calling model_main.py with proper parameters, and setting every parameter, directory, file, etc, in pipeline_config .
My data set has 594 images for training, and 150 for evaluating. One class. Using model ssdlite_mobilenet_v2_coco.
Labels were set with LabelImg tool, then created the CSV, and then converted to tfrecord data with the generate_tf_record script.
After a while the only output I've have are several lines like this:
WARNING:tensorflow:Ignoring ground truth with image id 534047036 since it was previously added
WARNING:tensorflow:Ignoring detection with image id 534047036 since it was previously added
and some times,
creating index...
index created!
creating index...
index created!
Running per image evaluation...
Evaluate annotation type *bbox*
DONE (t=0.69s).
Accumulating evaluation results...
DONE (t=0.11s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.000
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.003
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.000
Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.026
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.000
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.000
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.015
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.071
Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = -1.000
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.030
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.083
WARNING:tensorflow:num_readers has been reduced to 1 to match input file shards.
Tensorboard informs: No dashboards are active for the current data set. No checkpoint files are generated. It runs for hours/days with that output I've checked two, three, several times the configuration.
Is this a normal behavior / output? Why there is no output files, checkpoints, etc?
ssdlite_mobilenet_v2_coco.config :
model {
ssd {
num_classes: 1
image_resizer {
fixed_shape_resizer {
height: 300
width: 300
}
}
feature_extractor {
type: "ssd_mobilenet_v2"
depth_multiplier: 1.0
min_depth: 16
conv_hyperparams {
regularizer {
l2_regularizer {
weight: 3.99999989895e-05
}
}
initializer {
truncated_normal_initializer {
mean: 0.0
stddev: 0.0299999993294
}
}
activation: RELU_6
batch_norm {
decay: 0.999700009823
center: true
scale: true
epsilon: 0.0010000000475
train: true
}
}
use_depthwise: true
}
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 {
}
}
box_predictor {
convolutional_box_predictor {
conv_hyperparams {
regularizer {
l2_regularizer {
weight: 3.99999989895e-05
}
}
initializer {
truncated_normal_initializer {
mean: 0.0
stddev: 0.0299999993294
}
}
activation: RELU_6
batch_norm {
decay: 0.999700009823
center: true
scale: true
epsilon: 0.0010000000475
train: true
}
}
min_depth: 0
max_depth: 0
num_layers_before_predictor: 0
use_dropout: false
dropout_keep_probability: 0.800000011921
kernel_size: 3
box_code_size: 4
apply_sigmoid_to_scores: false
use_depthwise: true
}
}
anchor_generator {
ssd_anchor_generator {
num_layers: 6
min_scale: 0.20000000298
max_scale: 0.949999988079
aspect_ratios: 1.0
aspect_ratios: 2.0
aspect_ratios: 0.5
aspect_ratios: 3.0
aspect_ratios: 0.333299994469
}
}
post_processing {
batch_non_max_suppression {
score_threshold: 0.300000011921
iou_threshold: 0.600000023842
max_detections_per_class: 100
max_total_detections: 100
}
score_converter: SIGMOID
}
normalize_loss_by_num_matches: true
loss {
localization_loss {
weighted_smooth_l1 {
}
}
classification_loss {
weighted_sigmoid {
}
}
hard_example_miner {
num_hard_examples: 3000
iou_threshold: 0.990000009537
loss_type: CLASSIFICATION
max_negatives_per_positive: 3
min_negatives_per_image: 3
}
classification_weight: 1.0
localization_weight: 1.0
}
}
}
train_config {
batch_size: 24
data_augmentation_options {
random_horizontal_flip {
}
}
data_augmentation_options {
ssd_random_crop {
}
}
optimizer {
rms_prop_optimizer {
learning_rate {
exponential_decay_learning_rate {
initial_learning_rate: 0.00400000018999
decay_steps: 800720
decay_factor: 0.949999988079
}
}
momentum_optimizer_value: 0.899999976158
decay: 0.899999976158
epsilon: 1.0
}
}
fine_tune_checkpoint: "C:/proyectos/ml/tensorflow/models/tf_model_files/ssdlite_mobilenet_v2_coco_2018_05_09/model.ckpt"
num_steps: 200000
fine_tune_checkpoint_type: "detection"
}
train_input_reader {
label_map_path: "C:/proyectos/ml/tensorflow/models/models-master/research/object_detection/training/labelmap.pbtxt"
tf_record_input_reader {
input_path: "C:/proyectos/ml/samples/lics/training/training.record"
}
}
eval_config {
num_examples: 594
max_evals: 10
use_moving_averages: false
}
eval_input_reader {
label_map_path: "C:/proyectos/ml/tensorflow/models/models-master/research/object_detection/training/labelmap.pbtxt"
shuffle: false
num_readers: 1
tf_record_input_reader {
input_path: "C:/proyectos/ml/samples/lics/testing/testing.record"
}
}
generate_tf_record.py
"""
Usage:
# From tensorflow/models/
# Create train data:
python generate_tfrecord.py --csv_input=images/train_labels.csv --image_dir=images/train --output_path=train.record
# Create test data:
python generate_tfrecord.py --csv_input=images/test_labels.csv --image_dir=images/test --output_path=test.record
"""
from __future__ import division
from __future__ import print_function
from __future__ import absolute_import
import os
import io
import pandas as pd
import tensorflow as tf
from PIL import Image
from object_detection.utils import dataset_util
from collections import namedtuple, OrderedDict
flags = tf.app.flags
flags.DEFINE_string('csv_input', '', 'Path to the CSV input')
flags.DEFINE_string('image_dir', '', 'Path to the image directory')
flags.DEFINE_string('output_path', '', 'Path to output TFRecord')
FLAGS = flags.FLAGS
# TO-DO replace this with label map
def class_text_to_int(row_label):
if row_label == 'pat':
return 1
else:
None
def split(df, group):
data = namedtuple('data', ['filename', 'object'])
gb = df.groupby(group)
return [data(filename, gb.get_group(x)) for filename, x in zip(gb.groups.keys(), gb.groups)]
def create_tf_example(group, path):
with tf.gfile.GFile(os.path.join(path, '{}'.format(group.filename)), 'rb') as fid:
encoded_jpg = fid.read()
encoded_jpg_io = io.BytesIO(encoded_jpg)
image = Image.open(encoded_jpg_io)
width, height = image.size
filename = group.filename.encode('utf8')
image_format = b'jpg'
xmins = []
xmaxs = []
ymins = []
ymaxs = []
classes_text = []
classes = []
for index, row in group.object.iterrows():
xmins.append(row['xmin'] / width)
xmaxs.append(row['xmax'] / width)
ymins.append(row['ymin'] / height)
ymaxs.append(row['ymax'] / height)
classes_text.append(row['class'].encode('utf8'))
classes.append(class_text_to_int(row['class']))
tf_example = tf.train.Example(features=tf.train.Features(feature={
'image/height': dataset_util.int64_feature(height),
'image/width': dataset_util.int64_feature(width),
'image/filename': dataset_util.bytes_feature(filename),
'image/source_id': dataset_util.bytes_feature(filename),
'image/encoded': dataset_util.bytes_feature(encoded_jpg),
'image/format': dataset_util.bytes_feature(image_format),
'image/object/bbox/xmin': dataset_util.float_list_feature(xmins),
'image/object/bbox/xmax': dataset_util.float_list_feature(xmaxs),
'image/object/bbox/ymin': dataset_util.float_list_feature(ymins),
'image/object/bbox/ymax': dataset_util.float_list_feature(ymaxs),
'image/object/class/text': dataset_util.bytes_list_feature(classes_text),
'image/object/class/label': dataset_util.int64_list_feature(classes),
}))
return tf_example
def main(_):
writer = tf.python_io.TFRecordWriter(FLAGS.output_path)
print("!output_path Path="+FLAGS.output_path)
path = os.path.join(os.getcwd(), FLAGS.image_dir)
print("!image_dir="+path)
print("!csv_input="+FLAGS.csv_input)
examples = pd.read_csv(FLAGS.csv_input)
grouped = split(examples, 'filename')
for group in grouped:
tf_example = create_tf_example(group, path)
writer.write(tf_example.SerializeToString())
writer.close()
output_path = os.path.join(os.getcwd(), FLAGS.output_path)
print('Successfully created the TFRecords: {}'.format(output_path))
if __name__ == '__main__':
tf.app.run()
xml_to_csv
import os
import glob
import pandas as pd
import xml.etree.ElementTree as ET
def xml_to_csv(path):
xml_list = []
for xml_file in glob.glob(path + '/*.xml'):
Nombre = os.path.split(xml_file)[1]
NombreJpg = Nombre.split(".")[0] + "jpg"
if not os.path.exists(path+"/"+Nombre):
print("No esta la imagen " + path+"/"+NombreJpg)
continue
print("Adding "+Nombre)
tree = ET.parse(xml_file)
root = tree.getroot()
for member in root.findall('object'):
value = (root.find('filename').text,
int(root.find('size')[0].text),
int(root.find('size')[1].text),
member[0].text,
int(member[4][0].text),
int(member[4][1].text),
int(member[4][2].text),
int(member[4][3].text)
)
xml_list.append(value)
column_name = ['filename', 'width', 'height', 'class', 'xmin', 'ymin', 'xmax', 'ymax']
xml_df = pd.DataFrame(xml_list, columns=column_name)
return xml_df
def main():
image_path = "C:/proyectos/ml/samples/lics/training"
xml_df = xml_to_csv(image_path)
xml_df.to_csv("C:/proyectos/ml/samples/lics/training/training.csv", index=None)
print('Successfully converted training xml to csv.')
image_path = "C:/proyectos/ml/samples/lics/testing"
xml_df = xml_to_csv(image_path)
xml_df.to_csv("C:/proyectos/ml/samples/lics/testing/testing.csv", index=None)
print('Successfully converted testing xml to csv.')
main()
Example xml:
<annotation>
<folder>training</folder>
<filename>20180525_320_135_000142_6.jpg</filename>
<path>C:\proyectos\ml\samples\lics\training\20180525_320_135_000142_6.jpg</path>
<source>
<database>Unknown</database>
</source>
<size>
<width>1280</width>
<height>1024</height>
<depth>3</depth>
</size>
<segmented>0</segmented>
<object>
<name>pat</name>
<pose>Unspecified</pose>
<truncated>0</truncated>
<difficult>0</difficult>
<bndbox>
<xmin>949</xmin>
<ymin>290</ymin>
<xmax>1093</xmax>
<ymax>336</ymax>
</bndbox>
</object>
</annotation>
