- Clone the Tensorflow Research repository
- As a sandbox test, build your VSCode project in the
/researchdirectory - Within your project, prepare three main folders, they are
/annotations,/images, and/data - Annotate your images with online annotation tool such as Lotus, labelimg and many more.
- Inside
/annotationsgo insert all .xml outputs from your annotation - Convert your .xml to unitary .csv file with code below. Make sure to store your .xml data in
/annotations. Oh by the way, this code should be built as a separated file outside the three main folders, aforementioned. You can name it asconversion.py(PS: Don't forget to configure your Python and Tensorflow for VSCode if you use this editor)
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'):
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 = os.path.join(os.getcwd(), 'annotations')
xml_df = xml_to_csv(image_path)
xml_df.to_csv('data_jagung.csv', index=None)
print('Successfully converted xml to csv.')
main()
- Once the conversion finished, you can see the .csv right away. Can't you?
- Its time convert your .csv to .TFRecord file with this code below. Anyway, name it
generator.pyfor short haha
Usage:
# From tensorflow/models/
# Create train data:
python generate_tfrecord.py --csv_input=data/train_labels.csv --output_path=train.record
# Create test data:
python generate_tfrecord.py --csv_input=data/test_labels.csv --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('output_path', '', 'Path to output TFRecord')
flags.DEFINE_string('image_dir', '', 'Path to images')
FLAGS = flags.FLAGS
def class_text_to_int(row_label):
if row_label == 'jagung':
return 1
else:
return 0
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(_):
print(os.getcwd())
writer = tf.python_io.TFRecordWriter('data/test.record')
# path = os.path.join(FLAGS.image_dir)
path = 'images'
# examples = pd.read_csv(FLAGS.csv_input)
examples = pd.read_csv('data_jagung.csv')
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()
- There you go, tfrecord is available on your
datafolder
ARCHITECTURE="mobilenet_0.50_${IMAGE_SIZE}"
python -m scripts.retrain \
--bottleneck_dir=tf_files/bottlenecks \
--how_many_training_steps=500 \
--model_dir=tf_files/models/ \
--summaries_dir=tf_files/training_summaries/"${ARCHITECTURE}" \
--output_graph=tf_files/retrained_graph_jagung.pb \
--output_labels=tf_files/retrained_labels_jagung.txt \
--architecture="${ARCHITECTURE}" \
--image_dir=tf_files/foto_jagung
IMAGE_SIZE=224
tflite_convert \
--graph_def_file=tf_files/retrained_graph.pb \
--output_file=tf_files/custom_folowers.tflite \
--input_format=TENSORFLOW_GRAPHDEF \
--output_format=TFLITE \
--input_shape=1,${IMAGE_SIZE},${IMAGE_SIZE},3 \
--input_array=input \
--output_array=final_result \
--inference_type=FLOAT \
--input_data_type=FLOAT
bazel-bin/tensorflow/contrib/lite/toco/toco \
- -input_format=TENSORFLOW_GRAPHDEF \
--input_file=tf_files/restrained_graph.pb \
--output_format=TFLITE \
--output_file=tf_files/custom_flowers.tflite \
--inference_type=QUANTIZED_UINT8 \
--inference_input_type=QUANTIZED_UINT8 \
--input_arrays=input \
--output_arrays=final_result \
--input_shapes=1,224,224,3\
--mean_values=128 \
--std_values=128 \
--default_ranges_min=0 \
--default_ranges_max=6
Additional Notes: Use venv to avoid missing components during training and conversion (optional)
source ./venv/bin/activate