def create_dataset():
print('criando dataset')
data = ImageClassifierDataLoader.from_folder('./images_dataset')
train_data, rest_data = data.split(0.8)
validation_data, test_data = rest_data.split(0.5)
model = image_classifier.create(train_data,
validation_data=validation_data,
epochs=10)
model.summary()
model.export(export_dir='model/', export_format=ExportFormat.LABEL)
model.export(export_dir='model/')
def run(data_dir, export_dir, spec='efficientnet_lite0', **kwargs):
"""Runs demo."""
spec = model_spec.get(spec)
data = image_classifier.DataLoader.from_folder(data_dir)
train_data, rest_data = data.split(0.8)
validation_data, test_data = rest_data.split(0.5)
model = image_classifier.create(
train_data, model_spec=spec, validation_data=validation_data, **kwargs)
_, acc = model.evaluate(test_data)
print('Test accuracy: %f' % acc)
# Exports to TFLite and SavedModel, with label file.
export_format = [
ExportFormat.TFLITE,
ExportFormat.SAVED_MODEL,
]
model.export(export_dir, export_format=export_format)
"""generates tfjs model from files"""
import os
import numpy as np
import tensorflow as tf
assert tf.__version__.startswith("2")
from tflite_model_maker import model_spec
from tflite_model_maker import image_classifier
from tflite_model_maker.config import ExportFormat
from tflite_model_maker.config import QuantizationConfig
from tflite_model_maker.image_classifier import DataLoader
import matplotlib.pyplot as plt
IMAGE_PATH = "/home/uni/storage/p-block/"
MODEL_PATH = "/home/uni/repo/p-block/generated-model"
data = DataLoader.from_folder(IMAGE_PATH)
train_data, test_data = data.split(0.9)
model = image_classifier.create(train_data)
# loss, accuracy = model.evaluate(test_data)
# 0.9273504018783569
# 0.3174212872982025
# model.export('/home/uni/repo/p-block/generated-model')
model.export(MODEL_PATH, export_format=ExportFormat.TFJS)
Here, we use EfficientNet Lite1 as a base model for image classification and build custom tflite model
More info, click here:
<br>
https://blog.tensorflow.org/2020/03/higher-accuracy-on-vision-models-with-efficientnet-lite.html
"""
efficientnet_model = model_spec.get("efficientnet_lite1")
"""### **8). Training and Creating**
Here, we begin train the entire datasets and also create custom model based on pre-trained model
"""
model = image_classifier.create(train_data,
epochs=10,
validation_data=validation_data,
use_augmentation=True,
shuffle=True,
model_spec=efficientnet_model)
"""### **9). Display Model Summary**"""
model.summary()
"""### **10). Display Model Training-Validation Loss and Accuracy**"""
# Commented out IPython magic to ensure Python compatibility.
# %matplotlib inline
#Loss graph
plt.figure(figsize=(8, 6))
plt.plot(model.history.history["loss"])
Tensorflow Lite Model Maker ile Uygulama için Hazır Edilmesi (Mobil Kullanım için)
pip install tflite-model-maker
import tensorflow as tf
assert tf.__version__.startswith('2')
from tflite_model_maker import configs
from tflite_model_maker import ExportFormat
from tflite_model_maker import image_classifier
from tflite_model_maker import ImageClassifierDataLoader
from tflite_model_maker import model_spec
import matplotlib.pyplot as plt
# Eğitim, Değerlendirme ve Test Verilerinin Yüklenmesi
training_data = ImageClassifierDataLoader.from_folder('/content/drive/MyDrive/onayli_onaysiz_512x512/training/')
validation_data = ImageClassifierDataLoader.from_folder('/content/drive/MyDrive/onayli_onaysiz_512x512/validation/')
test_data = ImageClassifierDataLoader.from_folder('/content/drive/MyDrive/onayli_onaysiz_512x512/test/')
print(len(validation_data))
print(len(test_data))
# Modelin Eğitilmesi
model = image_classifier.create(training_data, validation_data=validation_data, epochs=10)
# Modelin Test Verileriye Onaylanması
loss, accuracy = model.evaluate(test_data)
# Tensorflow Lite Modelin Yüklenmesi ve bu Dosya Android Studio tarafından kullanılabilecektir
model.export(export_dir='.')
seed=12,
image_size=(imgHeight, imgWidth),
batch_size=batchSize)
# Save labels
classNames = trainDataset.class_names
print("Classes names:%s" % classNames)
np.savetxt(labelsPath, classNames, '%s', delimiter='\n')
print("Control class names:")
for s in classNames:
print(s, end=', ')
print()
print("Control labels save to path: %s" % labelsPath)
# Load data
data = ImageClassifierDataLoader.from_folder(path)
trainData, testData = data.split(0.62)
trainData = data
# Create model
model = image_classifier.create(trainDataset)
model.summary()
# Evaluate the model
loss, acc = model.evaluate(testData)
print("Model accuracy: %.4f" % acc)
print("Model loss: %.4f" % loss)
# Export model
model.export(modelPath)
'https://storage.googleapis.com/download.tensorflow.org/example_images/flower_photos.tgz',
untar=True)
data = ImageClassifierDataLoader.from_folder(image_path)
train_data, rest_data = data.split(0.8)
validation_data, test_data = rest_data.split(0.5)
plt.figure(figsize=(10, 10))
for i, (image, label) in enumerate(data.dataset.take(25)):
plt.subplot(5, 5, i + 1)
plt.xticks([])
plt.yticks([])
plt.grid(False)
plt.imshow(image.numpy(), cmap=plt.cm.gray)
plt.xlabel(data.index_to_label[label.numpy()])
plt.show()
model = image_classifier.create(train_data, validation_data=validation_data)
model.summary()
loss, accuracy = model.evaluate(test_data)
def get_label_color(val1, val2):
if val1 == val2:
return 'black'
else:
return 'red'
# Then plot 100 test images and their predicted labels.
# If a prediction result is different from the label provided label in "test"
# dataset, we will highlight it in red color.
plt.figure(figsize=(20, 20))
# for i, (image, label) in enumerate(data.gen_dataset().unbatch().take(25)):
# plt.subplot(5,5,i+1)
# plt.xticks([])
# plt.yticks([])
# plt.grid(False)
# plt.imshow(image.numpy(), cmap=plt.cm.gray)
# plt.xlabel(data.index_to_label[label.numpy()])
# plt.show()
# model = image_classifier.create(train_data, validation_data=validation_data)
# model.summary()
train_data, test_data = data.split(0.8)
#SINGLE MODEL
model = image_classifier.create(train_data, dropout_rate=0.1)
#Evaluating model
loss, accuracy = model.evaluate(test_data)
# #Export TFLite file
model.export(export_dir='.')
# #Export LABEL file
model.export(export_dir='.', export_format=ExportFormat.LABEL)
#Post Training optimization
config = configs.QuantizationConfig.create_full_integer_quantization(
representative_data=test_data, is_integer_only=True)
#Exporting optimized model
import tensorflow as tf
assert tf.__version__.startswith('2')
from tflite_model_maker import model_spec
from tflite_model_maker import image_classifier
from tflite_model_maker.config import ExportFormat
from tflite_model_maker.config import QuantizationConfig
from tflite_model_maker.image_classifier import DataLoader
import matplotlib.pyplot as plt
#%% Config
# Based on: https://codelabs.developers.google.com/tflite-computer-vision-train-model
image_path = "D:\\GIT\\machine-learning-project\\main\\_data_224\\"
EPOCHS = 1
data = DataLoader.from_folder(image_path)
train_data, test_data = data.split(0.9)
#%% Training
model = image_classifier.create(train_data, epochs=EPOCHS)
#%% evaluation
loss, accuracy = model.evaluate(test_data)
#%% export
model.export(export_dir='.', export_format=ExportFormat.SAVED_MODEL)
model.export(export_dir='.', export_format=ExportFormat.LABEL)
# %%