satnogs-wut/wut-ml

#!/usr/bin/python3
# wut-ml
#
# Vet a SatNOGS image using machine learning (guessing).
# It will vet the image located at test/unvetted/waterfall.png.
#
# Note, there is an issue to fix where it will vet everything
# under the data/test directory, so fix that. For now, just delete
# everything else. :)
#
# Usage:
# wut-ml
# Example:
# wut-ml

import os
import numpy as np
import tensorflow.python.keras
from tensorflow.python.keras import Sequential
from tensorflow.python.keras.layers import Activation, Dropout, Flatten, Dense
from tensorflow.python.keras.preprocessing.image import ImageDataGenerator
from tensorflow.python.keras.layers import Convolution2D, MaxPooling2D, ZeroPadding2D
from tensorflow.python.keras import optimizers
from tensorflow.python.keras.preprocessing import image
from tensorflow.python.keras.models import load_model
from tensorflow.python.keras.preprocessing.image import load_img
from tensorflow.python.keras.preprocessing.image import img_to_array

# https://keras.io/preprocessing/image/
print("datagen")
datagen = ImageDataGenerator(
	featurewise_center=False,
	samplewise_center=False,
	featurewise_std_normalization=False,
	samplewise_std_normalization=False,
	zca_whitening=False,
	zca_epsilon=1e-06,
	rescale=1./255,
	shear_range=0.0,
	zoom_range=0.0,
	rotation_range=0,
	width_shift_range=0.0,
	height_shift_range=0.0,
	brightness_range=None,
	channel_shift_range=0.0,
	fill_mode='nearest',
	cval=0.0,
	horizontal_flip=False,
	vertical_flip=False,
	preprocessing_function=None,
	data_format='channels_last',
	validation_split=0.0,
	dtype='float32')

print("datagen.flow")
train_it = datagen.flow_from_directory('data/train/', class_mode='binary')
val_it = datagen.flow_from_directory('data/val/', class_mode='binary')
test_it = datagen.flow_from_directory('data/test/', class_mode='binary')
print("train_it.next()")
batchX, batchy = train_it.next()
print('Batch shape=%s, min=%.3f, max=%.3f' % (batchX.shape, batchX.min(), batchX.max()))
#img_width=823
#img_height=1606
img_width=256
img_height=256
print("Height", img_height, "Width", img_width)

# https://keras.io/models/sequential/
# https://keras.io/getting-started/sequential-model-guide/
print("Sequential")
model = Sequential()

print("add")
# https://keras.io/layers/convolutional/
model.add(Convolution2D(32, 3, 3, input_shape=(img_width, img_height,3)))
# https://keras.io/activations/
model.add(Activation('relu'))
# https://keras.io/layers/pooling/
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Convolution2D(32, 3, 3))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Convolution2D(64, 3, 3))
model.add(Activation('relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
# https://keras.io/layers/core/
model.add(Flatten())
# https://keras.io/layers/core/
model.add(Dense(64))
model.add(Activation('relu'))
# https://keras.io/layers/core/
model.add(Dropout(0.5))
model.add(Dense(1))
model.add(Activation('softmax'))

# https://keras.io/models/sequential/
print("compile")
model.compile(
	loss='categorical_crossentropy',
	loss_weights=None,
	sample_weight_mode=None,
	weighted_metrics=None,
	target_tensors=None,	
	optimizer='rmsprop',
	metrics=['accuracy'])

# https://keras.io/models/sequential/
print("fit")
model.fit(
	x=train_it,
	y=None,
	batch_size=None,
	epochs=1,
	verbose=2,
	callbacks=None,
	validation_split=0.0,
	validation_data=val_it,
	shuffle=True,
	class_weight=None,
	sample_weight=None,
	initial_epoch=0,
	steps_per_epoch=None,
	validation_steps=None,
	validation_freq=1,
	max_queue_size=10,
	workers=16,
	use_multiprocessing=True)

# https://keras.io/models/sequential/
# evaluate(x=None, y=None, batch_size=None, verbose=1, sample_weight=None, steps=None, callbacks=None, max_queue_size=10, workers=1, use_multiprocessing=False)

# https://keras.io/models/sequential/
print("predict")
prediction = model.predict(
	x=test_it,
	batch_size=None,
	verbose=2,
	steps=None,
	callbacks=None,
	max_queue_size=10,
	workers=16,
	use_multiprocessing=True)

print(prediction)

if prediction[0][0] == 1:
  rating = 'bad'
else:
  rating = 'good'
print('Observation: %s' % (rating))