#csv_name = "tomoto_and_unclass.csv"
csv = pd.read_csv(csv_name, header=None)
#test_csv_name = 'tomato_test_only_tomato.csv'
test_csv_name = 'tomato_test_only_tomato_original.csv'
test_csv = pd.read_csv(test_csv_name, header=None)
#path col=0
#label col=1
sample_size = csv.shape[0]
n_class = len(np.unique(csv[1]))
iteration_num = int(sample_size / a.batch_size * a.epoch)
seed = 1141919
#--------------ImageLoad-----------------#
with tf.name_scope('LoadImage'):
filename_queue = tf.train.string_input_producer([csv_name], shuffle=True)
reader = tf.TextLineReader()
_, val = reader.read(filename_queue)
record_defaults = [["a"], [0], ["a"]]
_, label, path = tf.decode_csv(val, record_defaults=record_defaults)
readfile = tf.read_file(path)
image = tf.image.decode_jpeg(readfile, channels=3)
image = tf.image.convert_image_dtype(image, dtype=tf.float32)
image = tf.cast(image, dtype=np.float32)
image = tf.image.resize_images(image, (256, 256))
def noisy(img):
img_f = tf.reshape(img, [-1])
a_m = tf.where(img_f > 1 / 255.0, tf.zeros_like(img_f),
tf.ones(img_f.shape))
a_m = tf.reshape(a_m, img.shape)
noise = tf.random_uniform(a_m.shape,
import tensorflow as tf
filename_queue = tf.train.string_input_producer(['data-01-test-score.csv'],
shuffle=False,
name='filename_queue')
reader = tf.TextLineReader()
key, value = reader.read(filename_queue)
# 값이 없을 경우의 기본값 설정
record_defaults = [[0.], [0.], [0.], [0.]]
xy = tf.decode_csv(value, record_defaults=record_defaults)
# collect batches of csv in
train_x_batch, train_y_batch = tf.train.batch([xy[0:-1], xy[-1:]],
batch_size=10)
# placeholders
X = tf.placeholder(tf.float32, shape=[None, 3])
Y = tf.placeholder(tf.float32, shape=[None, 1])
W = tf.Variable(tf.random_normal([3, 1]), name='weight')
b = tf.Variable(tf.random_normal([1]), name='bias')
# 가설
hypothesis = tf.matmul(X, W) + b
# Simplified cost/loss function
cost = tf.reduce_mean(tf.square(hypothesis - Y))
# Minimize
worker_device="/job:worker/task:%d" % FLAGS.task_index,
cluster=cluster)):
#More to come on is_chief...
is_chief = FLAGS.task_index == 0
# count the number of global steps
global_step = tf.get_variable('global_step', [],
initializer=tf.constant_initializer(0),
trainable=False)
#inout data
trainset_files = [
("/home/andy_shen/code/url_svmlight/Day%d" % i) + ".svm"
for i in range(121)
]
train_filename_queue = tf.train.string_input_producer(trainset_files)
train_reader = tf.TextLineReader()
train_data_line = train_reader.read(train_filename_queue)
with tf.name_scope('placeholder'):
y = tf.placeholder(tf.float32, [None, 1])
sp_indices = tf.placeholder(tf.int64)
shape = tf.placeholder(tf.int64)
ids_val = tf.placeholder(tf.int64)
weights_val = tf.placeholder(tf.float32)
with tf.name_scope('parameter'):
#x_data = tf.sparse_to_dense(sp_indices, shape, weights_val, 0., name=None)
x_data = tf.SparseTensor(sp_indices, weights_val, shape)
#weight = tf.Variable(tf.constant(0.0, shape = [num_features, 1]))
#b = tf.Variable(tf.constant(0.1, shape=[1]))
loss = SVMModel_with_linear(x_data, y, num_features)