def inference(it, num_workers, args):
from tensorflowonspark import util
# consume worker number from RDD partition iterator
for i in it:
worker_num = i
print("worker_num: {}".format(i))
# setup env for single-node TF
util.single_node_env()
# load saved_model using default tag and signature
sess = tf.Session()
tf.saved_model.loader.load(sess, ['serve'], args.export)
# parse function for TFRecords
def parse_tfr(example_proto):
feature_def = {"label": tf.FixedLenFeature(10, tf.int64),
"image": tf.FixedLenFeature(IMAGE_PIXELS * IMAGE_PIXELS, tf.int64)}
features = tf.parse_single_example(example_proto, feature_def)
norm = tf.constant(255, dtype=tf.float32, shape=(784,))
image = tf.div(tf.to_float(features['image']), norm)
label = tf.to_float(features['label'])
return (image, label)
# define a new tf.data.Dataset (for inferencing)
ds = tf.data.Dataset.list_files("{}/part-*".format(args.images_labels))
ds = ds.shard(num_workers, worker_num)
ds = ds.interleave(tf.data.TFRecordDataset, cycle_length=1)
ds = ds.map(parse_tfr).batch(10)
iterator = ds.make_one_shot_iterator()
image_label = iterator.get_next(name='inf_image')
# create an output file per spark worker for the predictions
tf.gfile.MakeDirs(args.output)
output_file = tf.gfile.GFile("{}/part-{:05d}".format(args.output, worker_num), mode='w')
while True:
try:
# get images and labels from tf.data.Dataset
img, lbl = sess.run(['inf_image:0', 'inf_image:1'])
# inference by feeding these images and labels into the input tensors
# you can view the exported model signatures via:
# saved_model_cli show --dir <export_dir> --all
# note that we feed directly into the graph tensors (bypassing the exported signatures)
# these tensors will be shown in the "name" field of the signature definitions
outputs = sess.run(['dense_2/Softmax:0'], feed_dict={'Placeholder:0': img})
for p in outputs[0]:
output_file.write("{}\n".format(np.argmax(p)))
except tf.errors.OutOfRangeError:
break
output_file.close()
def inference(it, num_workers, args):
from tensorflowonspark import util
# consume worker number from RDD partition iterator
for i in it:
worker_num = i
print("worker_num: {}".format(i))
# setup env for single-node TF
util.single_node_env()
# load saved_model
saved_model = tf.saved_model.load(args.export_dir, tags='serve')
predict = saved_model.signatures['serving_default']
# parse function for TFRecords
def parse_tfr(example_proto):
feature_def = {
"label": tf.io.FixedLenFeature(1, tf.int64),
"image": tf.io.FixedLenFeature(784, tf.int64)
}
features = tf.io.parse_single_example(serialized=example_proto,
features=feature_def)
image = tf.cast(features['image'], dtype=tf.float32) / 255.0
image = tf.reshape(image, [28, 28, 1])
label = tf.cast(features['label'], dtype=tf.float32)
return (image, label)
# define a new tf.data.Dataset (for inferencing)
ds = tf.data.Dataset.list_files("{}/part-*".format(args.images_labels),
shuffle=False)
ds = ds.shard(num_workers, worker_num)
ds = ds.interleave(tf.data.TFRecordDataset)
ds = ds.map(parse_tfr)
ds = ds.batch(10)
# create an output file per spark worker for the predictions
tf.io.gfile.makedirs(args.output)
output_file = tf.io.gfile.GFile("{}/part-{:05d}".format(
args.output, worker_num),
mode='w')
for batch in ds:
predictions = predict(conv2d_input=batch[0])
labels = np.reshape(batch[1], -1).astype(np.int)
preds = np.argmax(predictions['logits'], axis=1)
for x in zip(labels, preds):
output_file.write("{} {}\n".format(x[0], x[1]))
output_file.close()
def inference(it, num_workers, args):
from tensorflowonspark import util
os.environ['GOOGLE_APPLICATION_CREDENTIALS'] = args.keyfile
# consume worker number from RDD partition iterator
for i in it:
worker_num = i
print("worker_num: {}".format(i))
# setup env for single-node TF
util.single_node_env()
# load saved_model using default tag and signature
model_dir = args.model_dir
tag_set = 'serve'
signature_def_key = 'serving_default'
sess = tf.Session()
tf.saved_model.loader.load(sess, [tag_set], model_dir)
# define a new tf.data.Dataset (for inferencing)
ds = tf.data.Dataset.list_files("{}*".format(args.input_dir))
ds = ds.interleave(tf.data.TextLineDataset, cycle_length=1)
ds = ds.shard(num_workers, worker_num)
ds = ds.batch(100)
iterator = ds.make_one_shot_iterator()
input = iterator.get_next()
# create an output file per spark worker for the predictions
tf.gfile.MakeDirs(args.output_dir)
file_path = "{}/prediction.results-{:05d}-of-{:05d}".format(
args.output_dir, worker_num, num_workers)
print("worker({0}) write to {1}".format(worker_num, file_path))
output_file = tf.gfile.GFile(file_path, mode='w')
inputs, outputs = get_input_and_output_names(model_dir, tag_set,
signature_def_key)
sorted_outputs = sorted(outputs.items())
output_keys = [key for key, value in sorted_outputs]
output_names = [value for key, value in sorted_outputs]
while True:
try:
dataset = sess.run(input)
result = sess.run(output_names,
feed_dict={inputs.values()[0]: dataset})
cols = len(result)
rows = len(result[0])
for row_num in range(rows):
row = bytearray("{")
for col_num in range(cols):
if col_num > 0: row.extend(",")
key = output_keys[col_num]
value = result[col_num][row_num]
row.extend(bytearray("'"))
row.extend(
bytearray(key.decode(default_codec), default_codec))
row.extend(bytearray("':"))
if isinstance(value, str):
row.extend(bytearray("'"))
row.extend(
bytearray(value.decode(default_codec),
default_codec))
row.extend(bytearray("'"))
else:
row.extend(bytearray(str(value)))
row.extend(bytearray("}\n"))
output_file.write(bytes(row))
except tf.errors.OutOfRangeError:
break
output_file.close()