Skip to content

GitLab

Commit d160e2fd authored by Mark Daoust's avatar Mark Daoust
Browse files

Delete deprecated tutorials directory.

parent bf6d6f6f
Hide whitespace changes
Inline Side-by-side
Showing with 0 additions and 1542 deletions
tutorials/rnn/ptb/BUILD deleted 100644 → 0
View file @ bf6d6f6f
# Description:
# Python support for TensorFlow.
package(default_visibility = ["//tensorflow:internal"])
licenses(["notice"]) # Apache 2.0
exports_files(["LICENSE"])
py_library(
name = "package",
srcs = [
"__init__.py",
],
srcs_version = "PY2AND3",
deps = [
":reader",
],
)
py_library(
name = "reader",
srcs = ["reader.py"],
srcs_version = "PY2AND3",
deps = ["//tensorflow:tensorflow_py"],
)
py_test(
name = "reader_test",
size = "small",
srcs = ["reader_test.py"],
srcs_version = "PY2AND3",
deps = [
":reader",
"//tensorflow:tensorflow_py",
],
)
py_library(
name = "util",
srcs = ["util.py"],
srcs_version = "PY2AND3",
deps = ["//tensorflow:tensorflow_py"],
)
py_binary(
name = "ptb_word_lm",
srcs = [
"ptb_word_lm.py",
],
srcs_version = "PY2AND3",
deps = [
":reader",
":util",
"//tensorflow:tensorflow_py",
],
)
filegroup(
name = "all_files",
srcs = glob(
["**/*"],
exclude = [
"**/METADATA",
"**/OWNERS",
],
),
visibility = ["//tensorflow:__subpackages__"],
)
tutorials/rnn/ptb/__init__.py deleted 100644 → 0
View file @ bf6d6f6f
# Copyright 2015 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Makes helper libraries available in the ptb package."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import reader
import util
tutorials/rnn/ptb/ptb_word_lm.py deleted 100644 → 0
View file @ bf6d6f6f
# Copyright 2015 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Example / benchmark for building a PTB LSTM model.
Trains the model described in:
(Zaremba, et. al.) Recurrent Neural Network Regularization
http://arxiv.org/abs/1409.2329
There are 3 supported model configurations:
===========================================
| config | epochs | train | valid | test
===========================================
| small | 13 | 37.99 | 121.39 | 115.91
| medium | 39 | 48.45 | 86.16 | 82.07
| large | 55 | 37.87 | 82.62 | 78.29
The exact results may vary depending on the random initialization.
The hyperparameters used in the model:
- init_scale - the initial scale of the weights
- learning_rate - the initial value of the learning rate
- max_grad_norm - the maximum permissible norm of the gradient
- num_layers - the number of LSTM layers
- num_steps - the number of unrolled steps of LSTM
- hidden_size - the number of LSTM units
- max_epoch - the number of epochs trained with the initial learning rate
- max_max_epoch - the total number of epochs for training
- keep_prob - the probability of keeping weights in the dropout layer
- lr_decay - the decay of the learning rate for each epoch after "max_epoch"
- batch_size - the batch size
- rnn_mode - the low level implementation of lstm cell: one of CUDNN,
BASIC, or BLOCK, representing cudnn_lstm, basic_lstm, and
lstm_block_cell classes.
The data required for this example is in the data/ dir of the
PTB dataset from Tomas Mikolov's webpage:
$ wget http://www.fit.vutbr.cz/~imikolov/rnnlm/simple-examples.tgz
$ tar xvf simple-examples.tgz
To run:
$ python ptb_word_lm.py --data_path=simple-examples/data/
"""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import time
import numpy as np
import tensorflow as tf
import reader
import util
from tensorflow.python.client import device_lib
from distutils.version import StrictVersion
flags = tf.flags
logging = tf.logging
flags.DEFINE_string(
"model", "small",
"A type of model. Possible options are: small, medium, large.")
flags.DEFINE_string("data_path", None,
"Where the training/test data is stored.")
flags.DEFINE_string("save_path", None,
"Model output directory.")
flags.DEFINE_bool("use_fp16", False,
"Train using 16-bit floats instead of 32bit floats")
flags.DEFINE_integer("num_gpus", 1,
"If larger than 1, Grappler AutoParallel optimizer "
"will create multiple training replicas with each GPU "
"running one replica.")
flags.DEFINE_string("rnn_mode", None,
"The low level implementation of lstm cell: one of CUDNN, "
"BASIC, and BLOCK, representing cudnn_lstm, basic_lstm, "
"and lstm_block_cell classes.")
FLAGS = flags.FLAGS
BASIC = "basic"
CUDNN = "cudnn"
BLOCK = "block"
def data_type():
return tf.float16 if FLAGS.use_fp16 else tf.float32
class PTBInput(object):
"""The input data."""
def __init__(self, config, data, name=None):
self.batch_size = batch_size = config.batch_size
self.num_steps = num_steps = config.num_steps
self.epoch_size = ((len(data) // batch_size) - 1) // num_steps
self.input_data, self.targets = reader.ptb_producer(
data, batch_size, num_steps, name=name)
class PTBModel(object):
"""The PTB model."""
def __init__(self, is_training, config, input_):
self._is_training = is_training
self._input = input_
self._rnn_params = None
self._cell = None
self.batch_size = input_.batch_size
self.num_steps = input_.num_steps
size = config.hidden_size
vocab_size = config.vocab_size
with tf.device("/cpu:0"):
embedding = tf.get_variable(
"embedding", [vocab_size, size], dtype=data_type())
inputs = tf.nn.embedding_lookup(embedding, input_.input_data)
if is_training and config.keep_prob < 1:
inputs = tf.nn.dropout(inputs, config.keep_prob)
output, state = self._build_rnn_graph(inputs, config, is_training)
softmax_w = tf.get_variable(
"softmax_w", [size, vocab_size], dtype=data_type())
softmax_b = tf.get_variable("softmax_b", [vocab_size], dtype=data_type())
logits = tf.nn.xw_plus_b(output, softmax_w, softmax_b)
# Reshape logits to be a 3-D tensor for sequence loss
logits = tf.reshape(logits, [self.batch_size, self.num_steps, vocab_size])
# Use the contrib sequence loss and average over the batches
loss = tf.contrib.seq2seq.sequence_loss(
logits,
input_.targets,
tf.ones([self.batch_size, self.num_steps], dtype=data_type()),
average_across_timesteps=False,
average_across_batch=True)
# Update the cost
self._cost = tf.reduce_sum(loss)
self._final_state = state
if not is_training:
return
self._lr = tf.Variable(0.0, trainable=False)
tvars = tf.trainable_variables()
grads, _ = tf.clip_by_global_norm(tf.gradients(self._cost, tvars),
config.max_grad_norm)
optimizer = tf.train.GradientDescentOptimizer(self._lr)
self._train_op = optimizer.apply_gradients(
zip(grads, tvars),
global_step=tf.train.get_or_create_global_step())
self._new_lr = tf.placeholder(
tf.float32, shape=[], name="new_learning_rate")
self._lr_update = tf.assign(self._lr, self._new_lr)
def _build_rnn_graph(self, inputs, config, is_training):
if config.rnn_mode == CUDNN:
return self._build_rnn_graph_cudnn(inputs, config, is_training)
else:
return self._build_rnn_graph_lstm(inputs, config, is_training)
def _build_rnn_graph_cudnn(self, inputs, config, is_training):
"""Build the inference graph using CUDNN cell."""
inputs = tf.transpose(inputs, [1, 0, 2])
self._cell = tf.contrib.cudnn_rnn.CudnnLSTM(
num_layers=config.num_layers,
num_units=config.hidden_size,
input_size=config.hidden_size,
dropout=1 - config.keep_prob if is_training else 0)
params_size_t = self._cell.params_size()
self._rnn_params = tf.get_variable(
"lstm_params",
initializer=tf.random_uniform(
[params_size_t], -config.init_scale, config.init_scale),
validate_shape=False)
c = tf.zeros([config.num_layers, self.batch_size, config.hidden_size],
tf.float32)
h = tf.zeros([config.num_layers, self.batch_size, config.hidden_size],
tf.float32)
self._initial_state = (tf.contrib.rnn.LSTMStateTuple(h=h, c=c),)
outputs, h, c = self._cell(inputs, h, c, self._rnn_params, is_training)
outputs = tf.transpose(outputs, [1, 0, 2])
outputs = tf.reshape(outputs, [-1, config.hidden_size])
return outputs, (tf.contrib.rnn.LSTMStateTuple(h=h, c=c),)
def _get_lstm_cell(self, config, is_training):
if config.rnn_mode == BASIC:
return tf.contrib.rnn.BasicLSTMCell(
config.hidden_size, forget_bias=0.0, state_is_tuple=True,
reuse=not is_training)
if config.rnn_mode == BLOCK:
return tf.contrib.rnn.LSTMBlockCell(
config.hidden_size, forget_bias=0.0)
raise ValueError("rnn_mode %s not supported" % config.rnn_mode)
def _build_rnn_graph_lstm(self, inputs, config, is_training):
"""Build the inference graph using canonical LSTM cells."""
# Slightly better results can be obtained with forget gate biases
# initialized to 1 but the hyperparameters of the model would need to be
# different than reported in the paper.
def make_cell():
cell = self._get_lstm_cell(config, is_training)
if is_training and config.keep_prob < 1:
cell = tf.contrib.rnn.DropoutWrapper(
cell, output_keep_prob=config.keep_prob)
return cell
cell = tf.contrib.rnn.MultiRNNCell(
[make_cell() for _ in range(config.num_layers)], state_is_tuple=True)
self._initial_state = cell.zero_state(config.batch_size, data_type())
state = self._initial_state
# Simplified version of tf.nn.static_rnn().
# This builds an unrolled LSTM for tutorial purposes only.
# In general, use tf.nn.static_rnn() or tf.nn.static_state_saving_rnn().
#
# The alternative version of the code below is:
#
# inputs = tf.unstack(inputs, num=self.num_steps, axis=1)
# outputs, state = tf.nn.static_rnn(cell, inputs,
# initial_state=self._initial_state)
outputs = []
with tf.variable_scope("RNN"):
for time_step in range(self.num_steps):
if time_step > 0: tf.get_variable_scope().reuse_variables()
(cell_output, state) = cell(inputs[:, time_step, :], state)
outputs.append(cell_output)
output = tf.reshape(tf.concat(outputs, 1), [-1, config.hidden_size])
return output, state
def assign_lr(self, session, lr_value):
session.run(self._lr_update, feed_dict={self._new_lr: lr_value})
def export_ops(self, name):
"""Exports ops to collections."""
self._name = name
ops = {util.with_prefix(self._name, "cost"): self._cost}
if self._is_training:
ops.update(lr=self._lr, new_lr=self._new_lr, lr_update=self._lr_update)
if self._rnn_params:
ops.update(rnn_params=self._rnn_params)
for name, op in ops.items():
tf.add_to_collection(name, op)
self._initial_state_name = util.with_prefix(self._name, "initial")
self._final_state_name = util.with_prefix(self._name, "final")
util.export_state_tuples(self._initial_state, self._initial_state_name)
util.export_state_tuples(self._final_state, self._final_state_name)
def import_ops(self):
"""Imports ops from collections."""
if self._is_training:
self._train_op = tf.get_collection_ref("train_op")[0]
self._lr = tf.get_collection_ref("lr")[0]
self._new_lr = tf.get_collection_ref("new_lr")[0]
self._lr_update = tf.get_collection_ref("lr_update")[0]
rnn_params = tf.get_collection_ref("rnn_params")
if self._cell and rnn_params:
params_saveable = tf.contrib.cudnn_rnn.RNNParamsSaveable(
self._cell,
self._cell.params_to_canonical,
self._cell.canonical_to_params,
rnn_params,
base_variable_scope="Model/RNN")
tf.add_to_collection(tf.GraphKeys.SAVEABLE_OBJECTS, params_saveable)
self._cost = tf.get_collection_ref(util.with_prefix(self._name, "cost"))[0]
num_replicas = FLAGS.num_gpus if self._name == "Train" else 1
self._initial_state = util.import_state_tuples(
self._initial_state, self._initial_state_name, num_replicas)
self._final_state = util.import_state_tuples(
self._final_state, self._final_state_name, num_replicas)
@property
def input(self):
return self._input
@property
def initial_state(self):
return self._initial_state
@property
def cost(self):
return self._cost
@property
def final_state(self):
return self._final_state
@property
def lr(self):
return self._lr
@property
def train_op(self):
return self._train_op
@property
def initial_state_name(self):
return self._initial_state_name
@property
def final_state_name(self):
return self._final_state_name
class SmallConfig(object):
"""Small config."""
init_scale = 0.1
learning_rate = 1.0
max_grad_norm = 5
num_layers = 2
num_steps = 20
hidden_size = 200
max_epoch = 4
max_max_epoch = 13
keep_prob = 1.0
lr_decay = 0.5
batch_size = 20
vocab_size = 10000
rnn_mode = BLOCK
class MediumConfig(object):
"""Medium config."""
init_scale = 0.05
learning_rate = 1.0
max_grad_norm = 5
num_layers = 2
num_steps = 35
hidden_size = 650
max_epoch = 6
max_max_epoch = 39
keep_prob = 0.5
lr_decay = 0.8
batch_size = 20
vocab_size = 10000
rnn_mode = BLOCK
class LargeConfig(object):
"""Large config."""
init_scale = 0.04
learning_rate = 1.0
max_grad_norm = 10
num_layers = 2
num_steps = 35
hidden_size = 1500
max_epoch = 14
max_max_epoch = 55
keep_prob = 0.35
lr_decay = 1 / 1.15
batch_size = 20
vocab_size = 10000
rnn_mode = BLOCK
class TestConfig(object):
"""Tiny config, for testing."""
init_scale = 0.1
learning_rate = 1.0
max_grad_norm = 1
num_layers = 1
num_steps = 2
hidden_size = 2
max_epoch = 1
max_max_epoch = 1
keep_prob = 1.0
lr_decay = 0.5
batch_size = 20
vocab_size = 10000
rnn_mode = BLOCK
def run_epoch(session, model, eval_op=None, verbose=False):
"""Runs the model on the given data."""
start_time = time.time()
costs = 0.0
iters = 0
state = session.run(model.initial_state)
fetches = {
"cost": model.cost,
"final_state": model.final_state,
}
if eval_op is not None:
fetches["eval_op"] = eval_op
for step in range(model.input.epoch_size):
feed_dict = {}
for i, (c, h) in enumerate(model.initial_state):
feed_dict[c] = state[i].c
feed_dict[h] = state[i].h
vals = session.run(fetches, feed_dict)
cost = vals["cost"]
state = vals["final_state"]
costs += cost
iters += model.input.num_steps
if verbose and step % (model.input.epoch_size // 10) == 10:
print("%.3f perplexity: %.3f speed: %.0f wps" %
(step * 1.0 / model.input.epoch_size, np.exp(costs / iters),
iters * model.input.batch_size * max(1, FLAGS.num_gpus) /
(time.time() - start_time)))
return np.exp(costs / iters)
def get_config():
"""Get model config."""
config = None
if FLAGS.model == "small":
config = SmallConfig()
elif FLAGS.model == "medium":
config = MediumConfig()
elif FLAGS.model == "large":
config = LargeConfig()
elif FLAGS.model == "test":
config = TestConfig()
else:
raise ValueError("Invalid model: %s", FLAGS.model)
if FLAGS.rnn_mode:
config.rnn_mode = FLAGS.rnn_mode
if FLAGS.num_gpus != 1 or StrictVersion(tf.__version__) < StrictVersion("1.3.0") :
config.rnn_mode = BASIC
return config
def main(_):
if not FLAGS.data_path:
raise ValueError("Must set --data_path to PTB data directory")
gpus = [
x.name for x in device_lib.list_local_devices() if x.device_type == "GPU"
]
if FLAGS.num_gpus > len(gpus):
raise ValueError(
"Your machine has only %d gpus "
"which is less than the requested --num_gpus=%d."
% (len(gpus), FLAGS.num_gpus))
raw_data = reader.ptb_raw_data(FLAGS.data_path)
train_data, valid_data, test_data, _ = raw_data
config = get_config()
eval_config = get_config()
eval_config.batch_size = 1
eval_config.num_steps = 1
with tf.Graph().as_default():
initializer = tf.random_uniform_initializer(-config.init_scale,
config.init_scale)
with tf.name_scope("Train"):
train_input = PTBInput(config=config, data=train_data, name="TrainInput")
with tf.variable_scope("Model", reuse=None, initializer=initializer):
m = PTBModel(is_training=True, config=config, input_=train_input)
tf.summary.scalar("Training Loss", m.cost)
tf.summary.scalar("Learning Rate", m.lr)
with tf.name_scope("Valid"):
valid_input = PTBInput(config=config, data=valid_data, name="ValidInput")
with tf.variable_scope("Model", reuse=True, initializer=initializer):
mvalid = PTBModel(is_training=False, config=config, input_=valid_input)
tf.summary.scalar("Validation Loss", mvalid.cost)
with tf.name_scope("Test"):
test_input = PTBInput(
config=eval_config, data=test_data, name="TestInput")
with tf.variable_scope("Model", reuse=True, initializer=initializer):
mtest = PTBModel(is_training=False, config=eval_config,
input_=test_input)
models = {"Train": m, "Valid": mvalid, "Test": mtest}
for name, model in models.items():
model.export_ops(name)
metagraph = tf.train.export_meta_graph()
if StrictVersion(tf.__version__) < StrictVersion("1.1.0") and FLAGS.num_gpus > 1:
raise ValueError("num_gpus > 1 is not supported for TensorFlow versions "
"below 1.1.0")
soft_placement = False
if FLAGS.num_gpus > 1:
soft_placement = True
util.auto_parallel(metagraph, m)
with tf.Graph().as_default():
tf.train.import_meta_graph(metagraph)
for model in models.values():
model.import_ops()
sv = tf.train.Supervisor(logdir=FLAGS.save_path)
config_proto = tf.ConfigProto(allow_soft_placement=soft_placement)
with sv.managed_session(config=config_proto) as session:
for i in range(config.max_max_epoch):
lr_decay = config.lr_decay ** max(i + 1 - config.max_epoch, 0.0)
m.assign_lr(session, config.learning_rate * lr_decay)
print("Epoch: %d Learning rate: %.3f" % (i + 1, session.run(m.lr)))
train_perplexity = run_epoch(session, m, eval_op=m.train_op,
verbose=True)
print("Epoch: %d Train Perplexity: %.3f" % (i + 1, train_perplexity))
valid_perplexity = run_epoch(session, mvalid)
print("Epoch: %d Valid Perplexity: %.3f" % (i + 1, valid_perplexity))
test_perplexity = run_epoch(session, mtest)
print("Test Perplexity: %.3f" % test_perplexity)
if FLAGS.save_path:
print("Saving model to %s." % FLAGS.save_path)
sv.saver.save(session, FLAGS.save_path, global_step=sv.global_step)
if __name__ == "__main__":
tf.app.run()
tutorials/rnn/ptb/reader.py deleted 100644 → 0
View file @ bf6d6f6f
# Copyright 2015 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Utilities for parsing PTB text files."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import collections
import os
import sys
import tensorflow as tf
Py3 = sys.version_info[0] == 3
def _read_words(filename):
with tf.gfile.GFile(filename, "r") as f:
if Py3:
return f.read().replace("\n", "<eos>").split()
else:
return f.read().decode("utf-8").replace("\n", "<eos>").split()
def _build_vocab(filename):
data = _read_words(filename)
counter = collections.Counter(data)
count_pairs = sorted(counter.items(), key=lambda x: (-x[1], x[0]))
words, _ = list(zip(*count_pairs))
word_to_id = dict(zip(words, range(len(words))))
return word_to_id
def _file_to_word_ids(filename, word_to_id):
data = _read_words(filename)
return [word_to_id[word] for word in data if word in word_to_id]
def ptb_raw_data(data_path=None):
"""Load PTB raw data from data directory "data_path".
Reads PTB text files, converts strings to integer ids,
and performs mini-batching of the inputs.
The PTB dataset comes from Tomas Mikolov's webpage:
http://www.fit.vutbr.cz/~imikolov/rnnlm/simple-examples.tgz
Args:
data_path: string path to the directory where simple-examples.tgz has
been extracted.
Returns:
tuple (train_data, valid_data, test_data, vocabulary)
where each of the data objects can be passed to PTBIterator.
"""
train_path = os.path.join(data_path, "ptb.train.txt")
valid_path = os.path.join(data_path, "ptb.valid.txt")
test_path = os.path.join(data_path, "ptb.test.txt")
word_to_id = _build_vocab(train_path)
train_data = _file_to_word_ids(train_path, word_to_id)
valid_data = _file_to_word_ids(valid_path, word_to_id)
test_data = _file_to_word_ids(test_path, word_to_id)
vocabulary = len(word_to_id)
return train_data, valid_data, test_data, vocabulary
def ptb_producer(raw_data, batch_size, num_steps, name=None):
"""Iterate on the raw PTB data.
This chunks up raw_data into batches of examples and returns Tensors that
are drawn from these batches.
Args:
raw_data: one of the raw data outputs from ptb_raw_data.
batch_size: int, the batch size.
num_steps: int, the number of unrolls.
name: the name of this operation (optional).
Returns:
A pair of Tensors, each shaped [batch_size, num_steps]. The second element
of the tuple is the same data time-shifted to the right by one.
Raises:
tf.errors.InvalidArgumentError: if batch_size or num_steps are too high.
"""
with tf.name_scope(name, "PTBProducer", [raw_data, batch_size, num_steps]):
raw_data = tf.convert_to_tensor(raw_data, name="raw_data", dtype=tf.int32)
data_len = tf.size(raw_data)
batch_len = data_len // batch_size
data = tf.reshape(raw_data[0 : batch_size * batch_len],
[batch_size, batch_len])
epoch_size = (batch_len - 1) // num_steps
assertion = tf.assert_positive(
epoch_size,
message="epoch_size == 0, decrease batch_size or num_steps")
with tf.control_dependencies([assertion]):
epoch_size = tf.identity(epoch_size, name="epoch_size")
i = tf.train.range_input_producer(epoch_size, shuffle=False).dequeue()
x = tf.strided_slice(data, [0, i * num_steps],
[batch_size, (i + 1) * num_steps])
x.set_shape([batch_size, num_steps])
y = tf.strided_slice(data, [0, i * num_steps + 1],
[batch_size, (i + 1) * num_steps + 1])
y.set_shape([batch_size, num_steps])
return x, y
tutorials/rnn/ptb/reader_test.py deleted 100644 → 0
View file @ bf6d6f6f
# Copyright 2015 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Tests for models.tutorials.rnn.ptb.reader."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import os.path
import tensorflow as tf
import reader
class PtbReaderTest(tf.test.TestCase):
def setUp(self):
self._string_data = "\n".join(
[" hello there i am",
" rain as day",
" want some cheesy puffs ?"])
def testPtbRawData(self):
tmpdir = tf.test.get_temp_dir()
for suffix in "train", "valid", "test":
filename = os.path.join(tmpdir, "ptb.%s.txt" % suffix)
with tf.gfile.GFile(filename, "w") as fh:
fh.write(self._string_data)
# Smoke test
output = reader.ptb_raw_data(tmpdir)
self.assertEqual(len(output), 4)
def testPtbProducer(self):
raw_data = [4, 3, 2, 1, 0, 5, 6, 1, 1, 1, 1, 0, 3, 4, 1]
batch_size = 3
num_steps = 2
x, y = reader.ptb_producer(raw_data, batch_size, num_steps)
with self.test_session() as session:
coord = tf.train.Coordinator()
tf.train.start_queue_runners(session, coord=coord)
try:
xval, yval = session.run([x, y])
self.assertAllEqual(xval, [[4, 3], [5, 6], [1, 0]])
self.assertAllEqual(yval, [[3, 2], [6, 1], [0, 3]])
xval, yval = session.run([x, y])
self.assertAllEqual(xval, [[2, 1], [1, 1], [3, 4]])
self.assertAllEqual(yval, [[1, 0], [1, 1], [4, 1]])
finally:
coord.request_stop()
coord.join()
if __name__ == "__main__":
tf.test.main()
tutorials/rnn/ptb/util.py deleted 100644 → 0
View file @ bf6d6f6f
# Copyright 2017 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
"""Utilities for Grappler autoparallel optimizer."""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import tensorflow as tf
from tensorflow.core.framework import variable_pb2
from tensorflow.core.protobuf import rewriter_config_pb2
FLAGS = tf.flags.FLAGS
def export_state_tuples(state_tuples, name):
for state_tuple in state_tuples:
tf.add_to_collection(name, state_tuple.c)
tf.add_to_collection(name, state_tuple.h)
def import_state_tuples(state_tuples, name, num_replicas):
restored = []
for i in range(len(state_tuples) * num_replicas):
c = tf.get_collection_ref(name)[2 * i + 0]
h = tf.get_collection_ref(name)[2 * i + 1]
restored.append(tf.contrib.rnn.LSTMStateTuple(c, h))
return tuple(restored)
def with_prefix(prefix, name):
"""Adds prefix to name."""
return "/".join((prefix, name))
def with_autoparallel_prefix(replica_id, name):
return with_prefix("AutoParallel-Replica-%d" % replica_id, name)
class UpdateCollection(object):
"""Update collection info in MetaGraphDef for AutoParallel optimizer."""
def __init__(self, metagraph, model):
self._metagraph = metagraph
self.replicate_states(model.initial_state_name)
self.replicate_states(model.final_state_name)
self.update_snapshot_name("variables")
self.update_snapshot_name("trainable_variables")
def update_snapshot_name(self, var_coll_name):
var_list = self._metagraph.collection_def[var_coll_name]
for i, value in enumerate(var_list.bytes_list.value):
var_def = variable_pb2.VariableDef()
var_def.ParseFromString(value)
# Somehow node Model/global_step/read doesn't have any fanout and seems to
# be only used for snapshot; this is different from all other variables.
if var_def.snapshot_name != "Model/global_step/read:0":
var_def.snapshot_name = with_autoparallel_prefix(
0, var_def.snapshot_name)
value = var_def.SerializeToString()
var_list.bytes_list.value[i] = value
def replicate_states(self, state_coll_name):
state_list = self._metagraph.collection_def[state_coll_name]
num_states = len(state_list.node_list.value)
for replica_id in range(1, FLAGS.num_gpus):
for i in range(num_states):
state_list.node_list.value.append(state_list.node_list.value[i])
for replica_id in range(FLAGS.num_gpus):
for i in range(num_states):
index = replica_id * num_states + i
state_list.node_list.value[index] = with_autoparallel_prefix(
replica_id, state_list.node_list.value[index])
def auto_parallel(metagraph, model):
from tensorflow.python.grappler import tf_optimizer
rewriter_config = rewriter_config_pb2.RewriterConfig()
rewriter_config.optimizers.append("autoparallel")
rewriter_config.auto_parallel.enable = True
rewriter_config.auto_parallel.num_replicas = FLAGS.num_gpus
optimized_graph = tf_optimizer.OptimizeGraph(rewriter_config, metagraph)
metagraph.graph_def.CopyFrom(optimized_graph)
UpdateCollection(metagraph, model)
tutorials/rnn/quickdraw/BUILD deleted 100644 → 0
View file @ bf6d6f6f
# Description:
# Example classification model on Quick, Draw! dataset.
package(default_visibility = ["//visibility:public"])
licenses(["notice"]) # Apache 2.0
exports_files(["LICENSE"])
py_binary(
name = "train_model",
srcs = [
"train_model.py",
],
srcs_version = "PY2AND3",
deps = [
"//third_party/py/tensorflow",
],
)
py_binary(
name = "create_dataset",
srcs = [
"create_dataset.py",
],
deps = [
"//third_party/py/numpy",
"//third_party/py/tensorflow",
],
)
filegroup(
name = "all_files",
srcs = glob(
["**/*"],
exclude = [
"**/METADATA",
"**/OWNERS",
],
),
visibility = ["//third_party/tensorflow:__subpackages__"],
)
tutorials/rnn/quickdraw/create_dataset.py deleted 100644 → 0
View file @ bf6d6f6f
# Copyright 2017 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
r"""Creates training and eval data from Quickdraw NDJSON files.
This tool reads the NDJSON files from https://quickdraw.withgoogle.com/data
and converts them into tensorflow.Example stored in TFRecord files.
The tensorflow example will contain 3 features:
shape - contains the shape of the sequence [length, dim] where dim=3.
class_index - the class index of the class for the example.
ink - a length * dim vector of the ink.
It creates disjoint training and evaluation sets.
python create_dataset.py \
--ndjson_path ${HOME}/ndjson \
--output_path ${HOME}/tfrecord
"""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import argparse
import json
import os
import random
import sys
import numpy as np
import tensorflow as tf
def parse_line(ndjson_line):
"""Parse an ndjson line and return ink (as np array) and classname."""
sample = json.loads(ndjson_line)
class_name = sample["word"]
if not class_name:
print ("Empty classname")
return None, None
inkarray = sample["drawing"]
stroke_lengths = [len(stroke[0]) for stroke in inkarray]
total_points = sum(stroke_lengths)
np_ink = np.zeros((total_points, 3), dtype=np.float32)
current_t = 0
if not inkarray:
print("Empty inkarray")
return None, None
for stroke in inkarray:
if len(stroke[0]) != len(stroke[1]):
print("Inconsistent number of x and y coordinates.")
return None, None
for i in [0, 1]:
np_ink[current_t:(current_t + len(stroke[0])), i] = stroke[i]
current_t += len(stroke[0])
np_ink[current_t - 1, 2] = 1 # stroke_end
# Preprocessing.
# 1. Size normalization.
lower = np.min(np_ink[:, 0:2], axis=0)
upper = np.max(np_ink[:, 0:2], axis=0)
scale = upper - lower
scale[scale == 0] = 1
np_ink[:, 0:2] = (np_ink[:, 0:2] - lower) / scale
# 2. Compute deltas.
np_ink[1:, 0:2] -= np_ink[0:-1, 0:2]
np_ink = np_ink[1:, :]
return np_ink, class_name
def convert_data(trainingdata_dir,
observations_per_class,
output_file,
classnames,
output_shards=10,
offset=0):
"""Convert training data from ndjson files into tf.Example in tf.Record.
Args:
trainingdata_dir: path to the directory containin the training data.
The training data is stored in that directory as ndjson files.
observations_per_class: the number of items to load per class.
output_file: path where to write the output.
classnames: array with classnames - is auto created if not passed in.
output_shards: the number of shards to write the output in.
offset: the number of items to skip at the beginning of each file.
Returns:
classnames: the class names as strings. classnames[classes[i]] is the
textual representation of the class of the i-th data point.
"""
def _pick_output_shard():
return random.randint(0, output_shards - 1)
file_handles = []
# Open all input files.
for filename in sorted(tf.gfile.ListDirectory(trainingdata_dir)):
if not filename.endswith(".ndjson"):
print("Skipping", filename)
continue
file_handles.append(
tf.gfile.GFile(os.path.join(trainingdata_dir, filename), "r"))
if offset: # Fast forward all files to skip the offset.
count = 0
for _ in file_handles[-1]:
count += 1
if count == offset:
break
writers = []
for i in range(FLAGS.output_shards):
writers.append(
tf.python_io.TFRecordWriter("%s-%05i-of-%05i" % (output_file, i,
output_shards)))
reading_order = list(range(len(file_handles))) * observations_per_class
random.shuffle(reading_order)
for c in reading_order:
line = file_handles[c].readline()
ink = None
while ink is None:
ink, class_name = parse_line(line)
if ink is None:
print ("Couldn't parse ink from '" + line + "'.")
if class_name not in classnames:
classnames.append(class_name)
features = {}
features["class_index"] = tf.train.Feature(int64_list=tf.train.Int64List(
value=[classnames.index(class_name)]))
features["ink"] = tf.train.Feature(float_list=tf.train.FloatList(
value=ink.flatten()))
features["shape"] = tf.train.Feature(int64_list=tf.train.Int64List(
value=ink.shape))
f = tf.train.Features(feature=features)
example = tf.train.Example(features=f)
writers[_pick_output_shard()].write(example.SerializeToString())
# Close all files
for w in writers:
w.close()
for f in file_handles:
f.close()
# Write the class list.
with tf.gfile.GFile(output_file + ".classes", "w") as f:
for class_name in classnames:
f.write(class_name + "\n")
return classnames
def main(argv):
del argv
classnames = convert_data(
FLAGS.ndjson_path,
FLAGS.train_observations_per_class,
os.path.join(FLAGS.output_path, "training.tfrecord"),
classnames=[],
output_shards=FLAGS.output_shards,
offset=0)
convert_data(
FLAGS.ndjson_path,
FLAGS.eval_observations_per_class,
os.path.join(FLAGS.output_path, "eval.tfrecord"),
classnames=classnames,
output_shards=FLAGS.output_shards,
offset=FLAGS.train_observations_per_class)
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.register("type", "bool", lambda v: v.lower() == "true")
parser.add_argument(
"--ndjson_path",
type=str,
default="",
help="Directory where the ndjson files are stored.")
parser.add_argument(
"--output_path",
type=str,
default="",
help="Directory where to store the output TFRecord files.")
parser.add_argument(
"--train_observations_per_class",
type=int,
default=10000,
help="How many items per class to load for training.")
parser.add_argument(
"--eval_observations_per_class",
type=int,
default=1000,
help="How many items per class to load for evaluation.")
parser.add_argument(
"--output_shards",
type=int,
default=10,
help="Number of shards for the output.")
FLAGS, unparsed = parser.parse_known_args()
tf.app.run(main=main, argv=[sys.argv[0]] + unparsed)
tutorials/rnn/quickdraw/train_model.py deleted 100644 → 0
View file @ bf6d6f6f
# Copyright 2017 The TensorFlow Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================
r"""Binary for training a RNN-based classifier for the Quick, Draw! data.
python train_model.py \
--training_data train_data \
--eval_data eval_data \
--model_dir /tmp/quickdraw_model/ \
--cell_type cudnn_lstm
When running on GPUs using --cell_type cudnn_lstm is much faster.
The expected performance is ~75% in 1.5M steps with the default configuration.
"""
from __future__ import absolute_import
from __future__ import division
from __future__ import print_function
import argparse
import ast
import functools
import sys
import tensorflow as tf
def get_num_classes():
classes = []
with tf.gfile.GFile(FLAGS.classes_file, "r") as f:
classes = [x for x in f]
num_classes = len(classes)
return num_classes
def get_input_fn(mode, tfrecord_pattern, batch_size):
"""Creates an input_fn that stores all the data in memory.
Args:
mode: one of tf.contrib.learn.ModeKeys.{TRAIN, INFER, EVAL}
tfrecord_pattern: path to a TF record file created using create_dataset.py.
batch_size: the batch size to output.
Returns:
A valid input_fn for the model estimator.
"""
def _parse_tfexample_fn(example_proto, mode):
"""Parse a single record which is expected to be a tensorflow.Example."""
feature_to_type = {
"ink": tf.VarLenFeature(dtype=tf.float32),
"shape": tf.FixedLenFeature([2], dtype=tf.int64)
}
if mode != tf.estimator.ModeKeys.PREDICT:
# The labels won't be available at inference time, so don't add them
# to the list of feature_columns to be read.
feature_to_type["class_index"] = tf.FixedLenFeature([1], dtype=tf.int64)
parsed_features = tf.parse_single_example(example_proto, feature_to_type)
labels = None
if mode != tf.estimator.ModeKeys.PREDICT:
labels = parsed_features["class_index"]
parsed_features["ink"] = tf.sparse_tensor_to_dense(parsed_features["ink"])
return parsed_features, labels
def _input_fn():
"""Estimator `input_fn`.
Returns:
A tuple of:
- Dictionary of string feature name to `Tensor`.
- `Tensor` of target labels.
"""
dataset = tf.data.TFRecordDataset.list_files(tfrecord_pattern)
if mode == tf.estimator.ModeKeys.TRAIN:
dataset = dataset.shuffle(buffer_size=10)
dataset = dataset.repeat()
# Preprocesses 10 files concurrently and interleaves records from each file.
dataset = dataset.interleave(
tf.data.TFRecordDataset,
cycle_length=10,
block_length=1)
dataset = dataset.map(
functools.partial(_parse_tfexample_fn, mode=mode),
num_parallel_calls=10)
dataset = dataset.prefetch(10000)
if mode == tf.estimator.ModeKeys.TRAIN:
dataset = dataset.shuffle(buffer_size=1000000)
# Our inputs are variable length, so pad them.
dataset = dataset.padded_batch(
batch_size, padded_shapes=dataset.output_shapes)
features, labels = dataset.make_one_shot_iterator().get_next()
return features, labels
return _input_fn
def model_fn(features, labels, mode, params):
"""Model function for RNN classifier.
This function sets up a neural network which applies convolutional layers (as
configured with params.num_conv and params.conv_len) to the input.
The output of the convolutional layers is given to LSTM layers (as configured
with params.num_layers and params.num_nodes).
The final state of the all LSTM layers are concatenated and fed to a fully
connected layer to obtain the final classification scores.
Args:
features: dictionary with keys: inks, lengths.
labels: one hot encoded classes
mode: one of tf.estimator.ModeKeys.{TRAIN, INFER, EVAL}
params: a parameter dictionary with the following keys: num_layers,
num_nodes, batch_size, num_conv, conv_len, num_classes, learning_rate.
Returns:
ModelFnOps for Estimator API.
"""
def _get_input_tensors(features, labels):
"""Converts the input dict into inks, lengths, and labels tensors."""
# features[ink] is a sparse tensor that is [8, batch_maxlen, 3]
# inks will be a dense tensor of [8, maxlen, 3]
# shapes is [batchsize, 2]
shapes = features["shape"]
# lengths will be [batch_size]
lengths = tf.squeeze(
tf.slice(shapes, begin=[0, 0], size=[params.batch_size, 1]))
inks = tf.reshape(features["ink"], [params.batch_size, -1, 3])
if labels is not None:
labels = tf.squeeze(labels)
return inks, lengths, labels
def _add_conv_layers(inks, lengths):
"""Adds convolution layers."""
convolved = inks
for i in range(len(params.num_conv)):
convolved_input = convolved
if params.batch_norm:
convolved_input = tf.layers.batch_normalization(
convolved_input,
training=(mode == tf.estimator.ModeKeys.TRAIN))
# Add dropout layer if enabled and not first convolution layer.
if i > 0 and params.dropout:
convolved_input = tf.layers.dropout(
convolved_input,
rate=params.dropout,
training=(mode == tf.estimator.ModeKeys.TRAIN))
convolved = tf.layers.conv1d(
convolved_input,
filters=params.num_conv[i],
kernel_size=params.conv_len[i],
activation=None,
strides=1,
padding="same",
name="conv1d_%d" % i)
return convolved, lengths
def _add_regular_rnn_layers(convolved, lengths):
"""Adds RNN layers."""
if params.cell_type == "lstm":
cell = tf.nn.rnn_cell.BasicLSTMCell
elif params.cell_type == "block_lstm":
cell = tf.contrib.rnn.LSTMBlockCell
cells_fw = [cell(params.num_nodes) for _ in range(params.num_layers)]
cells_bw = [cell(params.num_nodes) for _ in range(params.num_layers)]
if params.dropout > 0.0:
cells_fw = [tf.contrib.rnn.DropoutWrapper(cell) for cell in cells_fw]
cells_bw = [tf.contrib.rnn.DropoutWrapper(cell) for cell in cells_bw]
outputs, _, _ = tf.contrib.rnn.stack_bidirectional_dynamic_rnn(
cells_fw=cells_fw,
cells_bw=cells_bw,
inputs=convolved,
sequence_length=lengths,
dtype=tf.float32,
scope="rnn_classification")
return outputs
def _add_cudnn_rnn_layers(convolved):
"""Adds CUDNN LSTM layers."""
# Convolutions output [B, L, Ch], while CudnnLSTM is time-major.
convolved = tf.transpose(convolved, [1, 0, 2])
lstm = tf.contrib.cudnn_rnn.CudnnLSTM(
num_layers=params.num_layers,
num_units=params.num_nodes,
dropout=params.dropout if mode == tf.estimator.ModeKeys.TRAIN else 0.0,
direction="bidirectional")
outputs, _ = lstm(convolved)
# Convert back from time-major outputs to batch-major outputs.
outputs = tf.transpose(outputs, [1, 0, 2])
return outputs
def _add_rnn_layers(convolved, lengths):
"""Adds recurrent neural network layers depending on the cell type."""
if params.cell_type != "cudnn_lstm":
outputs = _add_regular_rnn_layers(convolved, lengths)
else:
outputs = _add_cudnn_rnn_layers(convolved)
# outputs is [batch_size, L, N] where L is the maximal sequence length and N
# the number of nodes in the last layer.
mask = tf.tile(
tf.expand_dims(tf.sequence_mask(lengths, tf.shape(outputs)[1]), 2),
[1, 1, tf.shape(outputs)[2]])
zero_outside = tf.where(mask, outputs, tf.zeros_like(outputs))
outputs = tf.reduce_sum(zero_outside, axis=1)
return outputs
def _add_fc_layers(final_state):
"""Adds a fully connected layer."""
return tf.layers.dense(final_state, params.num_classes)
# Build the model.
inks, lengths, labels = _get_input_tensors(features, labels)
convolved, lengths = _add_conv_layers(inks, lengths)
final_state = _add_rnn_layers(convolved, lengths)
logits = _add_fc_layers(final_state)
# Add the loss.
cross_entropy = tf.reduce_mean(
tf.nn.sparse_softmax_cross_entropy_with_logits(
labels=labels, logits=logits))
# Add the optimizer.
train_op = tf.contrib.layers.optimize_loss(
loss=cross_entropy,
global_step=tf.train.get_global_step(),
learning_rate=params.learning_rate,
optimizer="Adam",
# some gradient clipping stabilizes training in the beginning.
clip_gradients=params.gradient_clipping_norm,
summaries=["learning_rate", "loss", "gradients", "gradient_norm"])
# Compute current predictions.
predictions = tf.argmax(logits, axis=1)
return tf.estimator.EstimatorSpec(
mode=mode,
predictions={"logits": logits, "predictions": predictions},
loss=cross_entropy,
train_op=train_op,
eval_metric_ops={"accuracy": tf.metrics.accuracy(labels, predictions)})
def create_estimator_and_specs(run_config):
"""Creates an Experiment configuration based on the estimator and input fn."""
model_params = tf.contrib.training.HParams(
num_layers=FLAGS.num_layers,
num_nodes=FLAGS.num_nodes,
batch_size=FLAGS.batch_size,
num_conv=ast.literal_eval(FLAGS.num_conv),
conv_len=ast.literal_eval(FLAGS.conv_len),
num_classes=get_num_classes(),
learning_rate=FLAGS.learning_rate,
gradient_clipping_norm=FLAGS.gradient_clipping_norm,
cell_type=FLAGS.cell_type,
batch_norm=FLAGS.batch_norm,
dropout=FLAGS.dropout)
estimator = tf.estimator.Estimator(
model_fn=model_fn,
config=run_config,
params=model_params)
train_spec = tf.estimator.TrainSpec(input_fn=get_input_fn(
mode=tf.estimator.ModeKeys.TRAIN,
tfrecord_pattern=FLAGS.training_data,
batch_size=FLAGS.batch_size), max_steps=FLAGS.steps)
eval_spec = tf.estimator.EvalSpec(input_fn=get_input_fn(
mode=tf.estimator.ModeKeys.EVAL,
tfrecord_pattern=FLAGS.eval_data,
batch_size=FLAGS.batch_size))
return estimator, train_spec, eval_spec
def main(unused_args):
estimator, train_spec, eval_spec = create_estimator_and_specs(
run_config=tf.estimator.RunConfig(
model_dir=FLAGS.model_dir,
save_checkpoints_secs=300,
save_summary_steps=100))
tf.estimator.train_and_evaluate(estimator, train_spec, eval_spec)
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.register("type", "bool", lambda v: v.lower() == "true")
parser.add_argument(
"--training_data",
type=str,
default="",
help="Path to training data (tf.Example in TFRecord format)")
parser.add_argument(
"--eval_data",
type=str,
default="",
help="Path to evaluation data (tf.Example in TFRecord format)")
parser.add_argument(
"--classes_file",
type=str,
default="",
help="Path to a file with the classes - one class per line")
parser.add_argument(
"--num_layers",
type=int,
default=3,
help="Number of recurrent neural network layers.")
parser.add_argument(
"--num_nodes",
type=int,
default=128,
help="Number of node per recurrent network layer.")
parser.add_argument(
"--num_conv",
type=str,
default="[48, 64, 96]",
help="Number of conv layers along with number of filters per layer.")
parser.add_argument(
"--conv_len",
type=str,
default="[5, 5, 3]",
help="Length of the convolution filters.")
parser.add_argument(
"--cell_type",
type=str,
default="lstm",
help="Cell type used for rnn layers: cudnn_lstm, lstm or block_lstm.")
parser.add_argument(
"--batch_norm",
type="bool",
default="False",
help="Whether to enable batch normalization or not.")
parser.add_argument(
"--learning_rate",
type=float,
default=0.0001,
help="Learning rate used for training.")
parser.add_argument(
"--gradient_clipping_norm",
type=float,
default=9.0,
help="Gradient clipping norm used during training.")
parser.add_argument(
"--dropout",
type=float,
default=0.3,
help="Dropout used for convolutions and bidi lstm layers.")
parser.add_argument(
"--steps",
type=int,
default=100000,
help="Number of training steps.")
parser.add_argument(
"--batch_size",
type=int,
default=8,
help="Batch size to use for training/evaluation.")
parser.add_argument(
"--model_dir",
type=str,
default="",
help="Path for storing the model checkpoints.")
parser.add_argument(
"--self_test",
type="bool",
default="False",
help="Whether to enable batch normalization or not.")
FLAGS, unparsed = parser.parse_known_args()
tf.app.run(main=main, argv=[sys.argv[0]] + unparsed)
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment