2019-05-25 01:23:11 +02:00
|
|
|
#!/usr/bin/env python
|
2019-03-18 19:45:05 +01:00
|
|
|
# -*- coding: utf-8 -*-
|
|
|
|
|
|
|
|
# Copyright (c) 2018, NVIDIA CORPORATION. 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.
|
|
|
|
|
|
|
|
import os
|
|
|
|
|
|
|
|
import warnings
|
|
|
|
warnings.simplefilter("ignore")
|
|
|
|
|
|
|
|
import tensorflow as tf
|
|
|
|
|
|
|
|
import horovod.tensorflow as hvd
|
|
|
|
from utils import hvd_utils
|
|
|
|
|
|
|
|
from runtime import Runner
|
|
|
|
|
|
|
|
from utils.cmdline_helper import parse_cmdline
|
|
|
|
|
|
|
|
if __name__ == "__main__":
|
|
|
|
|
|
|
|
tf.logging.set_verbosity(tf.logging.ERROR)
|
|
|
|
|
|
|
|
FLAGS = parse_cmdline()
|
|
|
|
|
|
|
|
RUNNING_CONFIG = tf.contrib.training.HParams(
|
|
|
|
mode=FLAGS.mode,
|
|
|
|
|
|
|
|
# ======= Directory HParams ======= #
|
|
|
|
log_dir=FLAGS.results_dir,
|
|
|
|
model_dir=FLAGS.results_dir,
|
|
|
|
summaries_dir=FLAGS.results_dir,
|
|
|
|
data_dir=FLAGS.data_dir,
|
2019-07-08 22:51:28 +02:00
|
|
|
data_idx_dir=FLAGS.data_idx_dir,
|
2019-03-18 19:45:05 +01:00
|
|
|
|
|
|
|
# ========= Model HParams ========= #
|
|
|
|
n_classes=1001,
|
|
|
|
input_format='NHWC',
|
|
|
|
compute_format=FLAGS.data_format,
|
|
|
|
dtype=tf.float32 if FLAGS.precision == "fp32" else tf.float16,
|
|
|
|
height=224,
|
|
|
|
width=224,
|
|
|
|
n_channels=3,
|
|
|
|
|
|
|
|
# ======= Training HParams ======== #
|
|
|
|
iter_unit=FLAGS.iter_unit,
|
|
|
|
num_iter=FLAGS.num_iter,
|
|
|
|
warmup_steps=FLAGS.warmup_steps,
|
|
|
|
batch_size=FLAGS.batch_size,
|
|
|
|
log_every_n_steps=FLAGS.display_every,
|
2019-08-13 15:01:10 +02:00
|
|
|
lr_init=FLAGS.lr_init,
|
|
|
|
lr_warmup_epochs=FLAGS.lr_warmup_epochs,
|
2019-03-18 19:45:05 +01:00
|
|
|
weight_decay=FLAGS.weight_decay,
|
|
|
|
momentum=FLAGS.momentum,
|
|
|
|
loss_scale=FLAGS.loss_scale,
|
2019-08-13 15:01:10 +02:00
|
|
|
label_smoothing=FLAGS.label_smoothing,
|
|
|
|
use_cosine_lr=FLAGS.use_cosine_lr,
|
2019-05-25 01:23:11 +02:00
|
|
|
use_static_loss_scaling=FLAGS.use_static_loss_scaling,
|
2019-03-18 19:45:05 +01:00
|
|
|
distort_colors=False,
|
2019-08-13 15:01:10 +02:00
|
|
|
|
2019-03-18 19:45:05 +01:00
|
|
|
# ======= Optimization HParams ======== #
|
|
|
|
use_xla=FLAGS.use_xla,
|
|
|
|
use_tf_amp=FLAGS.use_tf_amp,
|
2019-07-08 22:51:28 +02:00
|
|
|
use_dali=FLAGS.use_dali,
|
|
|
|
gpu_memory_fraction=FLAGS.gpu_memory_fraction,
|
2019-03-18 19:45:05 +01:00
|
|
|
|
|
|
|
seed=FLAGS.seed,
|
|
|
|
)
|
|
|
|
|
|
|
|
# ===================================
|
|
|
|
|
|
|
|
runner = Runner(
|
|
|
|
# ========= Model HParams ========= #
|
|
|
|
n_classes=RUNNING_CONFIG.n_classes,
|
|
|
|
input_format=RUNNING_CONFIG.input_format,
|
|
|
|
compute_format=RUNNING_CONFIG.compute_format,
|
|
|
|
dtype=RUNNING_CONFIG.dtype,
|
|
|
|
n_channels=RUNNING_CONFIG.n_channels,
|
|
|
|
height=RUNNING_CONFIG.height,
|
|
|
|
width=RUNNING_CONFIG.width,
|
|
|
|
distort_colors=RUNNING_CONFIG.distort_colors,
|
|
|
|
log_dir=RUNNING_CONFIG.log_dir,
|
|
|
|
model_dir=RUNNING_CONFIG.model_dir,
|
|
|
|
data_dir=RUNNING_CONFIG.data_dir,
|
2019-07-08 22:51:28 +02:00
|
|
|
data_idx_dir=RUNNING_CONFIG.data_idx_dir,
|
2019-03-18 19:45:05 +01:00
|
|
|
|
|
|
|
# ======= Optimization HParams ======== #
|
|
|
|
use_xla=RUNNING_CONFIG.use_xla,
|
|
|
|
use_tf_amp=RUNNING_CONFIG.use_tf_amp,
|
2019-07-08 22:51:28 +02:00
|
|
|
use_dali=RUNNING_CONFIG.use_dali,
|
|
|
|
gpu_memory_fraction=RUNNING_CONFIG.gpu_memory_fraction,
|
2019-03-18 19:45:05 +01:00
|
|
|
seed=RUNNING_CONFIG.seed
|
|
|
|
)
|
|
|
|
|
|
|
|
if RUNNING_CONFIG.mode in ["train", "train_and_evaluate", "training_benchmark"]:
|
|
|
|
|
|
|
|
runner.train(
|
|
|
|
iter_unit=RUNNING_CONFIG.iter_unit,
|
|
|
|
num_iter=RUNNING_CONFIG.num_iter,
|
|
|
|
batch_size=RUNNING_CONFIG.batch_size,
|
|
|
|
warmup_steps=RUNNING_CONFIG.warmup_steps,
|
|
|
|
log_every_n_steps=RUNNING_CONFIG.log_every_n_steps,
|
|
|
|
weight_decay=RUNNING_CONFIG.weight_decay,
|
2019-08-13 15:01:10 +02:00
|
|
|
lr_init=RUNNING_CONFIG.lr_init,
|
|
|
|
lr_warmup_epochs=RUNNING_CONFIG.lr_warmup_epochs,
|
2019-03-18 19:45:05 +01:00
|
|
|
momentum=RUNNING_CONFIG.momentum,
|
2019-08-13 15:01:10 +02:00
|
|
|
loss_scale=RUNNING_CONFIG.loss_scale,
|
|
|
|
label_smoothing=RUNNING_CONFIG.label_smoothing,
|
|
|
|
use_static_loss_scaling=RUNNING_CONFIG.use_static_loss_scaling,
|
|
|
|
use_cosine_lr=RUNNING_CONFIG.use_cosine_lr,
|
2019-03-18 19:45:05 +01:00
|
|
|
is_benchmark=RUNNING_CONFIG.mode == 'training_benchmark',
|
2019-08-13 15:01:10 +02:00
|
|
|
|
2019-03-18 19:45:05 +01:00
|
|
|
)
|
|
|
|
|
|
|
|
if RUNNING_CONFIG.mode in ["train_and_evaluate", 'evaluate', 'inference_benchmark']:
|
|
|
|
|
|
|
|
if RUNNING_CONFIG.mode == 'inference_benchmark' and hvd_utils.is_using_hvd():
|
|
|
|
raise NotImplementedError("Only single GPU inference is implemented.")
|
|
|
|
|
|
|
|
elif not hvd_utils.is_using_hvd() or hvd.rank() == 0:
|
|
|
|
|
|
|
|
runner.evaluate(
|
|
|
|
iter_unit=RUNNING_CONFIG.iter_unit if RUNNING_CONFIG.mode != "train_and_evaluate" else "epoch",
|
|
|
|
num_iter=RUNNING_CONFIG.num_iter if RUNNING_CONFIG.mode != "train_and_evaluate" else 1,
|
|
|
|
warmup_steps=RUNNING_CONFIG.warmup_steps,
|
|
|
|
batch_size=RUNNING_CONFIG.batch_size,
|
|
|
|
log_every_n_steps=RUNNING_CONFIG.log_every_n_steps,
|
|
|
|
is_benchmark=RUNNING_CONFIG.mode == 'inference_benchmark'
|
|
|
|
)
|