【发布时间】：2019-04-29 18:59:02
【问题描述】：
尝试运行基于 Tensorflow 对象检测 API 的程序。 Faster RCNN 模型停止在 GCP 上的训练，但在本地运行没有问题。对于任何反馈，我们都表示感谢。已按照不同帖子中的建议尝试了服务代理的 Logs Writer 角色权限。找不到更多反馈。
完整的错误信息：
副本主机 0 以非零状态 1 退出。终止 原因：错误。 Traceback（最近一次调用最后一次）：文件 “/usr/lib/python2.7/runpy.py”，第 174 行，在 _run_module_as_main "main", fname, loader, pkg_name) 文件 “/usr/lib/python2.7/runpy.py”，第 72 行，在 _run_code 执行代码中 run_globals 文件 "/root/.local/lib/python2.7/site-packages/object_detection/train.py", 第 198 行，在 tf.app.run() 文件中 "/usr/local/lib/python2.7/dist-packages/tensorflow/python/platform/app.py", 第 48 行，运行中 _sys.exit(main(_sys.argv[:1] + flags_passthrough)) 文件 "/root/.local/lib/python2.7/site-packages/object_detection/train.py", 第 194 行，在主 worker_job_name、is_chief、FLAGS.train_dir) 文件中 "/root/.local/lib/python2.7/site-packages/object_detection/trainer.py", 第 296 行，在火车 saver=saver) 文件中 “/usr/local/lib/python2.7/dist-packages/tensorflow/contrib/slim/python/slim/learning.py”， 第 763 行，在火车 sess、train_op、global_step、train_step_kwargs 中） 文件 “/usr/local/lib/python2.7/dist-packages/tensorflow/contrib/slim/python/slim/learning.py”， 第 487 行，在 train_step run_metadata=run_metadata) 文件中 "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", 第 889 行，在运行 run_metadata_ptr) 文件 "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", 第 1120 行，在 _run feed_dict_tensor、options、run_metadata) 文件中 "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", 第 1317 行，在 _do_run 选项中，run_metadata) 文件 "/usr/local/lib/python2.7/dist-packages/tensorflow/python/client/session.py", 第 1336 行，在 _do_call raise type(e)(node_def, op, message) UnavailableError: Endpoint read failed 了解更多关于为什么你的 作业退出，请检查日志： https://console.cloud.google.com/logs/viewer?project=1086278442266&resource=ml_job%2Fjob_id%2Fuav_object_detection_1543356760&advancedFilter=resource.type%3D%22ml_job%22%0Aresource.labels.job_id%3D%22uav_object_detection_1543356760%22
这是我在终端中运行以开始训练的内容：
gcloud ml-engine jobs submit training `whoami`_object_detection_`date +%s` \
--job-dir=gs://my_gcs_bucket/train \
--packages dist/object_detection-0.1.tar.gz,slim/dist/slim-0.1.tar.gz \
--module-name object_detection.train \
--region us-central1 \
--config object_detection/samples/cloud/cloud.yml \
--runtime-version=1.4 \
-- \
--train_dir=gs://my_gcs_bucket/train \
--pipeline_config_path=gs://my_gcs_bucket/data/faster_rcnn_resnet101.config
这是我在 GCP 存储桶中的文件结构
+ data/
- faster_rcnn_resnet101.config
- model.ckpt.index
- model.ckpt.meta
- model.ckpt.data-00000-of-00001
- pet_label_map.pbtxt
- train.record
- val.record
+ train/
这是我正在运行的文件夹中的文件结构
+dist/
-object_detection-0.1.tar.gz
+object_detection/
+object_detection.egg-info/
+slim/
setup.py
配置文件：
# Faster R-CNN with Resnet-101 (v1) configured for the Oxford-IIIT Pet Dataset.
# Users should configure the fine_tune_checkpoint field in the train config as
# well as the label_map_path and input_path fields in the train_input_reader and
# eval_input_reader. Search for "PATH_TO_BE_CONFIGURED" to find the fields that
# should be configured.
model {
faster_rcnn {
num_classes: 1
image_resizer {
keep_aspect_ratio_resizer {
min_dimension: 600
max_dimension: 1024
}
}
feature_extractor {
type: 'faster_rcnn_resnet101'
first_stage_features_stride: 16
}
first_stage_anchor_generator {
grid_anchor_generator {
scales: [0.25, 0.5, 1.0, 2.0]
aspect_ratios: [0.5, 1.0, 2.0]
height_stride: 16
width_stride: 16
}
}
first_stage_box_predictor_conv_hyperparams {
op: CONV
regularizer {
l2_regularizer {
weight: 0.0
}
}
initializer {
truncated_normal_initializer {
stddev: 0.01
}
}
}
first_stage_nms_score_threshold: 0.0
first_stage_nms_iou_threshold: 0.7
first_stage_max_proposals: 300
first_stage_localization_loss_weight: 2.0
first_stage_objectness_loss_weight: 1.0
initial_crop_size: 14
maxpool_kernel_size: 2
maxpool_stride: 2
second_stage_box_predictor {
mask_rcnn_box_predictor {
use_dropout: false
dropout_keep_probability: 1.0
fc_hyperparams {
op: FC
regularizer {
l2_regularizer {
weight: 0.0
}
}
initializer {
variance_scaling_initializer {
factor: 1.0
uniform: true
mode: FAN_AVG
}
}
}
}
}
second_stage_post_processing {
batch_non_max_suppression {
score_threshold: 0.0
iou_threshold: 0.6
max_detections_per_class: 100
max_total_detections: 300
}
score_converter: SOFTMAX
}
second_stage_localization_loss_weight: 2.0
second_stage_classification_loss_weight: 1.0
}
}
train_config: {
batch_size: 1
batch_queue_capacity: 1
num_batch_queue_threads: 1
prefetch_queue_capacity: 1
optimizer {
momentum_optimizer: {
learning_rate: {
manual_step_learning_rate {
initial_learning_rate: 0.0003
schedule {
step: 0
learning_rate: .0003
}
schedule {
step: 900000
learning_rate: .00003
}
schedule {
step: 1200000
learning_rate: .000003
}
}
}
momentum_optimizer_value: 0.9
}
use_moving_average: false
}
gradient_clipping_by_norm: 10.0
fine_tune_checkpoint: "gs://my_gcs_bucket/data/model.ckpt"
from_detection_checkpoint: true
# Note: The below line limits the training process to 200K steps, which we
# empirically found to be sufficient enough to train the pets dataset. This
# effectively bypasses the learning rate schedule (the learning rate will
# never decay). Remove the below line to train indefinitely.
num_steps:2000
data_augmentation_options {
random_horizontal_flip {
}
}
}
train_input_reader: {
tf_record_input_reader {
input_path: "gs://my_gcs_bucket/data/data/train.record"
}
label_map_path: "gs://my_gcs_bucket/data/data/label_map.pbtxt"
queue_capacity: 10
min_after_dequeue: 5
}
eval_config: {
num_examples: 4
# Note: The below line limits the evaluation process to 10 evaluations.
# Remove the below line to evaluate indefinitely.
max_evals: 10
}
eval_input_reader: {
tf_record_input_reader {
input_path: "gs://my_gcs_bucket/data/data/val.record"
}
label_map_path: "gs://my_gcs_bucket/data/data/label_map.pbtxt"
shuffle: false
num_readers: 1
}
【问题讨论】：
-
您能否使用运行时版本 1.2 或其他版本成功运行作业？从类似的故障来看，问题似乎与各个节点之间的 gRPC 通信有关。
-
是的，使用 1.2 版是解决方案。有趣的是，我可以在本地机器上运行 1.10 就好了。
标签： tensorflow google-cloud-platform object-detection google-cloud-ml