-
-
Save byronyi/ca1b55e5a5423d5b3abb9efc6fd34b80 to your computer and use it in GitHub Desktop.
I also tried these three script in my develop machine. (a rather clean environment)
The #3 script still fail with error. Does this feature needs any pre-requisite packages?
Log from #3:
WARNING:tensorflow:Not all devices in tf.distribute.Strategy
are visible to TensorFlow.
WARNING:tensorflow:Not all devices in tf.distribute.Strategy
are visible to TensorFlow.
WARNING:tensorflow:From /home/sunbear/miniconda2/lib/python2.7/site-packages/tensorflow/python/data/ops/dataset_ops.py:1763: make_initializable_iterator (from tensorflow.python.data.ops.dataset_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use for ... in dataset:
to iterate over a dataset. If using tf.estimator
, return the Dataset
object directly from your input function. As a last resort, you can use tf.compat.v1.data.make_initializable_iterator(dataset)
.
WARNING:tensorflow:From /home/sunbear/miniconda2/lib/python2.7/site-packages/tensorflow/python/data/ops/dataset_ops.py:1458: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.
Instructions for updating:
Colocations handled automatically by placer.
WARNING:tensorflow:From /home/sunbear/miniconda2/lib/python2.7/site-packages/tensorflow/python/ops/init_ops.py:1253: calling init (from tensorflow.python.ops.init_ops) with dtype is deprecated and will be removed in a future version.
Instructions for updating:
Call initializer instance with the dtype argument instead of passing it to the constructor
E0117 19:12:52.497343637 4517 http_proxy.cc:62] 'https' scheme not supported in proxy URI
E0117 19:12:53.510391454 4515 http_proxy.cc:62] 'https' scheme not supported in proxy URI
WARNING:tensorflow:It seems that global step (tf.train.get_global_step) has not been increased. Current value (could be stable): 0 vs previous value: 0. You could increase the global step by passing tf.train.get_global_step() to Optimizer.apply_gradients or Optimizer.minimize.
Log from #1 or #2:
2019-01-17 19:12:29.215574: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA
2019-01-17 19:12:29.237936: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 3600000000 Hz
2019-01-17 19:12:29.238280: I tensorflow/compiler/xla/service/service.cc:162] XLA service 0x55a4bafcda60 executing computations on platform Host. Devices:
2019-01-17 19:12:29.238303: I tensorflow/compiler/xla/service/service.cc:169] StreamExecutor device (0): ,
2019-01-17 19:12:29.239407: I tensorflow/core/distributed_runtime/rpc/grpc_channel.cc:250] Initialize GrpcChannelCache for job worker -> {0 -> localhost:6000, 1 -> localhost:6001}
2019-01-17 19:12:29.240302: I tensorflow/core/distributed_runtime/rpc/grpc_server_lib.cc:387] Started server with target: grpc://localhost:6000
2019-01-17 19:12:29.240321: I tensorflow/core/distributed_runtime/rpc/grpc_server_lib.cc:391] Server already started (target: grpc://localhost:6000)
2019-01-17 19:12:29.240340: I tensorflow/core/distributed_runtime/rpc/grpc_server_lib.cc:391] Server already started (target: grpc://localhost:6000)
2019-01-17 19:12:53.513087: I tensorflow/core/distributed_runtime/master_session.cc:1192] Start master session 5a3ad4f61cec82ac with config: device_filters: "/job:worker/task:0" device_filters: "/job:worker/task:0" allow_soft_placement: true graph_options { rewrite_options { meta_optimizer_iterations: ONE scoped_allocator_optimization: ON scoped_allocator_opts { enable_op: "CollectiveReduce" enable_op: "CollectiveReduce" } } } experimental { collective_group_leader: "/job:worker/replica:0/task:0" }
E0117 19:12:53.525456321 4578 http_proxy.cc:62] 'https' scheme not supported in proxy URI
E0117 19:12:53.525458685 4577 http_proxy.cc:62] 'https' scheme not supported in proxy URI
E0117 19:12:53.525475494 4579 http_proxy.cc:62] 'https' scheme not supported in proxy URI
2019-01-17 19:12:53.701830: W tensorflow/core/common_runtime/base_collective_executor.cc:203] BaseCollectiveExecutor::StartAbort Out of range: End of sequence
[[{{node IteratorGetNext}}]]
byronyi,
I tried you scirpts and steps exactly same.
The worker script (#3 ) always stuck with following log : (any ideas? )
(tf-estimator-nightly 1.13.0.dev2019010910 tf-nightly 1.13.0.dev20190116)
Log:
[zhouhaiy@mlt-skx052 temp]$ python worker.py
2019-01-17 15:28:47.684458: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 AVX512F FMA
2019-01-17 15:28:47.699155: I tensorflow/core/platform/profile_utils/cpu_utils.cc:94] CPU Frequency: 2500000000 Hz
2019-01-17 15:28:47.709979: I tensorflow/compiler/xla/service/service.cc:162] XLA service 0x7e17fa0 executing computations on platfo rm Host. Devices:
2019-01-17 15:28:47.710022: I tensorflow/compiler/xla/service/service.cc:169] StreamExecutor device (0): ,
WARNING:tensorflow:Not all devices in
tf.distribute.Strategy
are visible to TensorFlow.WARNING:tensorflow:Using temporary folder as model directory: /tmp/tmpQBujoJ
WARNING:tensorflow:Not all devices in
tf.distribute.Strategy
are visible to TensorFlow.WARNING:tensorflow:Not all devices in
tf.distribute.Strategy
are visible to TensorFlow.WARNING:tensorflow:From /home/zhouhaiy/.local/lib/python2.7/site-packages/tensorflow/python/data/ops/dataset_ops.py:1763: make_initi alizable_iterator (from tensorflow.python.data.ops.dataset_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use
for ... in dataset:
to iterate over a dataset. If usingtf.estimator
, return theDataset
object directly from your input f unction. As a last resort, you can usetf.compat.v1.data.make_initializable_iterator(dataset)
.WARNING:tensorflow:From /home/zhouhaiy/.local/lib/python2.7/site-packages/tensorflow/python/data/ops/dataset_ops.py:1458: colocate_w ith (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.
Instructions for updating:
Colocations handled automatically by placer.
WARNING:tensorflow:From /home/zhouhaiy/.local/lib/python2.7/site-packages/tensorflow/python/ops/init_ops.py:1253: calling init ( from tensorflow.python.ops.init_ops) with dtype is deprecated and will be removed in a future version.
Instructions for updating:
Call initializer instance with the dtype argument instead of passing it to the constructor