TensorFlow
since 2015-11-10
リンク
-
- TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
最後まで読まないで自分がやってしまった手順をメモ：
$ virtualenv py
New python executable in py/bin/python
Installing setuptools, pip...done.
$ source py/bin/activate
$ which python
/Users/nishimotz/work/tensorFlow/py/bin/python
$ python -V
Python 2.7.10
$ which pip
/Users/nishimotz/work/tensorFlow/py/bin/pip
$ pip install https://storage.googleapis.com/tensorflow/mac/tensorflow-0.5.0-py2-none-any.whl
$ python
Python 2.7.10 (default, Jul 14 2015, 19:46:27)
[GCC 4.2.1 Compatible Apple LLVM 6.0 (clang-600.0.39)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow as tf
>>> hello = tf.constant('Hello, TensorFlow!')
>>> sess = tf.Session()
can't determine number of CPU cores: assuming 4
I tensorflow/core/common_runtime/local_device.cc:25] Local device intra op parallelism threads: 4
can't determine number of CPU cores: assuming 4
I tensorflow/core/common_runtime/local_session.cc:45] Local session inter op parallelism threads: 4
>>> print sess.run(hello)
Hello, TensorFlow!
>>> a = tf.constant(10)
>>> b = tf.constant(32)
>>> print sess.run(a+b)
42
Windows
since 2017-04-16
ひさしぶりに触る。
Windows Python 3.5.3 (32bit) がたまたま入っている。
https://www.tensorflow.org/install/install_windows
venv の説明はこちら http://bootcamp-text.readthedocs.io/textbook/6_pyvenv.html
> py -3.5 -m venv env > env\Scripts\activate > python -V Python 3.5.3 > pip install --upgrade tensorflow Collecting tensorflow Could not find a version that satisfies the requirement tensorflow (from versions: ) No matching distribution found for tensorflow
ないじゃん。。x64 でないからかな。。
Windows Python 3.5.3 (AMD64) を追加インストール。
> py -3.5 -m venv env
> env\Scripts\activate
> python -V
Python 3.5.3
> pip install --upgrade tensorflow
Collecting tensorflow
Downloading tensorflow-1.0.1-cp35-cp35m-win_amd64.whl (14.7MB)
100% |################################| 14.7MB 62kB/s
Collecting protobuf>=3.1.0 (from tensorflow)
Downloading protobuf-3.2.0-py2.py3-none-any.whl (360kB)
100% |################################| 368kB 858kB/s
Collecting wheel>=0.26 (from tensorflow)
Downloading wheel-0.29.0-py2.py3-none-any.whl (66kB)
100% |################################| 71kB 1.3MB/s
Collecting numpy>=1.11.0 (from tensorflow)
Downloading numpy-1.12.1-cp35-none-win_amd64.whl (7.7MB)
100% |################################| 7.7MB 144kB/s
Collecting six>=1.10.0 (from tensorflow)
Using cached six-1.10.0-py2.py3-none-any.whl
Collecting setuptools (from protobuf>=3.1.0->tensorflow)
Downloading setuptools-35.0.0-py2.py3-none-any.whl (390kB)
100% |################################| 399kB 936kB/s
Collecting packaging>=16.8 (from setuptools->protobuf>=3.1.0->tensorflow)
Using cached packaging-16.8-py2.py3-none-any.whl
Collecting appdirs>=1.4.0 (from setuptools->protobuf>=3.1.0->tensorflow)
Downloading appdirs-1.4.3-py2.py3-none-any.whl
Collecting pyparsing (from packaging>=16.8->setuptools->protobuf>=3.1.0->tensorflow)
Downloading pyparsing-2.2.0-py2.py3-none-any.whl (56kB)
100% |################################| 61kB 2.4MB/s
Installing collected packages: six, pyparsing, packaging, appdirs, setuptools, protobuf, wheel, numpy, tensorflow
Found existing installation: setuptools 28.8.0
Uninstalling setuptools-28.8.0:
Successfully uninstalled setuptools-28.8.0
Successfully installed appdirs-1.4.3 numpy-1.12.1 packaging-16.8 protobuf-3.2.0 pyparsing-2.2.0 setuptools-35.0.0 six-1.10.0 tensorflow-1.0.1 wheel-0.29.0
うまくいった。
> pip install -U tensorflow-gpu
Collecting tensorflow-gpu
Downloading tensorflow_gpu-1.0.1-cp35-cp35m-win_amd64.whl (43.1MB)
100% |################################| 43.1MB 23kB/s
Requirement already up-to-date: numpy>=1.11.0 in c:\work\2017\tf\env\lib\site-packages (from tensorflow-gpu)
Requirement already up-to-date: six>=1.10.0 in c:\work\2017\tf\env\lib\site-packages (from tensorflow-gpu)
Requirement already up-to-date: wheel>=0.26 in c:\work\2017\tf\env\lib\site-packages (from tensorflow-gpu)
Requirement already up-to-date: protobuf>=3.1.0 in c:\work\2017\tf\env\lib\site-packages (from tensorflow-gpu)
Requirement already up-to-date: setuptools in c:\work\2017\tf\env\lib\site-packages (from protobuf>=3.1.0->tensorflow-gpu)
Requirement already up-to-date: appdirs>=1.4.0 in c:\work\2017\tf\env\lib\site-packages (from setuptools->protobuf>=3.1.0->tensorflow-gpu)
Requirement already up-to-date: packaging>=16.8 in c:\work\2017\tf\env\lib\site-packages (from setuptools->protobuf>=3.1.0->tensorflow-gpu)
Requirement already up-to-date: pyparsing in c:\work\2017\tf\env\lib\site-packages (from packaging>=16.8->setuptools->protobuf>=3.1.0->tensorflow-gpu)
Installing collected packages: tensorflow-gpu
Successfully installed tensorflow-gpu-1.0.1
こっちも入った。
しかし、対応ハードウェアが必要だったらしい。。
> python
Python 3.5.3 (v3.5.3:1880cb95a742, Jan 16 2017, 16:02:32) [MSC v.1900 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow as tf
I c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\stream_executor\dso_loader.cc:126] Couldn't open CUDA library cublas64_80.dll
I c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\stream_executor\cuda\cuda_blas.cc:2294] Unable to load cuBLAS DSO.
I c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\stream_executor\dso_loader.cc:126] Couldn't open CUDA library cudnn64_5.dll
I c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\stream_executor\cuda\cuda_dnn.cc:3517] Unable to load cuDNN DSO
I c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\stream_executor\dso_loader.cc:126] Couldn't open CUDA library cufft64_80.dll
I c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\stream_executor\cuda\cuda_fft.cc:344] Unable to load cuFFT DSO.
I c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\stream_executor\dso_loader.cc:126] Couldn't open CUDA library nvcuda.dll
I c:\tf_jenkins\home\workspace\release-win\device\gpu\os\windows\tensorflow\stream_executor\cuda\cuda_diagnostics.cc:165] hostname: ・誹・
Traceback (most recent call last):
File "C:\work\2017\tf\env\lib\site-packages\tensorflow\python\pywrap_tensorflow.py", line 18, in swig_import_helper
return importlib.import_module(mname)
File "C:\Users\nishimotz\AppData\Local\Programs\Python\Python35\lib\importlib\__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
File "<frozen importlib._bootstrap>", line 986, in _gcd_import
File "<frozen importlib._bootstrap>", line 969, in _find_and_load
File "<frozen importlib._bootstrap>", line 958, in _find_and_load_unlocked
File "<frozen importlib._bootstrap>", line 666, in _load_unlocked
File "<frozen importlib._bootstrap>", line 577, in module_from_spec
File "<frozen importlib._bootstrap_external>", line 914, in create_module
File "<frozen importlib._bootstrap>", line 222, in _call_with_frames_removed
ImportError: DLL load failed: A dynamic link library (DLL) initialization routine failed.
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "C:\work\2017\tf\env\lib\site-packages\tensorflow\python\__init__.py", line 66, in <module>
from tensorflow.python import pywrap_tensorflow
File "C:\work\2017\tf\env\lib\site-packages\tensorflow\python\pywrap_tensorflow.py", line 21, in <module>
_pywrap_tensorflow = swig_import_helper()
File "C:\work\2017\tf\env\lib\site-packages\tensorflow\python\pywrap_tensorflow.py", line 20, in swig_import_helper
return importlib.import_module('_pywrap_tensorflow')
File "C:\Users\nishimotz\AppData\Local\Programs\Python\Python35\lib\importlib\__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
ImportError: No module named '_pywrap_tensorflow'
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
File "C:\work\2017\tf\env\lib\site-packages\tensorflow\__init__.py", line 24, in <module>
from tensorflow.python import *
File "C:\work\2017\tf\env\lib\site-packages\tensorflow\python\__init__.py", line 72, in <module>
raise ImportError(msg)
ImportError: Traceback (most recent call last):
File "C:\work\2017\tf\env\lib\site-packages\tensorflow\python\pywrap_tensorflow.py", line 18, in swig_import_helper
return importlib.import_module(mname)
File "C:\Users\nishimotz\AppData\Local\Programs\Python\Python35\lib\importlib\__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
File "<frozen importlib._bootstrap>", line 986, in _gcd_import
File "<frozen importlib._bootstrap>", line 969, in _find_and_load
File "<frozen importlib._bootstrap>", line 958, in _find_and_load_unlocked
File "<frozen importlib._bootstrap>", line 666, in _load_unlocked
File "<frozen importlib._bootstrap>", line 577, in module_from_spec
File "<frozen importlib._bootstrap_external>", line 914, in create_module
File "<frozen importlib._bootstrap>", line 222, in _call_with_frames_removed
ImportError: DLL load failed: A dynamic link library (DLL) initialization routine failed.
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "C:\work\2017\tf\env\lib\site-packages\tensorflow\python\__init__.py", line 66, in <module>
from tensorflow.python import pywrap_tensorflow
File "C:\work\2017\tf\env\lib\site-packages\tensorflow\python\pywrap_tensorflow.py", line 21, in <module>
_pywrap_tensorflow = swig_import_helper()
File "C:\work\2017\tf\env\lib\site-packages\tensorflow\python\pywrap_tensorflow.py", line 20, in swig_import_helper
return importlib.import_module('_pywrap_tensorflow')
File "C:\Users\nishimotz\AppData\Local\Programs\Python\Python35\lib\importlib\__init__.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
ImportError: No module named '_pywrap_tensorflow'
Failed to load the native TensorFlow runtime.
See https://github.com/tensorflow/tensorflow/blob/master/tensorflow/g3doc/get_started/os_setup.md#import_error
for some common reasons and solutions. Include the entire stack trace
above this error message when asking for help.
けっきょく pip uninstall tensorflow-gpu tensorflow して、インストールからやり直す。
> python
Python 3.5.3 (v3.5.3:1880cb95a742, Jan 16 2017, 16:02:32) [MSC v.1900 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> import tensorflow as tf
>>> hello = tf.constant('Hello, TensorFlow!')
>>> sess = tf.Session()
>>> print(sess.run(hello))
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "BestSplits" device_type: "CPU"') for unknown op: BestSplits
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "CountExtremelyRandomStats" device_type: "CPU"') for unknown op: CountExtremelyRandomStats
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "FinishedNodes" device_type: "CPU"') for unknown op: FinishedNodes
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "GrowTree" device_type: "CPU"') for unknown op: GrowTree
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "ReinterpretStringToFloat" device_type: "CPU"') for unknown op: ReinterpretStringToFloat
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "SampleInputs" device_type: "CPU"') for unknown op: SampleInputs
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "ScatterAddNdim" device_type: "CPU"') for unknown op: ScatterAddNdim
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "TopNInsert" device_type: "CPU"') for unknown op: TopNInsert
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "TopNRemove" device_type: "CPU"') for unknown op: TopNRemove
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "TreePredictions" device_type: "CPU"') for unknown op: TreePredictions
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "UpdateFertileSlots" device_type: "CPU"') for unknown op: UpdateFertileSlots
b'Hello, TensorFlow!'
まあいいんだろう、これで。。
tf-seq2seq
seq2seq もためす。
https://google.github.io/seq2seq/getting_started/
> git clone https://github.com/google/seq2seq.git
Cloning into 'seq2seq'...
remote: Counting objects: 5965, done.
remote: Compressing objects: 100% (48/48), done.
remote: Total 5965 (delta 22), reused 0 (delta 0), pack-reused 5917
Receiving objects: 100% (5965/5965), 1.62 MiB | 92.00 KiB/s, done.
Resolving deltas: 100% (4158/4158), done.
> cd seq2seq
> pip install -e .
Obtaining file:///C:/work/2017/tf/seq2seq
Requirement already satisfied: numpy in c:\work\2017\tf\env\lib\site-packages (from seq2seq==0.1)
Collecting matplotlib (from seq2seq==0.1)
Downloading matplotlib-2.0.0-cp35-cp35m-win_amd64.whl (9.0MB)
100% |################################| 9.0MB 96kB/s
Collecting pyyaml (from seq2seq==0.1)
Downloading PyYAML-3.12-cp35-cp35m-win_amd64.whl (195kB)
100% |################################| 204kB 823kB/s
Collecting pyrouge (from seq2seq==0.1)
Downloading pyrouge-0.1.3.tar.gz (60kB)
100% |################################| 61kB 712kB/s
Collecting python-dateutil (from matplotlib->seq2seq==0.1)
Downloading python_dateutil-2.6.0-py2.py3-none-any.whl (194kB)
100% |################################| 194kB 474kB/s
Collecting cycler>=0.10 (from matplotlib->seq2seq==0.1)
Downloading cycler-0.10.0-py2.py3-none-any.whl
Collecting pytz (from matplotlib->seq2seq==0.1)
Downloading pytz-2017.2-py2.py3-none-any.whl (484kB)
100% |################################| 491kB 391kB/s
Requirement already satisfied: six>=1.10 in c:\work\2017\tf\env\lib\site-packages (from matplotlib->seq2seq==0.1)
Requirement already satisfied: pyparsing!=2.0.4,!=2.1.2,!=2.1.6,>=1.5.6 in c:\work\2017\tf\env\lib\site-packages (from matplotlib->seq2seq==0.1)
Building wheels for collected packages: pyrouge
Running setup.py bdist_wheel for pyrouge ... done
Stored in directory: C:\Users\nishimotz\AppData\Local\pip\Cache\wheels\3c\e1\54\9e59592032c08e38e9f700f3b430a06e02eb6fa77e699d02ce
Successfully built pyrouge
Installing collected packages: python-dateutil, cycler, pytz, matplotlib, pyyaml, pyrouge, seq2seq
Running setup.py develop for seq2seq
Successfully installed cycler-0.10.0 matplotlib-2.0.0 pyrouge-0.1.3 python-dateutil-2.6.0 pytz-2017.2 pyyaml-3.12 seq2seq
テスト
> python -m unittest seq2seq.test.pipeline_test
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "BestSplits" device_type: "CPU"') for unknown op: BestSplits
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "CountExtremelyRandomStats" device_type: "CPU"') for unknown op: CountExtremelyRandomStats
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "FinishedNodes" device_type: "CPU"') for unknown op: FinishedNodes
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "GrowTree" device_type: "CPU"') for unknown op: GrowTree
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "ReinterpretStringToFloat" device_type: "CPU"') for unknown op: ReinterpretStringToFloat
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "SampleInputs" device_type: "CPU"') for unknown op: SampleInputs
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "ScatterAddNdim" device_type: "CPU"') for unknown op: ScatterAddNdim
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "TopNInsert" device_type: "CPU"') for unknown op: TopNInsert
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "TopNRemove" device_type: "CPU"') for unknown op: TopNRemove
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "TreePredictions" device_type: "CPU"') for unknown op: TreePredictions
E c:\tf_jenkins\home\workspace\release-win\device\cpu\os\windows\tensorflow\core\framework\op_kernel.cc:943] OpKernel ('op: "UpdateFertileSlots" device_type: "CPU"') for unknown op: UpdateFertileSlots
.INFO:tensorflow:Loading config from C:\Users\NISHIM~1\AppData\Local\Temp\tmpa_i3l_wj\train_config.yml
INFO:tensorflow:Final Config:
input_pipeline_dev:
class: ParallelTextInputPipeline
params:
source_files: ['C:\Users\NISHIM~1\AppData\Local\Temp\tmp2ld4u5ss']
target_files: ['C:\Users\NISHIM~1\AppData\Local\Temp\tmpuo0hqt9l']
input_pipeline_train:
class: ParallelTextInputPipeline
params:
source_files: ['C:\Users\NISHIM~1\AppData\Local\Temp\tmp79bixs0q']
target_files: ['C:\Users\NISHIM~1\AppData\Local\Temp\tmphefrqhe0']
model_params:
decoder.params:
rnn_cell:
cell_class: GRUCell
cell_params: {num_units: 8}
embedding.dim: 10
encoder.params:
rnn_cell:
cell_class: GRUCell
cell_params: {num_units: 8}
train_steps: 50
INFO:tensorflow:Setting save_checkpoints_secs to 600
INFO:tensorflow:Creating ParallelTextInputPipeline in mode=train
INFO:tensorflow:
ParallelTextInputPipeline:
num_epochs: null
shuffle: true
source_delimiter: ' '
source_files: ['C:\Users\NISHIM~1\AppData\Local\Temp\tmp79bixs0q']
target_delimiter: ' '
target_files: ['C:\Users\NISHIM~1\AppData\Local\Temp\tmphefrqhe0']
INFO:tensorflow:Creating ParallelTextInputPipeline in mode=eval
INFO:tensorflow:
ParallelTextInputPipeline:
num_epochs: 1
shuffle: false
source_delimiter: ' '
source_files: ['C:\Users\NISHIM~1\AppData\Local\Temp\tmp2ld4u5ss']
target_delimiter: ' '
target_files: ['C:\Users\NISHIM~1\AppData\Local\Temp\tmpuo0hqt9l']
INFO:tensorflow:Using config: {'_keep_checkpoint_every_n_hours': 4, '_task_type': None, '_evaluation_master': '', '_master': '', '_environment': 'local', '_save_checkpoints_secs': 600, '_save_summary_steps': 100, '_task_id': 0, '_tf_random_seed': None, '_cluster_spec': <tensorflow.python.training.server_lib.ClusterSpec object at 0x00000214D1DFEB38>, '_save_checkpoints_steps': None, '_tf_config': gpu_options {
per_process_gpu_memory_fraction: 1.0
}
, '_keep_checkpoint_max': 5, '_num_ps_replicas': 0, '_is_chief': True}
C:\work\2017\tf\env\lib\site-packages\tensorflow\contrib\learn\python\learn\estimators\estimator.py:181: DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead
return inspect.getargspec(func).args
INFO:tensorflow:Creating PrintModelAnalysisHook in mode=train
INFO:tensorflow:
PrintModelAnalysisHook: {}
INFO:tensorflow:Creating MetadataCaptureHook in mode=train
INFO:tensorflow:
MetadataCaptureHook: {step: 10}
INFO:tensorflow:Creating TrainSampleHook in mode=train
INFO:tensorflow:
TrainSampleHook: {every_n_secs: null, every_n_steps: 1000, source_delimiter: ' ',
target_delimiter: ' '}
INFO:tensorflow:Creating LogPerplexityMetricSpec in mode=eval
INFO:tensorflow:
LogPerplexityMetricSpec: {}
INFO:tensorflow:Creating BleuMetricSpec in mode=eval
INFO:tensorflow:
BleuMetricSpec: {eos_token: SEQUENCE_END, postproc_fn: '', separator: ' ', sos_token: SEQUENCE_START}
INFO:tensorflow:Creating RougeMetricSpec in mode=eval
INFO:tensorflow:
RougeMetricSpec: {eos_token: SEQUENCE_END, postproc_fn: '', rouge_type: rouge_1/f_score,
separator: ' ', sos_token: SEQUENCE_START}
WARNING:tensorflow:From C:\work\2017\tf\env\lib\site-packages\tensorflow\contrib\learn\python\learn\monitors.py:267: BaseMonitor.__init__ (from tensorflow.contrib.learn.python.learn.monitors) is deprecated and will be removed after 2016-12-05.
Instructions for updating:
Monitors are deprecated. Please use tf.train.SessionRunHook.
INFO:tensorflow:Creating AttentionSeq2Seq in mode=train
INFO:tensorflow:
AttentionSeq2Seq:
attention.class: AttentionLayerBahdanau
attention.params: {num_units: 10}
bridge.class: seq2seq.models.bridges.ZeroBridge
bridge.params: {}
decoder.class: seq2seq.decoders.AttentionDecoder
decoder.params:
rnn_cell:
cell_class: GRUCell
cell_params: {num_units: 8}
embedding.dim: 10
embedding.init_scale: 0.04
embedding.share: false
encoder.class: seq2seq.encoders.BidirectionalRNNEncoder
encoder.params:
rnn_cell:
cell_class: GRUCell
cell_params: {num_units: 8}
inference.beam_search.beam_width: 0
inference.beam_search.choose_successors_fn: choose_top_k
inference.beam_search.length_penalty_weight: 0.0
optimizer.clip_embed_gradients: 0.1
optimizer.clip_gradients: 5.0
optimizer.learning_rate: 0.0001
optimizer.lr_decay_rate: 0.99
optimizer.lr_decay_steps: 100
optimizer.lr_decay_type: ''
optimizer.lr_min_learning_rate: 1.0e-12
optimizer.lr_staircase: false
optimizer.lr_start_decay_at: 0
optimizer.lr_stop_decay_at: 2147483647
optimizer.name: Adam
optimizer.params: {}
optimizer.sync_replicas: 0
optimizer.sync_replicas_to_aggregate: 0
source.max_seq_len: 50
source.reverse: true
target.max_seq_len: 50
vocab_source: C:\Users\NISHIM~1\AppData\Local\Temp\tmp2s79yt9f
vocab_target: C:\Users\NISHIM~1\AppData\Local\Temp\tmpz6_mvzmb
INFO:tensorflow:Creating vocabulary lookup table of size 7
INFO:tensorflow:Creating vocabulary lookup table of size 7
INFO:tensorflow:Creating BidirectionalRNNEncoder in mode=train
INFO:tensorflow:
BidirectionalRNNEncoder:
init_scale: 0.04
rnn_cell:
cell_class: GRUCell
cell_params: {num_units: 8}
dropout_input_keep_prob: 1.0
dropout_output_keep_prob: 1.0
num_layers: 1
residual_combiner: add
residual_connections: false
residual_dense: false
C:\work\2017\tf\seq2seq\seq2seq\training\utils.py:118: DeprecationWarning: inspect.getargspec() is deprecated, use inspect.signature() instead
cell_args = set(inspect.getargspec(cell_class.__init__).args[1:])
INFO:tensorflow:Creating AttentionLayerBahdanau in mode=train
INFO:tensorflow:
AttentionLayerBahdanau: {num_units: 10}
INFO:tensorflow:Creating AttentionDecoder in mode=train
INFO:tensorflow:
AttentionDecoder:
init_scale: 0.04
max_decode_length: 100
rnn_cell:
cell_class: GRUCell
cell_params: {num_units: 8}
dropout_input_keep_prob: 1.0
dropout_output_keep_prob: 1.0
num_layers: 1
residual_combiner: add
residual_connections: false
residual_dense: false
INFO:tensorflow:Creating ZeroBridge in mode=train
INFO:tensorflow:
ZeroBridge: {}
INFO:tensorflow:Create CheckpointSaverHook.
EC:\Users\nishimotz\AppData\Local\Programs\Python\Python35\lib\unittest\case.py:629: ResourceWarning: unclosed file <_io.BufferedRandom name=3>
outcome.errors.clear()
C:\Users\nishimotz\AppData\Local\Programs\Python\Python35\lib\unittest\case.py:629: ResourceWarning: unclosed file <_io.BufferedRandom name=4>
outcome.errors.clear()
C:\Users\nishimotz\AppData\Local\Programs\Python\Python35\lib\unittest\case.py:629: ResourceWarning: unclosed file <_io.BufferedRandom name=5>
outcome.errors.clear()
C:\Users\nishimotz\AppData\Local\Programs\Python\Python35\lib\unittest\case.py:629: ResourceWarning: unclosed file <_io.BufferedRandom name=6>
outcome.errors.clear()
C:\Users\nishimotz\AppData\Local\Programs\Python\Python35\lib\unittest\case.py:629: ResourceWarning: unclosed file <_io.BufferedRandom name=7>
outcome.errors.clear()
C:\Users\nishimotz\AppData\Local\Programs\Python\Python35\lib\unittest\case.py:629: ResourceWarning: unclosed file <_io.BufferedRandom name=8>
outcome.errors.clear()
======================================================================
ERROR: test_train_infer (seq2seq.test.pipeline_test.PipelineTest)
Tests training and inference scripts.
----------------------------------------------------------------------
Traceback (most recent call last):
File "C:\work\2017\tf\seq2seq\seq2seq\test\pipeline_test.py", line 148, in test_train_infer
train_script.main([])
File "C:\work\2017\tf\seq2seq\bin\train.py", line 271, in main
schedule=FLAGS.schedule)
File "C:\work\2017\tf\env\lib\site-packages\tensorflow\contrib\learn\python\learn\learn_runner.py", line 106, in run
return task()
File "C:\work\2017\tf\env\lib\site-packages\tensorflow\contrib\learn\python\learn\experiment.py", line 459, in train_and_evaluate
self.train(delay_secs=0)
File "C:\work\2017\tf\env\lib\site-packages\tensorflow\contrib\learn\python\learn\experiment.py", line 281, in train
monitors=self._train_monitors + extra_hooks)
File "C:\work\2017\tf\env\lib\site-packages\tensorflow\python\util\deprecation.py", line 280, in new_func
return func(*args, **kwargs)
File "C:\work\2017\tf\env\lib\site-packages\tensorflow\contrib\learn\python\learn\estimators\estimator.py", line 426, in fit
loss = self._train_model(input_fn=input_fn, hooks=hooks)
File "C:\work\2017\tf\env\lib\site-packages\tensorflow\contrib\learn\python\learn\estimators\estimator.py", line 981, in _train_model
config=self.config.tf_config) as mon_sess:
File "C:\work\2017\tf\env\lib\site-packages\tensorflow\python\training\monitored_session.py", line 315, in MonitoredTrainingSession
return MonitoredSession(session_creator=session_creator, hooks=all_hooks)
File "C:\work\2017\tf\env\lib\site-packages\tensorflow\python\training\monitored_session.py", line 601, in __init__
session_creator, hooks, should_recover=True)
File "C:\work\2017\tf\env\lib\site-packages\tensorflow\python\training\monitored_session.py", line 428, in __init__
h.begin()
File "C:\work\2017\tf\seq2seq\seq2seq\training\hooks.py", line 246, in begin
opts = tf.contrib.tfprof.model_analyzer.TRAINABLE_VARS_PARAMS_STAT_OPTIONS
AttributeError: module 'tensorflow.contrib.tfprof' has no attribute 'model_analyzer'
----------------------------------------------------------------------
Ran 2 tests in 7.095s
FAILED (errors=1)
エラーになった。さて。。
https://github.com/google/seq2seq/issues/159
ここはまだ Windows に対応してないという話。
Bash on Ubuntu on Windows で試す
python3.5 は自分で make install していた。
$ which python3.5 /usr/local/bin/python3.5 $ python3.5 -m venv env-linux $ . env-linux/bin/activate (env-linux) nishimotz@LAPTOP-SH8BQD3I:/mnt/c/work/2017/tf$ pip install -U tensorflow Collecting tensorflow Downloading tensorflow-1.0.1-cp35-cp35m-manylinux1_x86_64.whl (44.1MB)
中略。 seq2seq の unittest の結果
----------------------------------------------------------------------
Traceback (most recent call last):
File "/mnt/c/work/2017/tf/seq2seq/seq2seq/test/pipeline_test.py", line 185, in test_train_infer
infer_script.main([])
File "/mnt/c/work/2017/tf/seq2seq/bin/infer.py", line 125, in main
sess.run([])
File "/mnt/c/work/2017/tf/env-linux/lib/python3.5/site-packages/tensorflow/python/training/monitored_session.py", line 462, in run
run_metadata=run_metadata)
File "/mnt/c/work/2017/tf/env-linux/lib/python3.5/site-packages/tensorflow/python/training/monitored_session.py", line 786, in run
run_metadata=run_metadata)
File "/mnt/c/work/2017/tf/env-linux/lib/python3.5/site-packages/tensorflow/python/training/monitored_session.py", line 744, in run
return self._sess.run(*args, **kwargs)
File "/mnt/c/work/2017/tf/env-linux/lib/python3.5/site-packages/tensorflow/python/training/monitored_session.py", line 899, in run
run_metadata=run_metadata))
File "/mnt/c/work/2017/tf/seq2seq/seq2seq/tasks/dump_attention.py", line 126, in after_run
_create_figure(fetches)
File "/mnt/c/work/2017/tf/seq2seq/seq2seq/tasks/dump_attention.py", line 58, in _create_figure
fig = plt.figure(figsize=(8, 8))
File "/mnt/c/work/2017/tf/env-linux/lib/python3.5/site-packages/matplotlib/pyplot.py", line 535, in figure
**kwargs)
File "/mnt/c/work/2017/tf/env-linux/lib/python3.5/site-packages/matplotlib/backends/backend_tkagg.py", line 81, in new_figure_manager
return new_figure_manager_given_figure(num, figure)
File "/mnt/c/work/2017/tf/env-linux/lib/python3.5/site-packages/matplotlib/backends/backend_tkagg.py", line 89, in new_figure_manager_given_figure
window = Tk.Tk()
File "/usr/local/lib/python3.5/tkinter/__init__.py", line 1877, in __init__
self.tk = _tkinter.create(screenName, baseName, className, interactive, wantobjects, useTk, sync, use)
_tkinter.TclError: no display name and no $DISPLAY environment variable
----------------------------------------------------------------------
Ran 2 tests in 12.785s
FAILED (errors=1)
Segmentation fault (core dumped)
じゃあ vcXsrv を実行しておこうか。
$ export DISPLAY=localhost:0
やり直す。
INFO:tensorflow:Restored model from /tmp/tmp3j3759e5/model.ckpt-50 c c c c . ---------------------------------------------------------------------- Ran 2 tests in 19.904s OK *** Error in `python': double free or corruption (!prev): 0x00000000013d2da0 *** Aborted (core dumped)
たぶん正常終了なのだが、ウィンドウは表示されず、 core dumped で終わった。
やっぱり Mac か Ubuntu でやるのが素直らしい。。
だが matplotlib にこだわらなければ Windows でも BoW でも動くものはある。
> python -m unittest seq2seq.test.data_test
などなど。。詳細の調査はこれから。
