Open coderguanmingyang opened 7 years ago
thanks for replay~
Hey, I had the same issue, it's caused by versions. This combo finally worked:
Robert@gpu-instance-2:~/stackgan$ pip show tensorflow-gpu
Name: tensorflow-gpu
Version: 0.12.0
Summary: TensorFlow helps the tensors flow
Home-page: http://tensorflow.org/
Author: Google Inc.
Author-email: opensource@google.com
License: Apache 2.0
Location: /usr/local/lib/python2.7/dist-packages
Requires: mock, numpy, protobuf, wheel, six
Robert@gpu-instance-2:~/stackgan$ pip show prettytensor
Name: prettytensor
Version: 0.6.2
Summary: Pretty Tensor makes learning beautiful
Home-page: https://github.com/google/prettytensor
Author: Eider Moore
Author-email: opensource@google.com
License: Apache 2.0
Location: /usr/local/lib/python2.7/dist-packages
Requires: six, enum34, tensorflow
@robertlacok I've ran into the same issue and tried you're combo above, but not sure I'm doing it right:
I did pip install tensorflow-gpu==0.12.0
and that worked after upgrading to CUDA 8.0 (from 7.5)
and then pip install prettytensor==0.6.2
but the install process looked a little weird:
Collecting prettytensor==0.6.2
Downloading prettytensor-0.6.2-py2-none-any.whl (264kB)
100% |████████████████████████████████| 266kB 1.7MB/s
Collecting tensorflow>=0.9.0rc0 (from prettytensor==0.6.2)
Downloading tensorflow-1.2.0rc2-cp27-cp27m-macosx_10_11_x86_64.whl (34.2MB)
100% |████████████████████████████████| 34.2MB 38kB/s
Requirement already satisfied: enum34>=1.0.0 in ~/tf-venv/lib/python2.7/site-packages (from prettytensor==0.6.2)
Requirement already satisfied: six>=1.10.0 in /usr/local/lib/python2.7/site-packages (from prettytensor==0.6.2)
Collecting markdown==2.2.0 (from tensorflow>=0.9.0rc0->prettytensor==0.6.2)
Downloading Markdown-2.2.0.tar.gz (236kB)
100% |████████████████████████████████| 245kB 1.7MB/s
Collecting html5lib==0.9999999 (from tensorflow>=0.9.0rc0->prettytensor==0.6.2)
Downloading html5lib-0.9999999.tar.gz (889kB)
100% |████████████████████████████████| 890kB 863kB/s
Requirement already satisfied: werkzeug>=0.11.10 in ~/tf-venv/lib/python2.7/site-packages (from tensorflow>=0.9.0rc0->prettytensor==0.6.2)
Requirement already satisfied: backports.weakref==1.0rc1 in /usr/local/lib/python2.7/site-packages (from tensorflow>=0.9.0rc0->prettytensor==0.6.2)
Requirement already satisfied: protobuf>=3.2.0 in /usr/local/lib/python2.7/site-packages (from tensorflow>=0.9.0rc0->prettytensor==0.6.2)
Collecting bleach==1.5.0 (from tensorflow>=0.9.0rc0->prettytensor==0.6.2)
Downloading bleach-1.5.0-py2.py3-none-any.whl
Requirement already satisfied: numpy>=1.11.0 in /usr/local/lib/python2.7/site-packages (from tensorflow>=0.9.0rc0->prettytensor==0.6.2)
Requirement already satisfied: mock>=2.0.0 in /usr/local/lib/python2.7/site-packages (from tensorflow>=0.9.0rc0->prettytensor==0.6.2)
Requirement already satisfied: wheel in /usr/local/lib/python2.7/site-packages (from tensorflow>=0.9.0rc0->prettytensor==0.6.2)
Requirement already satisfied: setuptools in /usr/local/lib/python2.7/site-packages (from protobuf>=3.2.0->tensorflow>=0.9.0rc0->prettytensor==0.6.2)
Requirement already satisfied: funcsigs>=1; python_version < "3.3" in /usr/local/lib/python2.7/site-packages (from mock>=2.0.0->tensorflow>=0.9.0rc0->prettytensor==0.6.2)
Requirement already satisfied: pbr>=0.11 in /usr/local/lib/python2.7/site-packages (from mock>=2.0.0->tensorflow>=0.9.0rc0->prettytensor==0.6.2)
Building wheels for collected packages: markdown, html5lib
Running setup.py bdist_wheel for markdown ... done
Stored in directory: /Users/George/Library/Caches/pip/wheels/b9/4f/6c/f4c1c5207c1d0eeaaf7005f7f736620c6ded6617c9d9b94096
Running setup.py bdist_wheel for html5lib ... done
Stored in directory: /Users/George/Library/Caches/pip/wheels/6f/85/6c/56b8e1292c6214c4eb73b9dda50f53e8e977bf65989373c962
Successfully built markdown html5lib
Installing collected packages: markdown, html5lib, bleach, tensorflow, prettytensor
Found existing installation: Markdown 2.6.8
Not uninstalling markdown at /usr/local/lib/python2.7/site-packages, outside environment ~/tf-venv/bin/..
Found existing installation: html5lib 0.999999999
Not uninstalling html5lib at /usr/local/lib/python2.7/site-packages, outside environment ~/tf-venv/bin/..
Found existing installation: bleach 2.0.0
Not uninstalling bleach at /usr/local/lib/python2.7/site-packages, outside environment ~/tf-venv/bin/..
Successfully installed bleach-1.5.0 html5lib-0.9999999 markdown-2.2.0 prettytensor-0.6.2 tensorflow-1.2.0rc2
This is what pip shows:
pip show tensorflow-gpu
Name: tensorflow-gpu
Version: 0.12.0
Summary: TensorFlow helps the tensors flow
Home-page: http://tensorflow.org/
Author: Google Inc.
Author-email: opensource@google.com
License: Apache 2.0
Location: ~/tf-venv/lib/python2.7/site-packages
Requires: protobuf, six, wheel, mock, numpy
pip show prettytensor
Name: prettytensor
Version: 0.6.2
Summary: Pretty Tensor makes learning beautiful
Home-page: https://github.com/google/prettytensor
Author: Eider Moore
Author-email: opensource@google.com
License: Apache 2.0
Location: ~/tf-venv/lib/python2.7/site-packages
Requires: tensorflow, enum34, six
However in Python I get this:
>>> import tensorflow as tf
>>> tf.__version__
'1.2.0-rc2'
and now I'm back to the int32 vs prettytensor Layer issue:
./demo/birds_demo.sh
Successfully load sentences from: Data/birds/example_captions.t7
Total number of sentences: 6
num_embeddings: 6 (6, 1024)
2017-06-08 23:25:07.035773: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations.
2017-06-08 23:25:07.035791: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations.
2017-06-08 23:25:07.035796: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations.
2017-06-08 23:25:07.035801: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations.
2017-06-08 23:25:07.035806: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use FMA instructions, but these are available on your machine and could speed up CPU computations.
lr_imsize: 64
WARNING:tensorflow:VARIABLES collection name is deprecated, please use GLOBAL_VARIABLES instead; VARIABLES will be removed after 2017-03-02.
WARNING:tensorflow:VARIABLES collection name is deprecated, please use GLOBAL_VARIABLES instead; VARIABLES will be removed after 2017-03-02.
WARNING:tensorflow:VARIABLES collection name is deprecated, please use GLOBAL_VARIABLES instead; VARIABLES will be removed after 2017-03-02.
WARNING:tensorflow:VARIABLES collection name is deprecated, please use GLOBAL_VARIABLES instead; VARIABLES will be removed after 2017-03-02.
Traceback (most recent call last):
File "demo/demo.py", line 192, in <module>
build_model(sess, embeddings.shape[-1], batch_size)
File "demo/demo.py", line 64, in build_model
fake_images = model.get_generator(tf.concat(1, [c, z]))
File "~/tf-venv/lib/python2.7/site-packages/tensorflow/python/ops/array_ops.py", line 1043, in concat
dtype=dtypes.int32).get_shape(
File "~/tf-venv/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 676, in convert_to_tensor
as_ref=False)
File "~/tf-venv/lib/python2.7/site-packages/tensorflow/python/framework/ops.py", line 741, in internal_convert_to_tensor
ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
File "~/tf-venv/lib/python2.7/site-packages/tensorflow/python/framework/constant_op.py", line 113, in _constant_tensor_conversion_function
return constant(v, dtype=dtype, name=name)
File "~/tf-venv/lib/python2.7/site-packages/tensorflow/python/framework/constant_op.py", line 102, in constant
tensor_util.make_tensor_proto(value, dtype=dtype, shape=shape, verify_shape=verify_shape))
File "~/tf-venv/lib/python2.7/site-packages/tensorflow/python/framework/tensor_util.py", line 374, in make_tensor_proto
_AssertCompatible(values, dtype)
File "~/tf-venv/lib/python2.7/site-packages/tensorflow/python/framework/tensor_util.py", line 302, in _AssertCompatible
(dtype.name, repr(mismatch), type(mismatch).__name__))
TypeError: Expected int32, got <prettytensor.pretty_tensor_class.Layer object at 0x115b31990> of type 'Layer' instead.
I'm a bit confused: did installing prettytensor 0.6.2 install tensorflow 1.2.0-rc2
CPU and override the previous tensorflow 0.12.0
GPU install ? Is is still there ? Any tips on getting the demo working ?
I solve this problem by following https://github.com/google/prettytensor/issues/48 (Although there are still other problems)
I use "pip install prettytensor" but prettytensor aways make mistakes :
$ sh demo/flowers_demo.sh { doc_length : 201 filenames : "Data/flowers/example_captions.t7" queries : "Data/flowers/example_captions.txt" net_txt : "models/text_encoder/lm_sje_flowers_c10_hybrid_0.00070_1_10_trainvalids.txt_iter16400.t7" } Successfully load sentences from: Data/flowers/example_captions.t7 Total number of sentences: 5 num_embeddings: 5 (5, 1024) 2017-03-29 09:47:14.610481: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations. 2017-03-29 09:47:14.610495: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations. 2017-03-29 09:47:14.610500: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations. 2017-03-29 09:47:14.610505: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations. 2017-03-29 09:47:14.610509: W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use FMA instructions, but these are available on your machine and could speed up CPU computations. lr_imsize: 64 Traceback (most recent call last): File "demo/demo.py", line 192, in
build_model(sess, embeddings.shape[-1], batch_size)
File "demo/demo.py", line 64, in build_model
fake_images = model.get_generator(tf.concat(1, [c, z]))
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/ops/array_ops.py", line 1029, in concat
dtype=dtypes.int32).get_shape(
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/ops.py", line 639, in convert_to_tensor
as_ref=False)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/ops.py", line 704, in internal_convert_to_tensor
ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/constant_op.py", line 113, in _constant_tensor_conversion_function
return constant(v, dtype=dtype, name=name)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/constant_op.py", line 102, in constant
tensor_util.make_tensor_proto(value, dtype=dtype, shape=shape, verify_shape=verify_shape))
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/tensor_util.py", line 370, in make_tensor_proto
_AssertCompatible(values, dtype)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/tensor_util.py", line 302, in _AssertCompatible
(dtype.name, repr(mismatch), type(mismatch).name))
TypeError: Expected int32, got <prettytensor.pretty_tensor_class.Layer object at 0x7f74d41abd90> of type 'Layer' instead.