tensorflow: 1.10 build fails with "No module named 'keras_applications'"
System information
- Have I written custom code (as opposed to using a stock example script provided in TensorFlow): no
- OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Linux Fedora 28
- Mobile device (e.g. iPhone 8, Pixel 2, Samsung Galaxy) if the issue happens on mobile device: N/A
- TensorFlow installed from (source or binary): source
- TensorFlow version (use command below): 1.10
- Python version: 3.6
- Bazel version (if compiling from source): 0.16
- GCC/Compiler version (if compiling from source): 7.3
- CUDA/cuDNN version: 9.2
- GPU model and memory: Quadro M2200 4G
- Exact command to reproduce: bazel build --config=opt --config=cuda //tensorflow/tools/pip_package:build_pip_package
Hi, a recent checkout of master (~ 2 hours ago) fails (for me) with
ERROR: /home/key/code/tensorflow/tensorflow/BUILD:584:1: Executing genrule //tensorflow:tensorflow_python_api_gen failed (Exit 1)
/home/key/anaconda3/lib/python3.6/site-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
from ._conv import register_converters as _register_converters
Traceback (most recent call last):
File "/home/key/.cache/bazel/_bazel_key/8de93c72df6c3fec3e289f10fadff72b/execroot/org_tensorflow/bazel-out/host/bin/tensorflow/create_tensorflow.python_api.runfiles/org_tensorflow/tensorflow/python/tools/api/generator/create_python_api.py", line 27, in <module>
from tensorflow.python.tools.api.generator import doc_srcs
File "/home/key/.cache/bazel/_bazel_key/8de93c72df6c3fec3e289f10fadff72b/execroot/org_tensorflow/bazel-out/host/bin/tensorflow/create_tensorflow.python_api.runfiles/org_tensorflow/tensorflow/python/__init__.py", line 81, in <module>
from tensorflow.python import keras
File "/home/key/.cache/bazel/_bazel_key/8de93c72df6c3fec3e289f10fadff72b/execroot/org_tensorflow/bazel-out/host/bin/tensorflow/create_tensorflow.python_api.runfiles/org_tensorflow/tensorflow/python/keras/__init__.py", line 25, in <module>
from tensorflow.python.keras import applications
File "/home/key/.cache/bazel/_bazel_key/8de93c72df6c3fec3e289f10fadff72b/execroot/org_tensorflow/bazel-out/host/bin/tensorflow/create_tensorflow.python_api.runfiles/org_tensorflow/tensorflow/python/keras/applications/__init__.py", line 21, in <module>
import keras_applications
ModuleNotFoundError: No module named 'keras_applications'
Target //tensorflow/tools/pip_package:build_pip_package failed to build
Use --verbose_failures to see the command lines of failed build steps.
INFO: Elapsed time: 3855.754s, Critical Path: 156.93s
INFO: 8932 processes: 8932 local.
FAILED: Build did NOT complete successfully
Could you please advise?
About this issue
- Original URL
- State: closed
- Created 6 years ago
- Reactions: 14
- Comments: 36 (6 by maintainers)
Commits related to this issue
- Moving on with build, missing two packages not in python path https://github.com/tensorflow/tensorflow/issues/21518 — committed to mrodozov/cmsdist by mrodozov 6 years ago
I believe the following will fix this issue:
@yifeif any additional info?
You might also need
pip install h5py==2.8.0we should update our installation from sources documentation with this information.
I can confirm that installing the three packages resolves the issue as per this link
I had the same problem during the installation in virtualenv. It worked once I installed
pip install keras_applications==1.0.4 --no-deps pip install keras_preprocessing==1.0.2 --no-deps pip install h5py==2.8.0
outside of the virtual environment.
In my case it was the problem that there are two different versions of packages with
-and_after keras sokeras_applicationsandkeras-applications. I was looking at one and TF was requiring the other one. The same goes forkeras_preprocessingYou will either need to create the virtualenv with
--system-site-packages, or runpip install keras-applications(and other pip install commands) after you activate the virtualenv.works great, thank you!!