It is 50MiB, which doesn't belong in git.

It turns out there is a second copy of this graph in the android assets
already too.  That explains why the repo is so large.

(Since we have not accepted external commits yet, we may
 consider rewriting the history to purge these files if
 the size of the repo is a concern).

Base CL: 108016510

a few other changes.

Changes:
- Some improvements to convolution by using 32-bit indices by
  @benoitsteiner. Not all calls converted yet.  Also some
  improvements to pooling as well by @benoitsteiner.

- Improvements to sparse matmul CPU implementation by Ashish

- Some fixes to warnings by @vrv

- Doc fixes to padding by @Yangqing

- Some improvements to Tensor wrappers by Eider

- Speed up of matrix inverse on CPU by Rasmus

- Add an example of doing image inference from a pre-trained model
  by @petewarden.

- fixed formula in mnist example by nodir

- Updates to event accumulator by Cassandra

- Slight changes to tensor c api by @mrry

- Handling of strings in listdiff by Phil

- Fix negative fraction-of-queue-full stats by Frank

- Type-checking improvement to importer by Yaroslav

- logdir recursive search for Tensorboard by @danmane

- Session.run() checks for empty graph by Manoj

Base CL: 108013706

error handling, updates to website.

Changes:
- Removes redundant reshape from image models by @mrry
- Default TensorBoard to localhost by @danmane
- Reformatting of tensorflow/core by @josh11b
- Make tutorials backwards compatible to 0.5.0 by @girving
- Improve print documentation (md files not updated).
- Add proper scrolling to sitemap by @martinwicke

Base CL: 107956254

Changes:
- Bug fix to input.py by @dave-andersen with tests.

- Some fixes to names in the whitepaper

- Include cfloat for FLT_MAX by @benoitsteiner

- Fix broken link in mnist tutorials by @mrry

- Typos and fixes to documentation by @Sohl-Dickstein, @ebrevdo, @vrv,
  Yaroslav, @martinwicke

- Fixed unnecessary check around a delete by @girving

- More compatibility tests by @josh11b

- Misc other typos and fixes by Googlers.

Base CL: 107944070

Changes:
- Fix softmax formula in word2vec to remove an extra exp()
  by @gouwsmeister

- Python3 fixes to remove basestring / support for unicode by @mrry

- Remove some comments by Josh

- Specify exact versions of bower dependencies for TensorBoard by
  @danmane.

Base CL: 107742361

Changes:
- Ptrdiff -> DenseIndex change by @jiayq

- Fix to scoping the logging in logging.py by @dga

- Improvement to Conv2DBackpropFilter on CPU by Andy

- Remove lookup table wrappers for the time being (wasn't in our
  public API yet) by Yukata

- Add a check similar to numpy to make sure the user isn't in the
  tensorflow src directory by @vrv

- More changes for  python 3 compat by @girving

- Make dropout preserve shape info from input (@mrry)

- Significant speed improvements by @zheng-xq to BFC allocator to bring
  on par (CPU overhead-wise) to the region allocator.  Make BFC
  allocator the default now that it's working well for a variety
  of models.

- Fix a bunch of typos reported by users (@vrv)

- Enable concat for bfloat16 on GPU by Ashish.

Base CL: 107733123

Changes:
- Updates to op documentation and index by Josh

- More changes to BUILD files for python 3 support by @girving

- Fix to Eigen to use DenseIndex everywhere by @jiayq

- Enable configuration for cuda compute capability by @zheng-xq,
  including updates to docs.

- Route aggregation method through optimizer by schuster

- Updates to install instructions for bazel 0.1.1.

Base CL: 107702099

Changes:

- futurize --stage2 changes for Python 3 compatibility by @girving.

- Small updates to documentation by @vrv, schuster and others

- Account for failure of std::thread::hardware_concurrency by @ebrevdo.

- More changes for backwards-compatibility tests by Josh

- Updates to python op doc generation by Josh

- Added support for using the best-fit allocator via ConfigProto by @vrv.

- Rename LocalSession to DirectSession, since local was a bad name for
  it.

- Enable tf.nn.moments() to work with tensors of unknown shape by @mrry.
  GITHUB_ISSUE: 139

- Changes for Android build by Andrew.

Base CL: 107645181

Changes:
- Updates to README / os_setup for common installation issues
  by ebrevdo, vrv, wicke, anelia, dga

- Add a test for backwards compatibility checks by Josh

- Simplifications to tiling/padding of scalars by bsteiner

- Improve listing of devices when no devices are known by Matthieu

Base CL: 107607137

Base CL: 107547091

Base CL: 107543097

Changes:

- Fix typos across several files contributed by Erik Erwitt,
  and Michael R. Berstein

- Fix bug in translate example (fr->en typo) by schuster

- Updates to some documentation (mcoram,shlens,vrv,joshl)

- Fix to Android camera demo app window size detection (andrewharp)

- Fix to support lookup table of high rank tensors (yleon)

- Fix invalid op names for parse_example (dga)

Base CL: 107531031

Changes:

- Some changes to make our ability to handle external contributions
  simpler (e.g., adding some markers, adding empty __init__.py files).

- Fixing documentation of SummaryWriter.add_summary().

- Some input validation changes for queues/barriers.

- Fixing the ptb tutorial (thanks to @kentonl for reporting),
  fixes, github issue 52.

- Some documentation suggestions for dealing with a few install
  problems.

- Speed improvements to conv2d gradient kernels on CPU.

- More documentation fixes for the website.  Thanks to @makky3939 and
  @keonkim for reports.

- Changes / fixes to the Docker files.

- Changes build_pip_package to not create an sdist.

- Adds tensorboard example script.

Base CL: 107445267

bug fixes -- reindents to 2 for some of the files to match our internal
requirements.

Thanks to Martin Andrews for the basic_usage.md suggested fix via
Gerrit.

Base CL: 107394029

and add tentative Dockerfile for GPU.  Change install instructions
to also suggest upgrading six for Mac.

Base CL: 107379398

Changes:
- Lot of updates to docs.

Base CL: 107357203

Changes:
- Doc type fixes.

Base CL: 107356208

Changes:
- Update to OS X install instructions.

Base CL: 107355822

Changes:
- Correct number of CPUs reported on mac.
- Doc updates.

Base CL: 107355041

Changes:
Fix anchor for github.

Base CL: 107352236

Changes:
- Updates to installation instructions.

Base CL: 107352130

Changes:
- Documentation updates.
- Relax numpy requirement to 1.9.2

Base CL: 107349632

Changes:
- Docuementation changes.
- Update URL for protobuf submodule.

Base CL: 107345722

Changes:
- Documentation updates.
- Specify numpy version in required packages.

Base CL: 107344010

Changes:
- Documentation changes: adding some examples
  for adding_an_op, fixes to some of the markdown,
  updates to docstrings, etc.

- Remove Dockerfile for now -- still undergoing
  changes.

Base CL: 107341050

Changes:
- Update whitepaper
- Fix some names in docs.

Base CL: 107313143

Changes:
- More documentation edits, fixes to anchors,
  fixes to mathjax, new images, etc.
- Add rnn models to pip install package.

Base CL: 107312343

Changes:
- Updates to Documentation, README.md, installation
  instructions, anchor links, etc.

- Adds Readme for embedding directory.

Base CL: 107308461

Changes:

- Update a lot of documentation, installation instructions,
  requirements, etc.

- Add RNN models directory for recurrent neural network
  examples to go along with the tutorials.

Base CL: 107290480

Changes:
- Updates to installation instructions.
- Updates to documentation.
- Minor modifications and tests for word2vec.

Base CL: 107284192

TensorFlow is an open source software library for numerical computation
using data flow graphs.

Base CL: 107276108