| Commit message (Collapse) | Author | Age |
|---|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This fix fixes several issues related to `go fmt` and `go lint`
based on https://goreportcard.com/report/github.com/tensorflow/tensorflow
There are several changes:
- `gofmt -s tensorflow/go/tensor.go`
- `gofmt -s tensorflow/go/example_inception_inference_test.go`
- `golint tensorflow/go/genop/internal/lib.go`
At the moment there are still quite a few golint and ineffassign warnings
in the current go code base. However, all of them are from `tensorflow/go/op/wrappers.go`
which is machine generated code.
This fix does not cover `tensorflow/go/op/wrappers.go`.
Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
* Improve py_func error handling.
Automatically translate some python errors into corresponding TF errors at runtime.
Change: 152156821
* Update interaction with libpng so that we use the public API instead of
knowledge of the internal libpng data structures.
Change: 152167754
* TensorBoard plugins now contain their own name/route prefix.
Change: 152167807
* Passes trainable flag to separable_conv2d biases.
Change: 152170239
* Saving resource variables with a caching device.
Change: 152171539
* Drop loss from estimator_spec.eval_metric_ops, as required by core Estimator.
Change: 152179924
* sample_stats.percentile DOCFIX.
Change: 152182295
* Added a memory optimizer to grappler.
Change: 152184170
* Change default behavior of the tf runs selector:
- If there are fewer than 41 runs, enable them all by default
- If there are 41 runs or more, disable them all by default
This is in response to user complaints that having it enable only the first ten runs by default was confusing, because it was not obvious to users that some runs had been disabled.
However, it still solves the initial user complaint that having very many runs simultaneously enabled would lag the UI.
I also changed the "toggle all runs" button to try to turn everything off before turning everything on.
Also, I improved the logic for detecting when the runs selection is back in the default state, so that we can avoid generating long URI strings wherever possible.
Change: 152188948
* Autogenerated Change: Change TensorBoard TAG to 52
Change: 152189000
* Remove warning that only happening with config cuda.
Change: 152189205
* Make resource variable shared name consistent with non-resource variables.
Remove colocation constraint from resource variable cached value with the
variable itself.
Change: 152192203
* Add a way to specify the optimization order; refactor and add constant folding to meta optimizer.
Change: 152193646
* Backport fixes and improvements from external Keras.
Change: 152198296
* Merge changes from github.
Change: 152200430
* Go: Update generated wrapper functions for TensorFlow ops.
Change: 152200754
* Update ops-related pbtxt files.
Change: 152203174
* Make ImportGraphDef() work with functions.
In addition to modify graph_constructor.cc, this patch adds some other
functionality to enable importing fucntions:
* Ability to add FunctionDefLibraries to Graphs and
FunctionLibraryDefinitions (in addition to existing functions)
* FunctionDefsEqual() utility function
Change: 152205258
* Expand contrib test to more than just test targets.
Change: 152206822
* Preserve graph version during optimization
Change: 152213262
* Exclude enter and exit nodes from shape refiner's constant folding.
Change: 152213637
* Allow reshape_mover and algebraic_simplifier to make multiple mutations, by avoiding the short-circuit
std::any_of.
Change: 152232810
* fixing workspace.bzl
* workspace.bzl further fixes
* fixing tensorflow.bzl merge conflicts
* fixing typo in dnn.h
* fixing bad merge for dnn.h
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
| |
And use this support to simplify the Inception example as it can use the
DecodeJpeg op.
Also fixed a bug in generated op functions - A TensorFlow "int"
is a Go "int64".
Another step in #10
Change: 139809489
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Errors during graph construction are held in the Scope.
This makes the op construction code mode compact and nestable
as errors do not _need_ to be handled on every op addition.
To help ensure that the client does not miss the error completely,
the Scope is treated as a "builder" of Graphs, and once the Graph
is extracted from the Scope (using Scope.Finalize), the Scope is
rendered useless.
To help tracing failures to the op, the error stores the stacktrace
pointing to when the error occurred, which will help in identifying
the precise operation that failed.
To use scopes to enhance existing Graphs, the idea is to add
a:
NewScopeWithGraph(*tf.Graph) *Scope
function, but that is not included in this change.
Change: 137423318
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This commit makes the inception inference example use the graph construction
API, and in particular the generated functions for TensorFlow ops, to
normalize an image before sending for inference through the Inception model.
Besides demonstrating the use of the graph construction API, this also
makes the example usable on arbitrarily sized images. (Prior to this change,
the input image had to be precisely 224x224 pixels).
A slight variation of this would have involved the use of the ReadFile and
DecodeJpeg ops, however those are not usable in Go at the moment since
string valued tensors aren't supported yet.
Change: 137099275
|
| |
|
|
|
|
|
|
|
|
|
| |
This will make it easier to simply copy-paste the example
code shown on
https://godoc.org/github.com/tensorflow/tensorflow/tensorflow/go
and run it as a standalone program.
Without this change, the reader needs to manually download the model,
unzip it, place the files in a specific location on the filesystem etc.
Change: 136200192
|
| |
|
|
|
|
|
|
|
| |
godoc.org translates full-file examples into a "main" package. This change
makes the example so that the generated "main" code can be copy-pasted and thus
used to try out a standalone binary.
Another step towards #10
Change: 134709944
|
|
|
Add an example (that will appear in Go doc) for a real use of the Go TensorFlow
APIs - using a pre-defined image recognition model for inference.
While at it a couple of minor tweaks:
- NewSession now accepts 'nil' for options as the documentation says it does
- Convenience accessors for the Outputs of an Operation
- Ability to extract (possibly partial) shapes from an Output
Another step towards #10
Change: 134560938
|
Yong Tang
Aditya Dhulipala
Rohan Jain
zhengjiajin
Kai Sasaki
Asim Shankar