| 1a1fb0b556cac150b10a3bca91857bb25caae073 |
|
08-Jan-2018 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Merged commit includes the following changes:
181203530 by gunan:
Adding a CLIF library for generic_tree_model.proto
--
181201655 by annarev:
Internal change.
--
181197493 by akshayka:
Make benchmark names in eager's resnet50_test unique.
--
181196715 by A. Unique TensorFlower:
Correct the typo of "ny.arrange" to "ny.arange".
--
181194884 by A. Unique TensorFlower:
Update ops-related pbtxt files.
--
181193667 by power:
Avoid reading through the inputbuffer for large reads.
--
181192456 by A. Unique TensorFlower:
Adding a CLIF library for generic_tree_model.proto
--
PiperOrigin-RevId: 181203530
/external/tensorflow/tensorflow/go/op/generate.go
|
| ccbc8991db3943ef984405881a1c917c530f902f |
|
05-Apr-2017 |
A. Unique TensorFlower <gardener@tensorflow.org> |
Merge changes from github.
Change: 152200430
/external/tensorflow/tensorflow/go/op/generate.go
|
| 93a975e114ee1c35f01ed3bdd47170e6f7129014 |
|
18-Feb-2017 |
Vijay Vasudevan <vrv@google.com> |
Merge changes from github.
Change: 147897309
/external/tensorflow/tensorflow/go/op/generate.go
|
| 23265d67fe4daa5761b6a080139ef700f1d09d1d |
|
01-Feb-2017 |
Asim Shankar <ashankar@google.com> |
Go: Tweaks to the op code generator
- Create the directory if it doesn't exist
- Output formatted code
Change: 146286990
/external/tensorflow/tensorflow/go/op/generate.go
|
| 6586a66b1eac161221155b8bf1c2a75677623bc3 |
|
11-Oct-2016 |
Asim Shankar <ashankar@google.com> |
go: Scaffolding for a package containing wrapper functions for TensorFlow ops.
This includes:
(1) A 'genop' command-line tool that generates Go functions for each TensorFlow op registered in the address space of the process.
(2) An 'op' package that will host these generated functions and some hand-crafted
functions (like Const).
At this time, none of the generated files are being committed.
As a result, the installation instructions in README.md have been updated to
reflect how 'go generate' should be used to setup the 'op' package.
Note that even after this change, packages that clients are expected to
use ('tensorflow' and 'op') do not depend on protocol buffers (github.com/golang/protobuf).
However, the stand-alone command-line tool 'genop' does.
This change is focused on the scaffolding and introduces a dummy code generator.
The real implementation of the code generator will be done in a follow up change
as I wanted to separate these "mechanics" from the "implementation details".
Yet another step for #10
Change: 135735176
/external/tensorflow/tensorflow/go/op/generate.go
|