Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | Fix "from tensorflow._api.v1 import *". | Michael Case | 2018-09-21 |
| | | | | PiperOrigin-RevId: 214055060 | ||
* | Remove duplicate `from six import text_type` in upload_test_benchmarks.py | Yong Tang | 2018-05-17 |
| | | | | | | | The `from six import text_type` was imported twice in upload_test_benchmarks.py, this fix removes the duplication. Signed-off-by: Yong Tang <yong.tang.github@outlook.com> | ||
* | Remove all_opensource_files. It's not needed any more. | Martin Wicke | 2018-03-28 |
| | | | | PiperOrigin-RevId: 190878279 | ||
* | Merge changes from github. | Benoit Steiner | 2018-03-22 |
| | | | | PiperOrigin-RevId: 190161440 | ||
* | Update file due to changes in Bazel (PACKAGE_NAME is deprecated) | A. Unique TensorFlower | 2018-03-22 |
| | | | | PiperOrigin-RevId: 190051589 | ||
* | Merge changes from github. | Jacques Pienaar | 2018-03-21 |
| | | | | PiperOrigin-RevId: 189945839 | ||
* | Automated g4 rollback of changelist 189231636 | A. Unique TensorFlower | 2018-03-15 |
| | | | | PiperOrigin-RevId: 189258641 | ||
* | Merge changes from github. | Jacques Pienaar | 2018-03-15 |
| | | | | PiperOrigin-RevId: 189231636 | ||
* | Internal change. | Anna R | 2018-01-31 |
| | | | | PiperOrigin-RevId: 184047860 | ||
* | De-bazel filename_test. Part of the effort to remove all_opensource_files. | Yifei Feng | 2018-01-31 |
| | | | | PiperOrigin-RevId: 184021942 | ||
* | Merge changes from github. | Jianwei Xie | 2018-01-24 |
| | | | | PiperOrigin-RevId: 183100142 | ||
* | Fix some build incompatibilities with new versions of Bazel | Austin Anderson | 2017-12-05 |
| | | | | | | See #15137. PiperOrigin-RevId: 178037461 | ||
* | Internal change. | Anna R | 2017-09-30 |
| | | | | PiperOrigin-RevId: 170604029 | ||
* | Adding a copy binary script to GitHub to change python versions for whl files. | Amit Patankar | 2017-09-19 |
| | | | | PiperOrigin-RevId: 169264602 | ||
* | Fix build errors caused by #12502 | A. Unique TensorFlower | 2017-09-19 |
| | | | | PiperOrigin-RevId: 169264011 | ||
* | Adding a copy binary script to GitHub to change python versions for whl files. | Amit Patankar | 2017-09-19 |
| | | | | PiperOrigin-RevId: 169264602 | ||
* | PiperOrigin-RevId: 166184401 | A. Unique TensorFlower | 2017-08-23 |
| | |||
* | Converting update_version script to python. | A. Unique TensorFlower | 2017-08-03 |
| | | | | PiperOrigin-RevId: 164192125 | ||
* | Updating the regex match to catch p100 for run and gather logs." | A. Unique TensorFlower | 2017-07-20 |
| | | | | PiperOrigin-RevId: 162681705 | ||
* | BUILD cleanup in tensorflow/tools/... | A. Unique TensorFlower | 2017-06-23 |
| | | | | PiperOrigin-RevId: 160018623 | ||
* | Internal change. | Anna R | 2017-06-22 |
| | | | | PiperOrigin-RevId: 159876942 | ||
* | Internal change. | Anna R | 2017-06-21 |
| | | | | PiperOrigin-RevId: 159769520 | ||
* | Internal change. | Anna R | 2017-06-20 |
| | | | | PiperOrigin-RevId: 159642257 | ||
* | Fix #10371 | Asim Shankar | 2017-06-05 |
| | | | | | | | cpuinfo.get_cpu_info() doesn't seem to include the l2_cache_size key on some architectures. PiperOrigin-RevId: 158021008 | ||
* | Use test_adjusted_name when making the mangled_test_name in | A. Unique TensorFlower | 2017-05-31 |
| | | | | | | | run_and_gather_logs_lib.py, to avoid duplicate file names when the same test is run on multiple GPUs. PiperOrigin-RevId: 157630193 | ||
* | This is running all the benchmark tests on k20, k40, and k80. We had to allow | A. Unique TensorFlower | 2017-05-11 |
| | | | | | | all cpu-platforms for k80. PiperOrigin-RevId: 155777715 | ||
* | Merge changes from github. | A. Unique TensorFlower | 2017-04-04 |
| | | | | Change: 152200430 | ||
* | Splitting scalar strict test into an internal and external version. | A. Unique TensorFlower | 2017-04-03 |
| | | | | Change: 152064895 | ||
* | Add BenchmarkType to TestResults proto and fix logging of C++ microbenchmarks. | Eugene Brevdo | 2017-03-02 |
| | | | | | C++ microbenchmarks results are now normalized by iters (like python benchmarks). Change: 149045367 | ||
* | Use 'name' instead of 'test_name' when forming benchmark output file name. | A. Unique TensorFlower | 2017-02-28 |
| | | | | Change: 148834478 | ||
* | Replace pip testing script with bazel. | Gunhan Gulsoy | 2017-02-13 |
| | | | | Change: 147423640 | ||
* | Tag tf_cc_logged_benchmarks as manual. They should only run with benchmark | A. Unique TensorFlower | 2017-02-09 |
| | | | | | build. Change: 147100980 | ||
* | A few changes to get tf_cc_logged_benchmark's running under tensorflow/. ↵ | A. Unique TensorFlower | 2017-02-04 |
| | | | | | | Pass in benchmark argument to tf_cc_logged_benchmark. Also, add _gpu suffix to benchmarks. Change: 146552100 | ||
* | Do not log test results using the text format, the JSON format is sufficient. | A. Unique TensorFlower | 2017-02-03 |
| | | | | Change: 146539043 | ||
* | Split test_log_output flag into two flags: test_log_output_dir and | A. Unique TensorFlower | 2017-02-01 |
| | | | | | | test_log_output_filename. This lets us create test_log_output_filename automatically if it is not passed in. Change: 146320874 | ||
* | Updated run_and_gather_logs.py to use argparse. | A. Unique TensorFlower | 2017-01-31 |
| | | | | Change: 146194766 | ||
* | Use glob.glob instead of gfile.Glob in system_info_lib.py. | A. Unique TensorFlower | 2017-01-27 |
| | | | | Change: 145768535 | ||
* | Add a benchmark data uploader script to store test results in gcloud datastore. | A. Unique TensorFlower | 2017-01-26 |
| | | | | | | | | | | | | | | The uploader is typically run periodically as a cron job. It processes data files found in a specified directory where the run_and_gather_logs.py script stores the benchmark test results. After processing a file, it moves it to a specified archive directory for safe-keeping. File-locking (flock) is used to allow multiple instances of the uploader to run concurrently if needed, splitting the work among them. Two datastore composite indexes are created to allow fast retrieval of benchmark timing data (for lookup and field (column) projection). Change: 145729637 | ||
* | Adding support for 'b' in mode for FileIO. | Rohan Jain | 2017-01-25 |
| | | | | Change: 145590579 | ||
* | Remove so many more hourglass imports | Justine Tunney | 2016-12-29 |
| | | | | Change: 143230429 | ||
* | Add missing numpy and six deps | Justine Tunney | 2016-12-28 |
| | | | | Change: 143131671 | ||
* | Remove hourglass imports from kernel_tests | Justine Tunney | 2016-12-14 |
| | | | | Change: 142080137 | ||
* | Pass benchmark name (path:target_name) to TestResults proto. | A. Unique TensorFlower | 2016-09-22 |
| | | | | Change: 134000963 | ||
* | Redirecting gfile to file_io. | Rohan Jain | 2016-09-08 |
| | | | | | Also, changed the return type of FileIO.readline(), FileIO.readlines(), list_directories() and walk() to be strings rather than bytes. Change: 132634441 | ||
* | Give python/kernel_tests/ directory its own BUILD file. | A. Unique TensorFlower | 2016-06-14 |
| | | | | Change: 124868436 | ||
* | Update copyright for 3p/tf/tools. | A. Unique TensorFlower | 2016-06-02 |
| | | | | Change: 123889091 | ||
* | Reenable check_futures_test | Geoffrey Irving | 2016-04-29 |
| | | | | | | Also fix a bunch of files that ended up without future imports while it was off. Change: 121161570 | ||
* | Change tensorflow benchmark code to export results as serialized proto instead | A. Unique TensorFlower | 2016-04-22 |
| | | | | | | of text proto. This simplifies ues of LITE_RUNTIME for mobile, where text functions are not available natively. Change: 120545368 | ||
* | Merge changes from github. | Vijay Vasudevan | 2016-03-29 |
| | | | | Change: 118532471 | ||
* | Opensource all_opensource_files | Geoffrey Irving | 2016-03-25 |
| | | | | Change: 118243578 |