Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | Fix tf.quantize_and_dequantize_(v2|v3) to be consistent with the doc. | Jingyue Wu | 2018-08-03 |
| | | | | | | | | I choose round-half-to-even, which matches cudnnConvolutionBiasActivationForward and cudnnTransformTensor. I can add an attribute for the rounding mode in the future if necessary. PiperOrigin-RevId: 207316630 | ||
* | Fixes a bug in the quantize_and_dequantize_op kernel with getting the min ↵ | A. Unique TensorFlower | 2018-06-29 |
| | | | | | | | | and max range, when the op is on the GPU but the range tensor is on the host. PiperOrigin-RevId: 202748603 | ||
* | Make the quantize_and_dequantize op use the full quantized range when possible. | A. Unique TensorFlower | 2018-05-29 |
| | | | | PiperOrigin-RevId: 198450816 | ||
* | Automated g4 rollback of changelist 197487461 | A. Unique TensorFlower | 2018-05-22 |
| | | | | PiperOrigin-RevId: 197523867 | ||
* | Make the quantize_and_dequantize op use the full quantized range when possible. | A. Unique TensorFlower | 2018-05-21 |
| | | | | PiperOrigin-RevId: 197487461 | ||
* | Automated g4 rollback of changelist 197226707 | A. Unique TensorFlower | 2018-05-21 |
| | | | | PiperOrigin-RevId: 197415745 | ||
* | Make the quantize_and_dequantize op use the full quantized range when possible. | A. Unique TensorFlower | 2018-05-18 |
| | | | | PiperOrigin-RevId: 197226707 | ||
* | Cleanup: Ran clang-format on all *.{cc,h} files in tensorflow/core/kernels. | A. Unique TensorFlower | 2018-01-26 |
| | | | | PiperOrigin-RevId: 183423961 | ||
* | Make input_min and input_max Inputs of QuantizeAndDequantize rather than ↵ | Suharsh Sivakumar | 2017-02-15 |
| | | | | | | attributes. Change: 147653779 | ||
* | Automated rollback of change 147124625 | Asim Shankar | 2017-02-10 |
| | | | | Change: 147161509 | ||
* | Make input_min and input_max Inputs of QuantizeAndDequantize rather than ↵ | Suharsh Sivakumar | 2017-02-09 |
| | | | | | | attributes. Change: 147124625 | ||
* | Fix inclusion guards of quantize_and_dequantize_op. | Suharsh Sivakumar | 2017-02-02 |
| | | | | Change: 146441782 | ||
* | Fix QuantizeAndDequantize GPU segfault. | Zongheng Yang | 2016-10-10 |
| | | | | | | | | | The issue is that the previous input_min() = input_min_ lines are dereferencing GPU memory pointers. Also added: simple microbenchmarks for CPU/GPU, as naive sanity checks. Change: 135736169 | ||
* | Update copyright for 3p/tf/core. | A. Unique TensorFlower | 2016-06-02 |
| | | | | Change: 123900938 | ||
* | Move quantize_and_dequantize to core for the quantize training rewriting. | Jianmin Chen | 2016-05-31 |
Change: 123681962 |