Commit message (Expand) | Author | Age | |
---|---|---|---|
* | Update SCALED mode to use the full quantized range of -128..127 when possible. | 2018-05-14 | |
* | Add 'SCALED' mode to QuantizeV2 and Dequantize, to match QuantizeAndDequantiz... | 2017-08-23 | |
* | Fix dequantize_op benchmark. The requirement is that the ranges come in as | 2017-08-17 | |
* | Preallocate vector storage when the ultimate vector size is known in advance | 2017-06-01 | |
* | Speed up dequantize for combined, by using a simple loop and avoiding the cas... | 2017-04-17 |