- 23 4月, 2021 1 次提交
-
-
由 Scott Zhu 提交于
PiperOrigin-RevId: 369893606
-
- 23 3月, 2021 1 次提交
-
-
由 Yeqing Li 提交于
PiperOrigin-RevId: 364483380
-
- 17 3月, 2020 1 次提交
-
-
由 Allen Wang 提交于
PiperOrigin-RevId: 301237793
-
- 30 8月, 2019 1 次提交
-
-
由 Reed Wanderman-Milne 提交于
The old infer_float32_policies policy will be removed from TensorFlow soon. To test convergence, I ran the Resnet50KerasAccuracy.benchmark_8_gpu_fp16 benchmark. I got an accuracy of 0.76037 and an exp_per_second of 6908. PiperOrigin-RevId: 266191126
-
- 23 8月, 2019 1 次提交
-
-
由 Hongkun Yu 提交于
PiperOrigin-RevId: 264895439
-
- 20 8月, 2019 1 次提交
-
-
由 Haoyu Zhang 提交于
Moving Keras ResNet models to `official/vision/image_classification` and benchmarks to `official/benchmark`. PiperOrigin-RevId: 264268533
-
- 11 5月, 2019 2 次提交
-
-
由 Haoyu Zhang 提交于
* Fix trivial model to work properly with fp16 * Add comment on manual casting
-
由 Haoyu Zhang 提交于
Previously we had one dense layer in trivial model. The weight was [224*224*3, num_classes]. Using two dense layers, the weights are [224*224*3, 1] and [1, num_classes].
-
- 28 3月, 2019 1 次提交
-
-
由 Haoyu Zhang 提交于
-
- 03 4月, 2018 1 次提交
-
-
由 Zhichao Lu 提交于
PiperOrigin-RevId: 191112757
-
- 22 9月, 2017 1 次提交
-
-
由 Neal Wu 提交于
-
- 11 3月, 2016 1 次提交
-
-
由 Manjunath Kudlur 提交于
Updated README
-
- 10 3月, 2016 1 次提交
-
-
由 Martin Wicke 提交于
-