- 25 6月, 2016 24 次提交
-
-
由 A. Unique TensorFlower 提交于
Change: 125835079
-
由 A. Unique TensorFlower 提交于
Change: 125834391
-
由 Martin Wicke 提交于
Change: 125830950
-
由 Martin Wicke 提交于
Change: 125830566
-
由 Jianmin Chen 提交于
Change: 125829994
-
由 Dan Mané 提交于
- dist/index.html now uses only relative, not absolute, imports - tf-tensorboard now contains all the demo logic; tf-tensorboard-demo was removed. The vulcanization logic was simplified. - Fixed serialize_tensorboard, and cleaned it up a bit. Change: 125827649
-
由 Mustafa Ispir 提交于
Change: 125827302
-
由 A. Unique TensorFlower 提交于
Change: 125821416
-
由 Zongheng Yang 提交于
Change: 125816703
-
由 A. Unique TensorFlower 提交于
Fix linker errors by adding a dependence on constant_op to targets using fill_functor. This is needed because the templates in fill_functor.h are instantiated in constant_op.cc. Change: 125816082
-
由 A. Unique TensorFlower 提交于
Change: 125813148
-
由 Eugene Brevdo 提交于
This will allow error checking for subclasses of MeanValue, SampleValue, etc. For example, for ADVISampleValue, we can ensure that no inputs are descended from other DistributionTensors with intermediate tensor calculations (because these require a special Jacobian term added to the score function, which can only be done by TransformedDistribution objects). Change: 125812475
-
由 A. Unique TensorFlower 提交于
the DNNLinearCombinedEstimators work properly. Change: 125811654
-
由 A. Unique TensorFlower 提交于
RealValuedColumn.to_weighted_sum didn't use variable scopes, so it was impossible to reuse the values in a graph. This is inconsistent with the rest of layers. It also manually performed a matrix multiply and this was replaced with a more standard formulation. Change: 125808727
-
由 Mustafa Ispir 提交于
Change: 125803613
-
由 Renato Utsch 提交于
Change: 125799092
-
由 A. Unique TensorFlower 提交于
Change: 125795153
-
由 A. Unique TensorFlower 提交于
Change: 125794105
-
由 Sherry Moore 提交于
Added Python interfaces to reset resource containers. pywrap_tensorflow.TF_Reset(target, [containers]) will release resources in all listed containers, while pywrap_tensorflow.TF_Reset(target) will release resources in all the containers. In particular, Variables cached on the devices will be cleared by this call. Change: 125790643
-
由 A. Unique TensorFlower 提交于
Change: 125790131
-
由 A. Unique TensorFlower 提交于
Change: 125789944
-
由 Andrew Selle 提交于
StridedSliceGrad op implements the gradient of StridedSlice. Also implement python benchmark for StridedSlice and simple Slice. Fix bugs in special case optimizations in StridedSlice. (Toward resolving bug #206) Change: 125789921
-
由 Ben Lee 提交于
- Prevent unnecessary tensorflow.Example copies when this method is used in serving. Change: 125788885
-
由 Alexey Surkov 提交于
- adds support of paths pointing to the root of a bucket, e.g gs://bucket - adds pagination support for GetChildren - makes a more optimal HTTP request in GetChildren Change: 125785863
-
- 24 6月, 2016 16 次提交
-
-
由 A. Unique TensorFlower 提交于
Change: 125783676
-
由 A. Unique TensorFlower 提交于
Change: 125783541
-
由 A. Unique TensorFlower 提交于
Change: 125783336
-
由 A. Unique TensorFlower 提交于
Change: 125783060
-
由 A. Unique TensorFlower 提交于
Change: 125780396
-
由 A. Unique TensorFlower 提交于
Fix a flaky test by computing half-precision gamma samples with float precision intermediate calculations. Change: 125776264
-
由 Pete Warden 提交于
Change: 125761757
-
由 A. Unique TensorFlower 提交于
Change: 125752088
-
由 A. Unique TensorFlower 提交于
Change: 125738602
-
由 A. Unique TensorFlower 提交于
Change: 125737177
-
由 A. Unique TensorFlower 提交于
Change: 125735275
-
由 Benoit Steiner 提交于
Also updated the weight initialization to ensure that the initial weights are likely to be greater than the smaller possible value representable using a 16 bit float (~0.00006). Change: 125733195
-
由 A. Unique TensorFlower 提交于
Change: 125729946
-
由 A. Unique TensorFlower 提交于
Change: 125729599
-
由 A. Unique TensorFlower 提交于
Change: 125729413
-
由 Eugene Brevdo 提交于
One TODO for DirichletMultinomial. Change: 125729016
-