Commits

History for

Commits on Apr 19, 2016

Merge pull request #1902 from martinwicke/branch_119712558

martinwicke
authored and
caisq
committed

Commits on Apr 12, 2016

Fix bugs in the indexing operator with latest NumPy (#1895)

mrry
authored and
martinwicke
committed
Make diag_op_test faster (#1889)
martinwicke
committed

Commits on Apr 10, 2016

Surface control_flow_ops.while_loop to public.

yuanbyu
authored and
tensorflower-gardener
committed
This is another step to make TensorFlow more interactive and flexible to users. It allows a tensor produced by a run call to stay "in-place" so that a future run call can use it in-place. To achiev…

yuanbyu
authored and
tensorflower-gardener
committed

Commits on Apr 8, 2016

Added reduce_join op.

A. Unique TensorFlower
authored and
tensorflower-gardener
committed
Re-enable write-once, read-many semantics for TensorArray.

ebrevdo
authored and
tensorflower-gardener
committed
Add batch_diag and batch_diag_part ops.

ebrevdo
authored and
tensorflower-gardener
committed
Merge commit for internal changes
Vijay Vasudevan
committed
Add an attribute to {batch_}matrix_solve and {batch_}matrix_triangular_solve to allow solving A^* X = B instead of A X = B.

A. Unique TensorFlower
authored and
tensorflower-gardener
committed
Fixes `tf.reverse_sequence()` to work when input shape is unknown.

mrry
authored and
tensorflower-gardener
committed

Commits on Apr 7, 2016

Add gradients for {batch_}matrix_solve.

A. Unique TensorFlower
authored and
tensorflower-gardener
committed
Added is_variable_initialized(variable) function.

sherrym
authored and
tensorflower-gardener
committed
Pull request #1465. Implements reverse mode differentiation of the cholesky decomposition as an op in TensorFlow. (#1465)

alexggmatthews
authored and
Vijay Vasudevan
committed

Commits on Apr 5, 2016

Minor tweaks to TensorArray & gfile python API.

ebrevdo
authored and
tensorflower-gardener
committed
Use convert_to_tensor to do type/shape check instead of make_tensor_proto,

keveman
authored and
tensorflower-gardener
committed
Merge commit for internal changes
Vijay Vasudevan
committed

Commits on Apr 4, 2016

Change rnn benchmarks to use variables and initialize only once

ebrevdo
authored and
tensorflower-gardener
committed
Add igamma & igammac TF ops + gradients.

ebrevdo
authored and
tensorflower-gardener
committed
Merge commit for internal changes
Vijay Vasudevan
committed

Commits on Apr 3, 2016

Disable optimizations when running benchmarks.

keveman
authored and
tensorflower-gardener
committed

Commits on Mar 31, 2016

Merge commit for internal changes
Vijay Vasudevan
committed
Add type check for concat dim, add tests.

keveman
authored and
tensorflower-gardener
committed
Add gradient for SparseReorder

A. Unique TensorFlower
authored and
tensorflower-gardener
committed
TensorFlow: fix python3 with GPU test failures.
Vijay Vasudevan
committed

Commits on Mar 30, 2016

Allow non-rectangular strides for conv2d.

A. Unique TensorFlower
authored and
tensorflower-gardener
committed
Make embedding_size an explicit argument of embedding_* functions in seq2seq.

lukaszkaiser
authored and
tensorflower-gardener
committed
Reduce test size in sparse_add_op_test.

concretevitamin
authored and
tensorflower-gardener
committed
Add a subclass of Benchmark called TensorFlowBenchmark.

ebrevdo
authored and
tensorflower-gardener
committed
Better bounds checking for the segment reduction ops.

josh11b
authored and
tensorflower-gardener
committed