Testing.
Modules
experimental
module: Public API for tf.test.experimental namespace.
Classes
class Benchmark
: Abstract class that provides helpers for TensorFlow benchmarks.
class TestCase
: Base class for tests that need to test TensorFlow.
Functions
assert_equal_graph_def(...)
: Asserts that two GraphDef
s are (mostly) the same.
benchmark_config(...)
: Returns a tf.compat.v1.ConfigProto for disabling the dependency optimizer.
compute_gradient(...)
: Computes the theoretical and numeric Jacobian of f
.
create_local_cluster(...)
: Create and start local servers and return the associated Server
objects.
disable_with_predicate(...)
: Disables the test if pred is true.
gpu_device_name(...)
: Returns the name of a GPU device if available or a empty string.
is_built_with_cuda(...)
: Returns whether TensorFlow was built with CUDA (GPU) support.
is_built_with_gpu_support(...)
: Returns whether TensorFlow was built with GPU (CUDA or ROCm) support.
is_built_with_rocm(...)
: Returns whether TensorFlow was built with ROCm (GPU) support.
is_built_with_xla(...)
: Returns whether TensorFlow was built with XLA support.
is_gpu_available(...)
: Returns whether TensorFlow can access a GPU. (deprecated)
main(...)
: Runs all unit tests.
with_eager_op_as_function(...)
: Returns the same class. This will be removed once all usages are removed.