hub / github.com/alibaba/nann / switch_to_thread_local

Method switch_to_thread_local

tensorflow/tensorflow/python/framework/ops.py:4910–4925 · view source on GitHub ↗

Make device, colocation and dependencies stacks thread-local. Device, colocation and dependencies stacks are not thread-local be default. If multiple threads access them, then the state is shared. This means that one thread may affect the behavior of another thread. After this met

(self)

Source from the content-addressed store, hash-verified

source not stored for this graph (policy: none)

Callers 15

testDeviceFunctionStackMethod · 0.95

testColocateWithMethod · 0.95

testControlDependenciesMethod · 0.95

__init__Method · 0.95

testDistributedSimpleMethod · 0.80

testParallelEnqueueMethod · 0.80

testParallelDequeueMethod · 0.80

testEnqueueAndBlockingDequeueMethod · 0.80

testParallelEnqueueManyMethod · 0.80

testParallelDequeueManyMethod · 0.80

testParallelDequeueUpToMethod · 0.80

testParallelEnqueueAndDequeueMethod · 0.80

Calls

no outgoing calls

Tested by 15

testDeviceFunctionStackMethod · 0.76

testColocateWithMethod · 0.76

testControlDependenciesMethod · 0.76

testDistributedSimpleMethod · 0.64

testParallelEnqueueMethod · 0.64

testParallelDequeueMethod · 0.64

testEnqueueAndBlockingDequeueMethod · 0.64

testParallelEnqueueManyMethod · 0.64

testParallelDequeueManyMethod · 0.64

testParallelDequeueUpToMethod · 0.64

testParallelEnqueueAndDequeueMethod · 0.64

testBlockingDequeueManyMethod · 0.64