MCPcopy Create free account
hub / github.com/alibaba/nann / multi_gpu_model

Function multi_gpu_model

tensorflow/tensorflow/python/keras/utils/multi_gpu_utils.py:37–265  ·  view source on GitHub ↗

Replicates a model on different GPUs. Specifically, this function implements single-machine multi-GPU data parallelism. It works in the following way: - Divide the model's input(s) into multiple sub-batches. - Apply a model copy on each sub-batch. Every model copy is executed on a de

(model, gpus, cpu_merge=True, cpu_relocation=False)

Source from the content-addressed store, hash-verified

source not stored for this graph (policy: none)

Callers

nothing calls this directly

Calls 13

clone_modelFunction · 0.90
LambdaClass · 0.90
concatenateFunction · 0.90
ModelClass · 0.90
_get_available_devicesFunction · 0.85
_normalize_device_nameFunction · 0.85
tupleFunction · 0.85
modelFunction · 0.85
rangeFunction · 0.50
deviceMethod · 0.45
name_scopeMethod · 0.45
as_listMethod · 0.45

Tested by

no test coverage detected