FM / python

/usr/local/lib64/python3.6/site-packages/caffe2/python/

/usr/local/lib64/python3.6/site-packages/caffe2/python

Name	Size	Mode	Actions
docs/	-	0755	rm
examples/	-	0755	rm
fakelowp/	-	0755	rm
helpers/	-	0755	rm
ideep/	-	0755	rm
layers/	-	0755	rm
mint/	-	0755	rm
mkl/	-	0755	rm
modeling/	-	0755	rm
models/	-	0755	rm
onnx/	-	0755	rm
operator_test/	-	0755	rm
predictor/	-	0755	rm
rnn/	-	0755	rm
serialized_test/	-	0755	rm
test/	-	0755	rm
trt/	-	0755	rm
__pycache__/	-	0755	rm
allcompare_test.py	2255	0644	edit dl rm
attention.py	12359	0644	edit dl rm
benchmark_generator.py	4912	0644	edit dl rm
binarysize.py	5521	0644	edit dl rm
brew.py	4762	0644	edit dl rm
brew_test.py	11739	0644	edit dl rm
build.py	153	0644	edit dl rm
cached_reader.py	4394	0644	edit dl rm
caffe2_pybind11_state.cpython-36m-x86_64-linux-gnu.so	48299712	0755	edit dl rm
caffe2_pybind11_state_gpu.cpython-36m-x86_64-linux-gnu.so	49048144	0755	edit dl rm
caffe_translator.py	35227	0644	edit dl rm
caffe_translator_test.py	3553	0644	edit dl rm
checkpoint.py	32101	0644	edit dl rm
checkpoint_test.py	13405	0644	edit dl rm
cnn.py	7626	0644	edit dl rm
context.py	2841	0644	edit dl rm
context_test.py	1792	0644	edit dl rm
control.py	19309	0644	edit dl rm
control_ops_grad.py	28893	0644	edit dl rm
control_ops_grad_test.py	1752	0644	edit dl rm
control_ops_util.py	10863	0644	edit dl rm
control_test.py	12276	0644	edit dl rm
convert.py	55	0644	edit dl rm
convert_test.py	201	0644	edit dl rm
convnet_benchmarks.py	20533	0644	edit dl rm
convnet_benchmarks_test.py	839	0644	edit dl rm
core.py	119400	0644	edit dl rm
core_gradients_test.py	38022	0644	edit dl rm
core_test.py	47683	0644	edit dl rm
crf.py	13250	0644	edit dl rm
crf_predict.py	1159	0644	edit dl rm
crf_viterbi_test.py	1663	0644	edit dl rm
dataio.py	23532	0644	edit dl rm
dataio_test.py	17575	0644	edit dl rm
dataset.py	12886	0644	edit dl rm
data_parallel_model.py	83100	0644	edit dl rm
data_parallel_model_test.py	56145	0644	edit dl rm
data_workers.py	15941	0644	edit dl rm
data_workers_test.py	6561	0644	edit dl rm
db_file_reader.py	6608	0644	edit dl rm
db_test.py	1110	0644	edit dl rm
device_checker.py	5157	0644	edit dl rm
dyndep.py	1533	0644	edit dl rm
embedding_generation_benchmark.py	5256	0644	edit dl rm
experiment_util.py	3625	0644	edit dl rm
extension_loader.py	744	0644	edit dl rm
fakefp16_transform_lib.py	322	0644	edit dl rm
filler_test.py	748	0644	edit dl rm
functional.py	4415	0644	edit dl rm
functional_test.py	4204	0644	edit dl rm
fused_8bit_rowwise_conversion_ops_test.py	3945	0644	edit dl rm
gradient_checker.py	15377	0644	edit dl rm
gradient_check_test.py	20729	0644	edit dl rm
gru_cell.py	5129	0644	edit dl rm
hip_test_util.py	405	0644	edit dl rm
hsm_util.py	2259	0644	edit dl rm
hypothesis_test.py	105762	0644	edit dl rm
hypothesis_test_util.py	26853	0644	edit dl rm
ideep_test_util.py	998	0644	edit dl rm
layers_test.py	92931	0644	edit dl rm
layer_model_helper.py	29340	0644	edit dl rm
layer_model_instantiator.py	3935	0644	edit dl rm
layer_parameter_sharing_test.py	9148	0644	edit dl rm
layer_test_util.py	4875	0644	edit dl rm
lazy.py	277	0644	edit dl rm
lazy_dyndep.py	2562	0644	edit dl rm
lazy_dyndep_test.py	3914	0644	edit dl rm
lengths_reducer_fused_8bit_rowwise_ops_test.py	7575	0644	edit dl rm
lengths_reducer_rowwise_8bit_ops_test.py	5710	0644	edit dl rm
lstm_benchmark.py	10649	0644	edit dl rm
memonger.py	34041	0644	edit dl rm
memonger_test.py	36910	0644	edit dl rm
mkl_test_util.py	1142	0644	edit dl rm
model_device_test.py	4777	0644	edit dl rm
model_helper.py	23492	0644	edit dl rm
model_helper_test.py	2336	0644	edit dl rm
modifier_context.py	1772	0644	edit dl rm
muji.py	8131	0644	edit dl rm
muji_test.py	3058	0644	edit dl rm
net_builder.py	27679	0644	edit dl rm
net_builder_test.py	11382	0644	edit dl rm
net_drawer.py	14264	0644	edit dl rm
net_printer.py	12704	0644	edit dl rm
net_printer_test.py	3190	0644	edit dl rm
nomnigraph.py	4216	0644	edit dl rm
nomnigraph_test.py	15427	0644	edit dl rm
nomnigraph_transformations.py	3787	0644	edit dl rm
nomnigraph_transformations_test.py	5767	0644	edit dl rm
normalizer.py	1411	0644	edit dl rm
normalizer_context.py	1007	0644	edit dl rm
normalizer_test.py	487	0644	edit dl rm
numa_benchmark.py	2230	0644	edit dl rm
numa_test.py	1663	0644	edit dl rm
observer_test.py	5316	0644	edit dl rm
operator_fp_exceptions_test.py	1248	0644	edit dl rm
optimizer.py	78813	0644	edit dl rm
optimizer_context.py	1462	0644	edit dl rm
optimizer_test.py	30705	0644	edit dl rm
optimizer_test_util.py	9187	0644	edit dl rm
parallelize_bmuf_distributed_test.py	9908	0644	edit dl rm
parallel_workers.py	7682	0644	edit dl rm
parallel_workers_test.py	3501	0644	edit dl rm
pipeline.py	17283	0644	edit dl rm
pipeline_test.py	2542	0644	edit dl rm
predictor_constants.py	198	0644	edit dl rm
python_op_test.py	9169	0644	edit dl rm
queue_util.py	4459	0644	edit dl rm
record_queue.py	4453	0644	edit dl rm
recurrent.py	13297	0644	edit dl rm
regularizer.py	21120	0644	edit dl rm
regularizer_context.py	1013	0644	edit dl rm
regularizer_test.py	10266	0644	edit dl rm
rnn_cell.py	68233	0644	edit dl rm
schema.py	45621	0644	edit dl rm
schema_test.py	15754	0644	edit dl rm
scope.py	3623	0644	edit dl rm
scope_test.py	5249	0644	edit dl rm
session.py	7642	0644	edit dl rm
session_test.py	2078	0644	edit dl rm
sparse_to_dense_mask_test.py	6565	0644	edit dl rm
sparse_to_dense_test.py	3556	0644	edit dl rm
task.py	24274	0644	edit dl rm
task_test.py	870	0644	edit dl rm
test_util.py	3524	0644	edit dl rm
text_file_reader.py	1990	0644	edit dl rm
timeout_guard.py	4054	0644	edit dl rm
toy_regression_test.py	2822	0644	edit dl rm
transformations.py	1832	0644	edit dl rm
transformations_test.py	11960	0644	edit dl rm
tt_core.py	9349	0644	edit dl rm
tt_core_test.py	2518	0644	edit dl rm
utils.py	14181	0644	edit dl rm
utils_test.py	1399	0644	edit dl rm
visualize.py	6315	0644	edit dl rm
workspace.py	25263	0644	edit dl rm
workspace_test.py	34844	0644	edit dl rm
_import_c_extension.py	2250	0644	edit dl rm
__init__.py	3925	0644	edit dl rm

Edit: /usr/local/lib64/python3.6/site-packages/caffe2/python/muji.py (8131B)

## @package muji
# Module caffe2.python.muji
"""muji.py does multi-gpu training for caffe2 with no need to change the c++
side code. Everything is defined on the computation graph level.

We support the following use cases:
  - 2 gpus, where peer access is enabled between them.
  - 4 gpus, where peer access are enabled between all of them.
  - 4 gpus, where peer access are enabled in two groups,
    between {1, 2} and {3, 4}
  - 8 gpus, where peer access are enabled in two groups,
    between {1, 2, 3, 4} and {5, 6, 7, 8}.
If above cases are not satisfied, a fallback function which does not rely on
peer access will be called.
"""

import numpy as np

from caffe2.proto import caffe2_pb2
from caffe2.python import workspace

def OnGPU(gpu_id):
    """A utility function that returns a device option protobuf of the
  specified gpu id.
  """
    device_option = caffe2_pb2.DeviceOption()
    device_option.device_type = workspace.GpuDeviceType
    device_option.device_id = gpu_id
    return device_option

def OnCPU():
    device_option = caffe2_pb2.DeviceOption()
    device_option.device_type = caffe2_pb2.CPU
    return device_option

def Allreduce(net, blobs, reduced_affix="_reduced", gpu_indices=None):
    """The general Allreduce interface that reroutes the function calls.
    CPUs and AMD GPUs are not supported because
    GetGpuPeerAccessPattern is called to get gpu peer access pattern.
  """
    if gpu_indices is None:
        gpu_indices = list(range(len(blobs)))
    if len(gpu_indices) != len(blobs):
        raise RuntimeError(
            "gpu_indices length and blobs length mismatch: %d vs %d" %
            (len(gpu_indices), len(blobs))
        )
    pattern = workspace.GetGpuPeerAccessPattern()
    if len(blobs) == 2 and pattern.shape[0] >= 2 and np.all(pattern[:2, :2]):
        return Allreduce2(net, blobs, reduced_affix, gpu_indices)
    elif len(blobs) == 4 and pattern.shape[0] >= 4 and np.all(pattern[:4, :4]):
        return Allreduce4(net, blobs, reduced_affix, gpu_indices)
    elif len(blobs) == 4 and pattern.shape[0] >= 4 and np.all(pattern[:2, :2]) and np.all(pattern[2:4, 2:4]):
        return Allreduce4Group2(net, blobs, reduced_affix, gpu_indices)
    elif len(blobs) == 8 and pattern.shape[0] >= 8 and np.all(pattern[:8, :8]):
        return Allreduce8(net, blobs, reduced_affix, gpu_indices)
    else:
        return AllreduceFallback(net, blobs, reduced_affix, gpu_indices)

def Allreduce2(net, blobs, reduced_affix, gpu_indices):
    """Allreduce for 2 gpus.

Algorithm: 0r <- 0 + 1, 1r <- 0r, where r means "reduced"
  """
    a, b = blobs
    gpu_a, gpu_b = gpu_indices
    a_reduced = net.Add([a, b], a + reduced_affix, device_option=OnGPU(gpu_a))
    b_reduced = a_reduced.Copy(
        [],
        b + reduced_affix,
        device_option=OnGPU(gpu_b)
    )
    return a_reduced, b_reduced

def Allreduce4(net, blobs, reduced_affix, gpu_indices):
    """Allreduce for 4 gpus.

Algorithm: 2 level reduction.
      0r <- 0 + 1, 2r <- 2 + 3
      0r <- 0r + 2r
      2r <- 0r,
      1r <- 0r, 3r <- 2r
  """
    a, b, c, d = blobs
    gpu_a, gpu_b, gpu_c, gpu_d = gpu_indices
    # a_reduced <- a+b, c_reduced <- c + d
    a_reduced = net.Add(
        [a, b],
        str(a) + reduced_affix,
        device_option=OnGPU(gpu_a)
    )
    c_reduced = net.Add(
        [c, d],
        str(c) + reduced_affix,
        device_option=OnGPU(gpu_c)
    )
    # a_reduced <- a_reduced + c_reduced
    a_reduced = a_reduced.Add(c_reduced, a_reduced, device_option=OnGPU(gpu_a))
    # broadcast a_reduced to c_reduced
    c_reduced = a_reduced.Copy([], c_reduced, device_option=OnGPU(gpu_c))
    # broadcast to b and d
    b_reduced = a_reduced.Copy(
        [],
        str(b) + reduced_affix,
        device_option=OnGPU(gpu_b)
    )
    d_reduced = c_reduced.Copy(
        [],
        str(d) + reduced_affix,
        device_option=OnGPU(gpu_d)
    )
    return a_reduced, b_reduced, c_reduced, d_reduced

def Allreduce4Group2(net, blobs, reduced_affix, gpu_indices):
    """Allreduce for 4 gpus where peer access are enabled in {0,1} and {2,3}

Algorithm: 2 level reduction.
      0r <- 0 + 1, 2r <- 2 + 3
      0r <- 0r + 2r
      2r <- 0r,
      1r <- 0r, 3r <- 2r
  """
    a, b, c, d = blobs
    gpu_a, gpu_b, gpu_c, gpu_d = gpu_indices
    # a_reduced <- a+b, c_reduced <- c + d
    a_reduced = net.Add(
        [a, b],
        str(a) + reduced_affix,
        device_option=OnGPU(gpu_a)
    )
    c_reduced = net.Add(
        [c, d],
        str(c) + reduced_affix,
        device_option=OnGPU(gpu_c)
    )
    # copy from c_reduce(gpu_c) to c_reduce_copy(gpu_a)
    c_reduced_copy = c_reduced.Copy(
        [],
        str(c_reduced) + '_copy',
        device_option=OnGPU(gpu_a)
    )
    # a_reduced <- a_reduced + c_reduced_copy
    a_reduced = a_reduced.Add(c_reduced_copy, a_reduced, device_option=OnGPU(gpu_a))
    # broadcast a_reduced to c_reduced
    c_reduced = a_reduced.Copy([], c_reduced, device_option=OnGPU(gpu_c))
    # broadcast to b and d
    b_reduced = a_reduced.Copy(
        [],
        str(b) + reduced_affix,
        device_option=OnGPU(gpu_b)
    )
    d_reduced = c_reduced.Copy(
        [],
        str(d) + reduced_affix,
        device_option=OnGPU(gpu_d)
    )
    return a_reduced, b_reduced, c_reduced, d_reduced

def Allreduce8(net, blobs, reduced_affix, gpu_indices):
    """Allreduce for 8 gpus.

Algorithm: 3 level reduction.
      0r <- 0 + 1, 2r <- 2 + 3, 4r <- 4 + 5, 6r <- 6 + 7
      0r <- 0r + 2r, 4r <- 4r + 6r
      0r <- 0r + 4r
      4r <- 0r
      2r <- 0r, 6r <- 4r
      1r <- 0r, 3r <- 2r, 5r <- 4r, 7r <- 6r
  """
    reduced = [None] * 8
    # Reduction level 1
    for i in [0, 2, 4, 6]:
        reduced[i] = net.Add(
            [blobs[i], blobs[i + 1]],
            blobs[i] + reduced_affix,
            device_option=OnGPU(gpu_indices[i])
        )
    # Reduction level 2
    for i in [0, 4]:
        reduced[i] = net.Add(
            [reduced[i], reduced[i + 2]],
            str(blobs[i]) + reduced_affix,
            device_option=OnGPU(gpu_indices[i])
        )
    # Reduction level 3: this involves a copy.
    reduced_4_copy = reduced[4].Copy(
        [],
        str(reduced[4]) + '_copy',
        device_option=OnGPU(gpu_indices[0])
    )
    reduced[0] = reduced[0].Add(
        reduced_4_copy,
        reduced[0],
        device_option=OnGPU(gpu_indices[0])
    )
    # Broadcast level 1
    reduced[4] = reduced[0].Copy(
        [],
        reduced[4],
        device_option=OnGPU(gpu_indices[4])
    )
    # Broadcast level 2
    for i in [2, 6]:
        reduced[i] = reduced[i - 2].Copy(
            [],
            reduced[i],
            device_option=OnGPU(gpu_indices[i])
        )
    # Broadcast level 3
    for i in [1, 3, 5, 7]:
        reduced[i] = reduced[i - 1].Copy(
            [],
            blobs[i] + reduced_affix,
            device_option=OnGPU(gpu_indices[i])
        )
    return reduced

def AllreduceFallback(net, blobs, reduced_affix, gpu_indices):
    """A fallback option for Allreduce with no assumption on p2p.

Algorithm: a flat operation on gpu 0
      0r <- 0
      0r <- 0r + i for i in gpu_indices[1:]
      ir <- 0r for i in gpu_indices[1:]
  """
    reduced = [None] * len(gpu_indices)
    if reduced_affix != '':
        # copy first
        reduced[0] = net.Copy(
            blobs[0],
            blobs[0] + reduced_affix,
            device_option=OnGPU(gpu_indices[0])
        )
    else:
        reduced[0] = blobs[0]
    # do temp copy and add
    temp_name = reduced[0] + '_temp_copy'
    for i in range(1, len(gpu_indices)):
        temp = net.Copy(
            blobs[i],
            temp_name,
            device_option=OnGPU(gpu_indices[0])
        )
        reduced[0] = net.Add(
            [temp, reduced[0]],
            reduced[0],
            device_option=OnGPU(gpu_indices[0])
        )
    # Broadcast to everyone else
    for i in range(1, len(gpu_indices)):
        reduced[i] = net.Copy(
            reduced[0],
            blobs[i] + reduced_affix,
            device_option=OnGPU(gpu_indices[i])
        )
    return reduced

cmd: