talwalkarlab / paleo Goto Github PK

View Code? Open in Web Editor NEW

85.0 85.0 39.0 425 KB

An analytical performance modeling tool for deep neural networks.

Home Page: https://talwalkarlab.github.io/paleo/

License: Apache License 2.0

Shell 3.56% Python 96.37% Makefile 0.08%

deep-learning deep-neural-networks distributed-computing paleo

paleo's People

Contributors

Stargazers

Watchers

paleo's Issues

Cannot run profile against alex_v1

$ ./paleo.sh profile nets/alex_v2_1gpu.json 
Network: nets/alex_v2_1gpu.json
Direction: forward
Traceback (most recent call last):
  File "paleo/profiler.py", line 427, in <module>
    cli()
  File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 716, in __call__
    return self.main(*args, **kwargs)
  File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 696, in main
    rv = self.invoke(ctx)
  File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 1060, in invoke
    return _process_result(sub_ctx.command.invoke(sub_ctx))
  File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 889, in invoke
    return ctx.invoke(self.callback, **ctx.params)
  File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 534, in invoke
    return callback(*args, **kwargs)
  File "paleo/profiler.py", line 414, in profile
    _print_tabular(cu, tf)
  File "paleo/profiler.py", line 343, in _print_tabular
    assert len(cudnn_result) == len(tensorflow_result)
TypeError: object of type 'NoneType' has no len()

Cannot run fullpass against alex_v1

$ ./paleo.sh fullpass nets/alex_v2_1gpu.json
INFO /home/gaul/work/paleo/paleo/profilers/tensorflow_profiler.py:21] TensorFlow version: 1.6.0
Traceback (most recent call last):
  File "paleo/profiler.py", line 427, in <module>
    cli()
  File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 716, in __call__
    return self.main(*args, **kwargs)
  File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 696, in main
    rv = self.invoke(ctx)
  File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 1060, in invoke
    return _process_result(sub_ctx.command.invoke(sub_ctx))
  File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 889, in invoke
    return ctx.invoke(self.callback, **ctx.params)
  File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 534, in invoke
    return callback(*args, **kwargs)
  File "paleo/profiler.py", line 317, in fullpass
    device_name, num_warmup, num_iter, batch_size)
  File "paleo/profiler.py", line 132, in profile_full_pass
    return profiler.profile_full_pass(layers)
  File "/home/gaul/work/paleo/paleo/profilers/tensorflow_profiler.py", line 44, in profile_full_pass
    graph, end_points, variables = self._compose_full_graph(layers)
  File "/home/gaul/work/paleo/paleo/profilers/tensorflow_profiler.py", line 93, in _compose_full_graph
    ops, _ = self._ops_softmax(layer, graph, end_points)
  File "/home/gaul/work/paleo/paleo/profilers/tensorflow_profiler.py", line 246, in _ops_softmax
    layer.outputs[0], layer.outputs[1])))
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/ops/nn_ops.py", line 2013, in sparse_softmax_cross_entropy_with_logits
    labels, logits)
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/ops/nn_ops.py", line 1779, in _ensure_xent_args
    "named arguments (labels=..., logits=..., ...)" % name)
ValueError: Only call `sparse_softmax_cross_entropy_with_logits` with named arguments (labels=..., logits=..., ...)

Upgrade TensorFlow dependency

Presently Paleo can only use TensorFlow 0.12. Running with the latest 1.6.0 generates an error:

$ ./paleo.sh profile nets/inception_v3.json 
Network: nets/inception_v3.json
Direction: forward
Traceback (most recent call last):
  File "paleo/profiler.py", line 427, in <module>
    cli()
  File "/home/gaul/.virtualenvs/paleo/local/lib/python2.7/site-packages/click/core.py", line 722, in __call__
    return self.main(*args, **kwargs)
  File "/home/gaul/.virtualenvs/paleo/local/lib/python2.7/site-packages/click/core.py", line 697, in main
    rv = self.invoke(ctx)
  File "/home/gaul/.virtualenvs/paleo/local/lib/python2.7/site-packages/click/core.py", line 1066, in invoke
    return _process_result(sub_ctx.command.invoke(sub_ctx))
  File "/home/gaul/.virtualenvs/paleo/local/lib/python2.7/site-packages/click/core.py", line 895, in invoke
    return ctx.invoke(self.callback, **ctx.params)
  File "/home/gaul/.virtualenvs/paleo/local/lib/python2.7/site-packages/click/core.py", line 535, in invoke
    return callback(*args, **kwargs)
  File "paleo/profiler.py", line 414, in profile
    _print_tabular(cu, tf)
  File "paleo/profiler.py", line 343, in _print_tabular
    assert len(cudnn_result) == len(tensorflow_result)
TypeError: object of type 'NoneType' has no len()

Travis fails due to flake8 error

paleo/utils/convertors.py:89:36: E741 ambiguous variable name 'l'

Observed at: https://travis-ci.org/TalwalkarLab/paleo/builds/350960556

Lack of cuDNN 5 and WINOGRAD algorithm support

I tried to run
./scripts/exp_case_alexnet_scalability.sh
and raised a butch of Warning and got no profiling results:

WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
INFO build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:393] Tile FFT: 1 (32x32) 1D: False
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
INFO build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:393] Tile FFT: 1 (32x32) 1D: False
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
INFO build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:393] Tile FFT: 1 (32x32) 1D: False
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
INFO build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:393] Tile FFT: 1 (32x32) 1D: False
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
INFO build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:393] Tile FFT: 1 (32x32) 1D: False
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
INFO build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:393] Tile FFT: 1 (32x32) 1D: False
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
INFO build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:393] Tile FFT: 1 (32x32) 1D: False
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
INFO build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:393] Tile FFT: 1 (32x32) 1D: False
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
INFO build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:393] Tile FFT: 1 (32x32) 1D: False
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
INFO build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:393] Tile FFT: 1 (32x32) 1D: False
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:192] Unsupported algorithm: CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD
WARNING build/bdist.linux-x86_64/egg/paleo/profilers/flops_profiler.py:246] Unsupported algorithm: CUDNN_CONVOLUTION_BWD_DATA_ALGO_WINOGRAD

What might be the problem? Thanks

Integrate network latency

It'd be nice to have the latency for different networks to see how much that influences the results. Currently, the network object only looks at bandwidth:

paleo/paleo/device.py

Lines 38 to 54 in 984bbf5

 class Network(object): 

 def __init__(self, name, bandwidth): 

 """ 

  Args: 

  name: name of this network connection. 

  bandwidth: in Gbps. 

  """ 

 self._name = name 

 self._bandwidth_Gbps = bandwidth 

 @property 

 def name(self): 

 return self._name 

 @property 

 def bandwidth(self): 

 return self._bandwidth_Gbps

CuDNN error

Dear developers,

I installed paleo on a machine that did not have cuDNN (since the install instructions say it is optional). Then I ran

#! /bin/sh

NET_FILE=nets/inception_v3.json
OUT_FILE=results/case_inception

PPP_COMP=0.62
PPP_COMM=0.72

./paleo.sh simulate $NET_FILE
--use_only_gemm
--batch_size=256
--network_name=ethernet
--device_name=K40
--num_workers=1,2,4,8,16,50,100
--ppp_comp=$PPP_COMP
--ppp_comm=$PPP_COMM
--scaling=weak >> $OUT_FILE

but I got the following error.

File "paleo/profiler.py", line 427, in <module> cli() File "/Users/pbalapra/anaconda2/envs/paleo/lib/python2.7/site-packages/click/core.py", line 722, in __call__ return self.main(*args, **kwargs) File "/Users/pbalapra/anaconda2/envs/paleo/lib/python2.7/site-packages/click/core.py", line 697, in main rv = self.invoke(ctx) File "/Users/pbalapra/anaconda2/envs/paleo/lib/python2.7/site-packages/click/core.py", line 1066, in invoke return _process_result(sub_ctx.command.invoke(sub_ctx)) File "/Users/pbalapra/anaconda2/envs/paleo/lib/python2.7/site-packages/click/core.py", line 895, in invoke return ctx.invoke(self.callback, **ctx.params) File "/Users/pbalapra/anaconda2/envs/paleo/lib/python2.7/site-packages/click/core.py", line 535, in invoke return callback(*args, **kwargs) File "paleo/profiler.py", line 297, in simulate ppp_comm, parallel, hybrid_workers) File "paleo/profiler.py", line 149, in simulate ppp_comm) File "/Users/pbalapra/Projects/repos/2018/paleo/paleo/simulation.py", line 250, in simulate_scaling device) File "/Users/pbalapra/Projects/repos/2018/paleo/paleo/simulation.py", line 59, in _profile_for_apply_updates device) File "/Users/pbalapra/Projects/repos/2018/paleo/paleo/profilers/flops_profiler.py", line 24, in __init__ from paleo.profilers import cudnn_profiler as cudnn File "/Users/pbalapra/Projects/repos/2018/paleo/paleo/profilers/cudnn_profiler.py", line 10, in <module> from paleo.third_party import libcudnn File "/Users/pbalapra/Projects/repos/2018/paleo/paleo/third_party/libcudnn.py", line 52, in <module> raise OSError('cuDNN library not found') OSError: cuDNN library not found

Travis integration tests

I could not run 3 of the 4 Paleo commands with the provided test inputs as reported in #3 #4 #5. Could you add integration tests to your .travis.yml so that I could be sure whether my environment creates these issues?

Missing comma

Hi! I think there is a missing comma in file "paleo/profilers/cudnn_profiler.py" at line 188 (after "3: 'CUDNN_CONVOLUTION_BWD_DATA_ALGO_FFT_TILING'"). Thanks!

Recommend Projects

React

A declarative, efficient, and flexible JavaScript library for building user interfaces.
Vue.js

🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript

TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
TensorFlow

An Open Source Machine Learning Framework for Everyone
Django

The Web framework for perfectionists with deadlines.
Laravel

A PHP framework for web artisans
D3

Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

javascript

JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
web

Some thing interesting about web. New door for the world.
server

A server is a program made to process requests and deliver data to clients.
Machine learning

Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Visualization

Some thing interesting about visualization, use data art
Game

Some thing interesting about game, make everyone happy.

Recommend Org

Facebook

We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft

Open source projects and samples from Microsoft.
Google

Google ❤️ Open Source for everyone.
Alibaba

Alibaba Open Source for everyone
D3

Data-Driven Documents codes.
Tencent

China tencent open source team.

	class Network(object):
	def __init__(self, name, bandwidth):
	"""
	Args:
	name: name of this network connection.
	bandwidth: in Gbps.
	"""
	self._name = name
	self._bandwidth_Gbps = bandwidth

	@property
	def name(self):
	return self._name

	@property
	def bandwidth(self):
	return self._bandwidth_Gbps