model speed error

[jxncyym]

Describe the issue: when I compress the model: https://github.com/chenjun2hao/DDRNet.pytorch I meet the error:

use L1FilterPruner error in these codes: pruner._unwrap_model() m_speedup = ModelSpeedup(model, dummy_input, mask_path, device) m_speedup.speedup_model()

error info: exception: no description

[2021-09-06 19:24:34] INFO (nni.compression.pytorch.speedup.compressor/MainThread) start to speed up the model start to speed up the model /algdata02/yiming.yu/DDRNet.pytorch_pruner/envp_3/lib/python3.6/site-packages/torch/nn/functional.py:3063: UserWarning: Default upsampling behavior when mode=bilinear is changed to align_corners=False since 0.4.0. Please specify align_corners=True if the old behavior is desired. See the documentation of nn.Upsample for details. "See the documentation of nn.Upsample for details.".format(mode)) /algdata02/yiming.yu/DDRNet.pytorch_pruner/envp_3/lib/python3.6/site-packages/torch/jit/_trace.py:940: TracerWarning: Encountering a list at the output of the tracer might cause the trace to be incorrect, this is only valid if the container structure does not change based on the module's inputs. Consider using a constant container instead (e.g. for list, use a tuple instead. for dict, use a NamedTuple instead). If you absolutely need this and know the side effects, pass strict=False to trace() to allow this behavior. _forceoutplace, [2021-09-06 19:24:36] INFO (FixMaskConflict/MainThread){'conv1.0': 1, 'conv1.3': 1, 'layer1.0.conv1': 1, 'layer1.0.conv2': 1, 'layer1.1.conv1': 1, 'layer1.1.conv2': 1, 'layer2.0.conv1': 1, 'layer2.0.conv2': 1, 'layer2.0.downsample.0': 1, 'layer2.1.conv1': 1, 'layer2.1.conv2': 1, 'layer3.0.conv1': 1, 'layer3.0.conv2': 1, 'layer3.0.downsample.0': 1, 'layer3.1.conv1': 1, 'layer3.1.conv2': 1, 'layer3.0.conv1': 1, 'layer3.0.conv2': 1, 'layer3.1.conv1': 1, 'layer3.1.conv2': 1, 'down3.0': 1, 'compression3.0': 1, 'layer4.0.conv1': 1, 'layer4.0.conv2': 1, 'layer4.0.downsample.0': 1, 'layer4.1.conv1': 1, 'layer4.1.conv2': 1, 'layer4.0.conv1': 1, 'layer4.0.conv2': 1, 'layer4.1.conv1': 1, 'layer4.1.conv2': 1, 'down4.0': 1, 'down4.3': 1, 'compression4.0': 1, 'layer5.0.conv1': 1, 'layer5.0.conv2': 1, 'layer5.0.conv3': 1, 'layer5_.0.downsample.0': 1, 'layer5.0.conv1': 1, 'layer5.0.conv2': 1, 'layer5.0.conv3': 1, 'layer5.0.downsample.0': 1, 'spp.scale0.2': 1, 'spp.scale1.3': 1, 'spp.process1.2': 1, 'spp.scale2.3': 1, 'spp.process2.2': 1, 'spp.scale3.3': 1, 'spp.process3.2': 1, 'spp.scale4.3': 1, 'spp.process4.2': 1, 'spp.compression.2': 1, 'spp.shortcut.2': 1, 'final_layer.conv1': 1, 'final_layer.conv2': 1, 'seghead_extra.conv1': 1, 'segheadextra.conv2': 1} {'conv1.0': 1, 'conv1.3': 1, 'layer1.0.conv1': 1, 'layer1.0.conv2': 1, 'layer1.1.conv1': 1, 'layer1.1.conv2': 1, 'layer2.0.conv1': 1, 'layer2.0.conv2': 1, 'layer2.0.downsample.0': 1, 'layer2.1.conv1': 1, 'layer2.1.conv2': 1, 'layer3.0.conv1': 1, 'layer3.0.conv2': 1, 'layer3.0.downsample.0': 1, 'layer3.1.conv1': 1, 'layer3.1.conv2': 1, 'layer3.0.conv1': 1, 'layer3.0.conv2': 1, 'layer3.1.conv1': 1, 'layer3.1.conv2': 1, 'down3.0': 1, 'compression3.0': 1, 'layer4.0.conv1': 1, 'layer4.0.conv2': 1, 'layer4.0.downsample.0': 1, 'layer4.1.conv1': 1, 'layer4.1.conv2': 1, 'layer4.0.conv1': 1, 'layer4.0.conv2': 1, 'layer4.1.conv1': 1, 'layer4.1.conv2': 1, 'down4.0': 1, 'down4.3': 1, 'compression4.0': 1, 'layer5.0.conv1': 1, 'layer5.0.conv2': 1, 'layer5.0.conv3': 1, 'layer5_.0.downsample.0': 1, 'layer5.0.conv1': 1, 'layer5.0.conv2': 1, 'layer5.0.conv3': 1, 'layer5.0.downsample.0': 1, 'spp.scale0.2': 1, 'spp.scale1.3': 1, 'spp.process1.2': 1, 'spp.scale2.3': 1, 'spp.process2.2': 1, 'spp.scale3.3': 1, 'spp.process3.2': 1, 'spp.scale4.3': 1, 'spp.process4.2': 1, 'spp.compression.2': 1, 'spp.shortcut.2': 1, 'final_layer.conv1': 1, 'final_layer.conv2': 1, 'seghead_extra.conv1': 1, 'seghead_extra.conv2': 1} /algdata02/yiming.yu/DDRNet.pytorch_pruner/envp_3/lib/python3.6/site-packages/nni/compression/pytorch/utils/mask_conflict.py:120: UserWarning: This overload of nonzero is deprecated: nonzero() Consider using one of the following signatures instead: nonzero(*, bool as_tuple) (Triggered internally at /pytorch/torch/csrc/utils/python_arg_parser.cpp:882.) all_ones = (w_mask.flatten(1).sum(-1) == count).nonzero().squeeze(1).tolist() [2021-09-06 19:24:36] INFO (FixMaskConflict/MainThread) dim0 sparsity: 0.499858 dim0 sparsity: 0.499858 [2021-09-06 19:24:36] INFO (FixMaskConflict/MainThread) dim1 sparsity: 0.000000 dim1 sparsity: 0.000000 [2021-09-06 19:24:36] INFO (FixMaskConflict/MainThread) Dectected conv prune dim" 0 Dectected conv prune dim" 0

Environment:

• NNI version:
• Training service (local|remote|pai|aml|etc):
• Client OS: ubuntu 18.0
• Server OS (for remote mode only):
• Python version:3.6
• PyTorch/TensorFlow version:1.7.1
• Is conda/virtualenv/venv used?:virtualenv used
• Is running in Docker?:no

Configuration:

• Experiment config (remember to remove secrets!):
• Search space:

Log message:

• nnimanager.log:
• dispatcher.log:
• nnictl stdout and stderr:

How to reproduce it?:

[YukSing12]

I met the same problem when I pruned the ddrnet. Could you share your solution if you solved this problem?

[HappyPeanuts]

hi, i meet the same problem.Have you resolved this problem? And how to resolve it ?