I am trying to prune from mmdet (https://github.com/open-mmlab/mmdetection/blob/master/configs/atss/atss_r50_fpn_1x_coco.py)

But it throw the exception when forwarding with FPN.

Any idea?

By the way, I think it's better to let users to configure the whole block as a group (like neck and bbox_head) which sharing the mask, since these blocks are always complicated, and the parsers are hard to modify to deal with these cases.

I make a config for yolox to use autoslim, but get an error:

error:

Traceback (most recent call last): File "tools/mmdet/train_mmdet.py", line 199, in main() File "tools/mmdet/train_mmdet.py", line 175, in main datasets = [build_dataset(cfg.data.train)] File "/home/yangmin/share/openmmlab/mmdetection/mmdet/datasets/builder.py", line 77, in build_dataset dataset = MultiImageMixDataset(**cp_cfg) TypeError: init() got an unexpected keyword argument 'ann_file'

config

########################################### base = [ '../../base/datasets/mmdet/coco_detection.py', '../../base/schedules/mmdet/schedule_1x.py', '../../base/mmdet_runtime.py' ]

img_scale = (640, 640)

model = dict( type='mmdet.YOLOX', input_size=img_scale, random_size_range=(15, 25), random_size_interval=10, backbone=dict(type='CSPDarknet', deepen_factor=0.33, widen_factor=0.5), neck=dict( type='YOLOXPAFPN', in_channels=[128, 256, 512], out_channels=128, num_csp_blocks=1), bbox_head=dict( type='YOLOXHead', num_classes=80, in_channels=128, feat_channels=128), train_cfg=dict(assigner=dict(type='SimOTAAssigner', center_radius=2.5)), # In order to align the source code, the threshold of the val phase is # 0.01, and the threshold of the test phase is 0.001. test_cfg=dict(score_thr=0.01, nms=dict(type='nms', iou_threshold=0.65)))

data_root = 'data/coco/' dataset_type = 'CocoDataset'

train_pipeline = [ dict(type='Mosaic', img_scale=img_scale, pad_val=114.0), dict( type='RandomAffine', scaling_ratio_range=(0.1, 2), border=(-img_scale[0] // 2, -img_scale[1] // 2)), dict( type='MixUp', img_scale=img_scale, ratio_range=(0.8, 1.6), pad_val=114.0), dict(type='YOLOXHSVRandomAug'), dict(type='RandomFlip', flip_ratio=0.5), # According to the official implementation, multi-scale # training is not considered here but in the # 'mmdet/models/detectors/yolox.py'. dict(type='Resize', img_scale=img_scale, keep_ratio=True), dict( type='Pad', pad_to_square=True, # If the image is three-channel, the pad value needs # to be set separately for each channel. pad_val=dict(img=(114.0, 114.0, 114.0))), dict(type='FilterAnnotations', min_gt_bbox_wh=(1, 1), keep_empty=False), dict(type='DefaultFormatBundle'), dict(type='Collect', keys=['img', 'gt_bboxes', 'gt_labels']) ]

train_dataset = dict( type='MultiImageMixDataset', dataset=dict( type=dataset_type, ann_file=data_root + 'annotations/instances_train2017.json', img_prefix=data_root + 'train2017/', pipeline=[ dict(type='LoadImageFromFile'), dict(type='LoadAnnotations', with_bbox=True) ], filter_empty_gt=False, ), pipeline=train_pipeline)

test_pipeline = [ dict(type='LoadImageFromFile'), dict( type='MultiScaleFlipAug', img_scale=img_scale, flip=False, transforms=[ dict(type='Resize', keep_ratio=True), dict(type='RandomFlip'), dict( type='Pad', pad_to_square=True, pad_val=dict(img=(114.0, 114.0, 114.0))), dict(type='DefaultFormatBundle'), dict(type='Collect', keys=['img']) ]) ]

data = dict( samples_per_gpu=8, workers_per_gpu=4, persistent_workers=True, train=train_dataset, val=dict( type=dataset_type, ann_file=data_root + 'annotations/instances_val2017.json', img_prefix=data_root + 'val2017/', pipeline=test_pipeline), test=dict( type=dataset_type, ann_file=data_root + 'annotations/instances_val2017.json', img_prefix=data_root + 'val2017/', pipeline=test_pipeline))

optimizer = dict( type='SGD', lr=0.01, momentum=0.9, weight_decay=5e-4, nesterov=True, paramwise_cfg=dict(norm_decay_mult=0., bias_decay_mult=0.)) optimizer_config = dict(grad_clip=None)

max_epochs = 300 num_last_epochs = 15 resume_from = None interval = 10

lr_config = dict( delete=True, policy='YOLOX', warmup='exp', by_epoch=False, warmup_by_epoch=True, warmup_ratio=1, warmup_iters=5, # 5 epoch num_last_epochs=num_last_epochs, min_lr_ratio=0.05)

runner = dict(type='EpochBasedRunner', max_epochs=max_epochs)

custom_hooks = [ dict( type='YOLOXModeSwitchHook', num_last_epochs=num_last_epochs, priority=48), dict( type='SyncNormHook', num_last_epochs=num_last_epochs, interval=interval, priority=48), dict( type='ExpMomentumEMAHook', resume_from=resume_from, momentum=0.0001, priority=49) ] checkpoint_config = dict(interval=interval) evaluation = dict( save_best='auto', # The evaluation interval is 'interval' when running epoch is # less than ‘max_epochs - num_last_epochs’. # The evaluation interval is 1 when running epoch is greater than # or equal to ‘max_epochs - num_last_epochs’. interval=interval, dynamic_intervals=[(max_epochs - num_last_epochs, 1)], metric='bbox') log_config = dict(interval=50)

algorithm = dict( type='AutoSlim', architecture=dict(type='MMDetArchitecture', model=model), #distiller=dict( # type='SelfDistiller', # components=[ # dict( # student_module='bbox_head.cls_score', # teacher_module='bbox_head.cls_score', # losses=[ # dict( # type='KLDivergence', # name='loss_kd', # tau=1, # loss_weight=1, # ) # ]), # ]), pruner=dict( type='RatioPruner', ratios=(2 / 12, 3 / 12, 4 / 12, 5 / 12, 6 / 12, 7 / 12, 8 / 12, 9 / 12, 10 / 12, 11 / 12, 1.0)), retraining=False, bn_training_mode=True, input_shape=None)

runner = dict(type='EpochBasedRunner', max_epochs=50)

use_ddp_wrapper = True ###############################

Describe the bug

Refer to Pruning-Split Checkpoint(Optional) for the following command:

$ python ./tools/model_converters/split_checkpoint.py \
>   configs/pruning/autoslim/autoslim_mbv2_supernet_8xb256_ci10.py \
>   output/epoch_49.pth \
>   --channel-cfgs output/subnet_497881198.yaml,output/subnet_297081551.yaml,output/subnet_197465826.yaml
load checkpoint from local path: output/epoch_49.pth
Traceback (most recent call last):
  File "./tools/model_converters/split_checkpoint.py", line 69, in <module>
    main()
  File "./tools/model_converters/split_checkpoint.py", line 60, in main
    algorithm.channel_cfg[i])
  File "/home/tanghuayang/venv_torch/lib64/python3.6/site-packages/torch/nn/modules/module.py", line 948, in __getattr__
    type(self).__name__, name))
AttributeError: 'AutoSlim' object has no attribute 'channel_cfg'

I am trying to train autoslim on cifar10 with mobilenetV2.

But I found there are some errors when training

And I found some of root cause. For example,

You have already called optimized.step() here (https://github.com/open-mmlab/mmrazor/blob/master/mmrazor/models/algorithms/autoslim.py#L226)

but in mmcv here is already have one (https://github.com/open-mmlab/mmcv/blob/master/mmcv/runner/hooks/optimizer.py#L56).

Do i need to configure something?

thank you

Describe the bug

A clear and concise description of what the bug is.

when i running cls-AutoSlim code by step 1, training a cls supernet on my own dataset, it appeared a AttributeError.

To Reproduce

The command you executed.

Traceback (most recent call last):
  File "tools/mmcls/train_mmcls.py", line 170, in <module>
    main()
  File "tools/mmcls/train_mmcls.py", line 140, in main
    algorithm = build_algorithm(cfg.algorithm)
  File "/home/ubuntu/project/hzw/mmrazor-master/mmrazor/models/builder.py", line 20, in build_algorithm
    return ALGORITHMS.build(cfg)
  File "/home/ubuntu/miniconda/envs/mmseg/lib/python3.8/site-packages/mmcv/utils/registry.py", line 212, in build
    return self.build_func(*args, **kwargs, registry=self)
  File "/home/ubuntu/miniconda/envs/mmseg/lib/python3.8/site-packages/mmcv/cnn/builder.py", line 27, in build_model_from_cfg
    return build_from_cfg(cfg, registry, default_args)
  File "/home/ubuntu/miniconda/envs/mmseg/lib/python3.8/site-packages/mmcv/utils/registry.py", line 55, in build_from_cfg
    raise type(e)(f'{obj_cls.__name__}: {e}')
AttributeError: AutoSlim: 'SliceBackward' object has no attribute 'variable'

Based on my own dataset, I trained a model using mmseg1.x. Now I want to realize knowledge distillation, resnet101+deeplab--->resnet18+deeplab. I configured the mmrazor environment, but in the distill config file https://github.com/open-mmlab/mmrazor/blob/bbb58f1a5c2fe2878484856767dba540092bc7bf/configs/distill /mmseg/cwd/cwd_logits_pspnet_r101-d8_pspnet_r18-d8_4xb2-80k_cityscapes-512x1024.py#L2 I don't know where I can modify the default settings of mmseg, such as the number of categories to be segmented, data set selection, etc.? Can you give me some help?

Hi,

Is there a inference demo in the code base? Try to init model follow test/test_apis/test_inference.py.

result = inference_detector(model, img_path) show_result_pyplot(model, img_path, result, wait_time=1, score_thr=0.7, palette=(72, 101, 241))

The terminal shows model and checkpoints loaded state dict do not match exactly.
The bbox painted on image is not correct, and every time is different. However, I'm sure the mAP tested by test.py is correct.

When I follow train.md for distilling model with config.py, such errors happened. It seems that my input 'imgs' data type is error in mmdet.detectors.base.lineL78 (should be tensor but Datacontainers). So, does my envs configuration version not match? Following is my envs configuration: TorchVision: 0.9.0 OpenCV: 4.5.4 MMCV: 1.4.0 MMCV Compiler: GCC 7.3 MMCV CUDA Compiler: 11.1 MMDetection: 2.20.0

I downloaded the voc2012 dataset, and downloaded the weights（pspnet_r101-d8_512x512_20k_voc12aug_20200617_102003-4aef3c9a.pth ） in mmsegmentation, and ran the following code.

python tools/mmseg/train_mmseg.py configs/distill/cwd/cwd_cls_head_pspnet_r101_d8_pspnet_r18_d8_512x1024_voc2012_80k.py --cfg-options algorithm.distiller.teacher.init_cfg.type=Pretrained algorithm.distiller.teacher.init_cfg.checkpoint=pretrained/pspnet_r101-d8_512x512_20k_voc12aug_20200617_102003-4aef3c9a.pth

the validation results are as follows: +-------------+-------+-------+ | Class | IoU | Acc | +-------------+-------+-------+ | background | 73.31 | 99.89 | | aeroplane | 0.0 | 0.0 | | bicycle | 0.0 | 0.0 | | bird | 0.0 | 0.0 | | boat | 0.0 | 0.0 | | bottle | 0.0 | 0.0 | | bus | 0.0 | 0.0 | | car | 0.0 | 0.0 | | cat | 0.0 | 0.0 | | chair | 0.0 | 0.0 | | cow | 0.0 | 0.0 | | diningtable | 0.0 | 0.0 | | dog | 0.0 | 0.0 | | horse | 0.0 | 0.0 | | motorbike | 0.0 | 0.0 | | person | 0.32 | 0.33 | | pottedplant | 0.0 | 0.0 | | sheep | 0.0 | 0.0 | | sofa | 0.0 | 0.0 | | train | 0.0 | 0.0 | | tvmonitor | 0.0 | 0.0 | +-------------+-------+-------+ 2021-12-24 11:09:10,633 - mmseg - INFO - Summary: 2021-12-24 11:09:10,633 - mmseg - INFO - +-------+------+------+ | aAcc | mIoU | mAcc | +-------+------+------+ | 73.25 | 3.51 | 4.77 | +-------+------+------+

can you help me sove this? Thanks

model = torch.hub.load('facebookresearch/detr:main', 'detr_resnet50', pretrained=True) graph = custom_symbolic_trace(model)

/usr/bin/python3 /Users/unicorn/MyCodes/net_op/fx/mmrazor_tracer.py Using cache found in /Users/unicorn/.cache/torch/hub/facebookresearch_detr_main /Users/unicorn/Library/Python/3.9/lib/python/site-packages/torchvision/models/_utils.py:208: UserWarning: The parameter 'pretrained' is deprecated since 0.13 and may be removed in the future, please use 'weights' instead. warnings.warn( /Users/unicorn/Library/Python/3.9/lib/python/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or None for 'weights' are deprecated since 0.13 and may be removed in the future. The current behavior is equivalent to passing weights=ResNet50_Weights.IMAGENET1K_V1. You can also use weights=ResNet50_Weights.DEFAULT to get the most up-to-date weights. warnings.warn(msg) Downloading: "https://dl.fbaipublicfiles.com/detr/detr-r50-e632da11.pth" to /Users/unicorn/.cache/torch/hub/checkpoints/detr-r50-e632da11.pth 100%|██████████| 159M/159M [00:15<00:00, 10.5MB/s] Traceback (most recent call last): File "/Users/unicorn/MyCodes/net_op/fx/mmrazor_tracer.py", line 284, in graph = custom_symbolic_trace(model) File "/Users/unicorn/MyCodes/net_op/fx/mmrazor_tracer.py", line 75, in custom_symbolic_trace graph = tracer.trace(root, concrete_args) File "/Users/unicorn/MyCodes/net_op/fx/mmrazor_tracer.py", line 264, in trace 'output', (self.create_arg(fn(*args)), ), {}, File "/Users/unicorn/.cache/torch/hub/facebookresearch_detr_main/models/detr.py", line 61, in forward features, pos = self.backbone(samples) File "/Users/unicorn/MyCodes/net_op/fx/mmrazor_tracer.py", line 225, in module_getattr_wrapper return self._module_getattr(attr, attr_val, parameter_proxy_cache) AttributeError: 'CustomTracer' object has no attribute '_module_getattr'

Describe the bug

A clear and concise description of what the bug is.

在选择配置文件为：cwd_logits_pspnet_r101-d8_pspnet_r18-d8_4xb2-80k_cityscapes-512x1024.py进行模型蒸馏时（其他位置未作改变），在代码运行至 runner = Runner.from_cfg(cfg)处，后经debug至from_cfg函数内发现代码会在run = cls(...)位置卡顿，等待半小时后才开始运行后面的代码。

[here]

To Reproduce

The command you executed.

[here]

Post related information

1. The output of pip list | grep "mmcv\|mmrazor\|^torch" [here]
2. Your config file if you modified it or created a new one.

[here]

3. Your train log file if you meet the problem during training. [here]
4. Other code you modified in the mmrazor folder. [here]

Additional context

Add any other context about the problem here.

[here]

Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers.

Motivation

Please describe the motivation of this PR and the goal you want to achieve through this PR.

Modification

Please briefly describe what modification is made in this PR.

BC-breaking (Optional)

Does the modification introduce changes that break the backward compatibility of the downstream repositories? If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR.

Use cases (Optional)

If this PR introduces a new feature, it is better to list some use cases here and update the documentation.

Checklist

Before PR:

• [ ] Pre-commit or other linting tools are used to fix the potential lint issues.
• [ ] Bug fixes are fully covered by unit tests, the case that causes the bug should be added in the unit tests.
• [ ] The modification is covered by complete unit tests. If not, please add more unit tests to ensure the correctness.
• [ ] The documentation has been modified accordingly, like docstring or example tutorials.

After PR:

• [ ] If the modification has potential influence on downstream or other related projects, this PR should be tested with those projects, like MMDet or MMSeg.
• [ ] CLA has been signed and all committers have signed the CLA in this PR.

Motivation

SingleTeacherDistillValLoop calls after_val_epoch hooks twice during the evaluation. However, the metric in the second call do not contain the key named student.accuracy/top1. As a result, if we set save_best='auto' in CheckpointHook, the program will fail.

Please refer to this issue for more details.

Modification

Now we only call after_val_epoch hooks once in SingleTeacherDistillValLoop.

Modification

1. Refactor graph_utils: We support deleting fakequant before / after op / method / function / module in graph_utils.
2. Add 8 properties to BaseQuantizer. Each is a tuple. For example, the module_del_prev_fakequant property in OpenVINOQuantizer is (torch.nn.ReLU6, torch.nn.Identity). These properties are determined by the backend, so maybe they need to be regarded as properties.
3. Add the corresponding pytest and docstring.

Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers.

Motivation

update readme in master branch

Describe the bug

if the architecture or teacher of SingleTeacherDistill is a instance of TimmClassifier, the ckpt can not be load correctly.

To Reproduce

from mmengine.config import Config
from mmrazor.registry import MODELS
from mmengine.runner import load_checkpoint, save_checkpoint

model = MODELS.build(Config.fromfile('./debug_config.py')['model'])

save_checkpoint(model.state_dict(), './temp.pth')
load_checkpoint(model, './temp.pth', strict=True)

Config

student = dict(
    _scope_='mmcls',
    type='TimmClassifier',
    model_name='deit_tiny_patch16_224',
)

teacher = dict(
    _scope_='mmcls',
    type='TimmClassifier',
    model_name='cait_s24_224',
)

model = dict(
    _scope_='mmrazor',
    type='SingleTeacherDistill',
    architecture=student,
    teacher=teacher,
    distiller=dict(
        type='ConfigurableDistiller',
    ),
)

student = dict( type='mmcls.ImageClassifier', backbone=dict( type='SearchableShuffleNetV2', norm_cfg=dict(type='BN'), act_cfg=dict(type='ReLU'), ),

neck=dict(type='GlobalAveragePooling'),
head=dict(
    type='LinearClsHead',
    num_classes=9,
    in_channels=64,
    loss=dict(type='CrossEntropyLoss', loss_weight=1.0),
    topk=(1, 5)))

how to define the in_channels