Merge pull request meituan#488 from lippman1125/main

1. fix bug of qat

configs/repopt/yolov6_tiny_hs.py

@@ -0,0 +1,58 @@
+# YOLOv6t model
+model = dict(
+    type='YOLOv6t',
+    pretrained=None,
+    depth_multiple=0.33,
+    width_multiple=0.375,
+    backbone=dict(
+        type='EfficientRep',
+        num_repeats=[1, 6, 12, 18, 6],
+        out_channels=[64, 128, 256, 512, 1024],
+        ),
+    neck=dict(
+        type='RepPANNeck',
+        num_repeats=[12, 12, 12, 12],
+        out_channels=[256, 128, 128, 256, 256, 512],
+        ),
+    head=dict(
+        type='EffiDeHead',
+        in_channels=[128, 256, 512],
+        num_layers=3,
+        begin_indices=24,
+        anchors=1,
+        out_indices=[17, 20, 23],
+        strides=[8, 16, 32],
+        iou_type='siou',
+        use_dfl=False,
+        reg_max=0 #if use_dfl is False, please set reg_max to 0
+    )
+)
+
+solver = dict(
+    optim='SGD',
+    lr_scheduler='Cosine',
+    lr0=0.01,
+    lrf=0.01,
+    momentum=0.937,
+    weight_decay=0.0005,
+    warmup_epochs=3.0,
+    warmup_momentum=0.8,
+    warmup_bias_lr=0.1
+)
+
+data_aug = dict(
+    hsv_h=0.015,
+    hsv_s=0.7,
+    hsv_v=0.4,
+    degrees=0.0,
+    translate=0.1,
+    scale=0.5,
+    shear=0.0,
+    flipud=0.0,
+    fliplr=0.5,
+    mosaic=1.0,
+    mixup=0.0,
+)
+
+# Choose Rep-block by the training Mode, choices=["repvgg", "hyper-search", "repopt"]
+training_mode='hyper_search'

configs/repopt/yolov6_tiny_opt.py

@@ -0,0 +1,58 @@
+# YOLOv6t model
+model = dict(
+    type='YOLOv6t',
+    pretrained=None,
+    scales='../yolov6_assert/v6t_v2_scale_last.pt',
+    depth_multiple=0.33,
+    width_multiple=0.375,
+    backbone=dict(
+        type='EfficientRep',
+        num_repeats=[1, 6, 12, 18, 6],
+        out_channels=[64, 128, 256, 512, 1024],
+        ),
+    neck=dict(
+        type='RepPANNeck',
+        num_repeats=[12, 12, 12, 12],
+        out_channels=[256, 128, 128, 256, 256, 512],
+        ),
+    head=dict(
+        type='EffiDeHead',
+        in_channels=[128, 256, 512],
+        num_layers=3,
+        begin_indices=24,
+        anchors=1,
+        out_indices=[17, 20, 23],
+        strides=[8, 16, 32],
+        iou_type='siou',
+        use_dfl=False,
+        reg_max=0 #if use_dfl is False, please set reg_max to 0
+    )
+)
+
+solver = dict(
+    optim='SGD',
+    lr_scheduler='Cosine',
+    lr0=0.01,
+    lrf=0.01,
+    momentum=0.937,
+    weight_decay=0.0005,
+    warmup_epochs=3.0,
+    warmup_momentum=0.8,
+    warmup_bias_lr=0.1
+)
+
+data_aug = dict(
+    hsv_h=0.015,
+    hsv_s=0.7,
+    hsv_v=0.4,
+    degrees=0.0,
+    translate=0.1,
+    scale=0.5,
+    shear=0.0,
+    flipud=0.0,
+    fliplr=0.5,
+    mosaic=1.0,
+    mixup=0.0,
+)
+# Choose Rep-block by the training Mode, choices=["repvgg", "hyper-search", "repopt"]
+training_mode='repopt'

configs/repopt/yolov6_tiny_opt_qat.py

@@ -0,0 +1,82 @@
+# YOLOv6t model
+model = dict(
+    type='YOLOv6t',
+    pretrained='./assets/v6s_t.pt',
+    scales='./assets/v6t_v2_scale_last.pt',
+    depth_multiple=0.33,
+    width_multiple=0.375,
+    backbone=dict(
+        type='EfficientRep',
+        num_repeats=[1, 6, 12, 18, 6],
+        out_channels=[64, 128, 256, 512, 1024],
+        ),
+    neck=dict(
+        type='RepPANNeck',
+        num_repeats=[12, 12, 12, 12],
+        out_channels=[256, 128, 128, 256, 256, 512],
+        ),
+    head=dict(
+        type='EffiDeHead',
+        in_channels=[128, 256, 512],
+        num_layers=3,
+        begin_indices=24,
+        anchors=1,
+        out_indices=[17, 20, 23],
+        strides=[8, 16, 32],
+        iou_type='siou',
+        use_dfl=False,
+        reg_max=0, #if use_dfl is False, please set reg_max to 0
+        distill_weight={
+            'class': 1.0,
+            'dfl': 1.0,
+        },
+    )
+)
+
+solver = dict(
+    optim='SGD',
+    lr_scheduler='Cosine',
+    lr0=0.00001,
+    lrf=0.001,
+    momentum=0.937,
+    weight_decay=0.00005,
+    warmup_epochs=3.0,
+    warmup_momentum=0.8,
+    warmup_bias_lr=0.1
+)
+
+data_aug = dict(
+    hsv_h=0.015,
+    hsv_s=0.7,
+    hsv_v=0.4,
+    degrees=0.0,
+    translate=0.1,
+    scale=0.5,
+    shear=0.0,
+    flipud=0.0,
+    fliplr=0.5,
+    mosaic=1.0,
+    mixup=0.0,
+)
+
+ptq = dict(
+    num_bits = 8,
+    calib_batches = 4,
+    # 'max', 'histogram'
+    calib_method = 'histogram',
+    # 'entropy', 'percentile', 'mse'
+    histogram_amax_method='entropy',
+    histogram_amax_percentile=99.99,
+    calib_output_path='./',
+    sensitive_layers_skip=False,
+    sensitive_layers_list=[],
+)
+
+qat = dict(
+    calib_pt = './assets/v6s_t_calib_histogram.pt',
+    sensitive_layers_skip = False,
+    sensitive_layers_list=[],
+)
+
+# Choose Rep-block by the training Mode, choices=["repvgg", "hyper-search", "repopt"]
+training_mode='repopt'

configs/repopt/yolov6n_hs.py

@@ -0,0 +1,58 @@
+# YOLOv6n model
+model = dict(
+    type='YOLOv6n',
+    pretrained=None,
+    depth_multiple=0.33,
+    width_multiple=0.25,
+    backbone=dict(
+        type='EfficientRep',
+        num_repeats=[1, 6, 12, 18, 6],
+        out_channels=[64, 128, 256, 512, 1024],
+        ),
+    neck=dict(
+        type='RepPANNeck',
+        num_repeats=[12, 12, 12, 12],
+        out_channels=[256, 128, 128, 256, 256, 512],
+        ),
+    head=dict(
+        type='EffiDeHead',
+        in_channels=[128, 256, 512],
+        num_layers=3,
+        begin_indices=24,
+        anchors=1,
+        out_indices=[17, 20, 23],
+        strides=[8, 16, 32],
+        iou_type='siou',
+        use_dfl=False,
+        reg_max=0 #if use_dfl is False, please set reg_max to 0
+    )
+)
+
+solver = dict(
+    optim='SGD',
+    lr_scheduler='Cosine',
+    lr0=0.02, #0.01 # 0.02 
+    lrf=0.01,
+    momentum=0.937,
+    weight_decay=0.0005,
+    warmup_epochs=3.0,
+    warmup_momentum=0.8,
+    warmup_bias_lr=0.1
+)
+
+data_aug = dict(
+    hsv_h=0.015,
+    hsv_s=0.7,
+    hsv_v=0.4,
+    degrees=0.0,
+    translate=0.1,
+    scale=0.5,
+    shear=0.0,
+    flipud=0.0,
+    fliplr=0.5,
+    mosaic=1.0,
+    mixup=0.0,
+)
+
+# Choose Rep-block by the training Mode, choices=["repvgg", "hyper-search", "repopt"]
+training_mode='hyper_search'

configs/repopt/yolov6n_opt.py

@@ -0,0 +1,58 @@
+# YOLOv6n model
+model = dict(
+    type='YOLOv6n',
+    pretrained=None,
+    scales='../yolov6_assert/v6n_v2_scale_last.pt',
+    depth_multiple=0.33,
+    width_multiple=0.25,
+    backbone=dict(
+        type='EfficientRep',
+        num_repeats=[1, 6, 12, 18, 6],
+        out_channels=[64, 128, 256, 512, 1024],
+        ),
+    neck=dict(
+        type='RepPANNeck',
+        num_repeats=[12, 12, 12, 12],
+        out_channels=[256, 128, 128, 256, 256, 512],
+        ),
+    head=dict(
+        type='EffiDeHead',
+        in_channels=[128, 256, 512],
+        num_layers=3,
+        begin_indices=24,
+        anchors=1,
+        out_indices=[17, 20, 23],
+        strides=[8, 16, 32],
+        iou_type='siou',
+        use_dfl=False,
+        reg_max=0 #if use_dfl is False, please set reg_max to 0
+    )
+)
+
+solver = dict(
+    optim='SGD',
+    lr_scheduler='Cosine',
+    lr0=0.02, #0.01 # 0.02 
+    lrf=0.01,
+    momentum=0.937,
+    weight_decay=0.0005,
+    warmup_epochs=3.0,
+    warmup_momentum=0.8,
+    warmup_bias_lr=0.1
+)
+
+data_aug = dict(
+    hsv_h=0.015,
+    hsv_s=0.7,
+    hsv_v=0.4,
+    degrees=0.0,
+    translate=0.1,
+    scale=0.5,
+    shear=0.0,
+    flipud=0.0,
+    fliplr=0.5,
+    mosaic=1.0,
+    mixup=0.0,
+)
+# Choose Rep-block by the training Mode, choices=["repvgg", "hyper-search", "repopt"]
+training_mode='repopt'