exp : hpe model configs #17

AI/pose-estimation/mmpose/configs/_base_/default_runtime.py

@@ -4,13 +4,18 @@
     interval=50,
     hooks=[
         dict(type='TextLoggerHook'),
+        dict(type='WandbLoggerHook',
+             init_kwargs={
+                 'project': 'HPE modesl',
+                 'entity' : 'somang',
+                 'name' : 'mobile_net3_bottom_up'},
+            )
         # dict(type='TensorboardLoggerHook')
         # dict(type='PaviLoggerHook') # for internal services
     ])
 
 log_level = 'INFO'
 load_from = None
-resume_from = None
 dist_params = dict(backend='nccl')
 workflow = [('train', 1)]
 

AI/pose-estimation/mmpose/configs/_base_/default_runtime_cid_mobile.py

@@ -0,0 +1,25 @@
+checkpoint_config = dict(interval=10)
+
+log_config = dict(
+    interval=50,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        dict(type='WandbLoggerHook',
+             init_kwargs={
+                 'project': 'HPE modesl',
+                 'entity' : 'somang',
+                 'name' : 'CID_mobile'},
+            )
+        # dict(type='TensorboardLoggerHook')
+        # dict(type='PaviLoggerHook') # for internal services
+    ])
+
+log_level = 'INFO'
+load_from = None
+dist_params = dict(backend='nccl')
+workflow = [('train', 1)]
+
+# disable opencv multithreading to avoid system being overloaded
+opencv_num_threads = 0
+# set multi-process start method as `fork` to speed up the training
+mp_start_method = 'fork'

AI/pose-estimation/mmpose/configs/_base_/default_runtime_hrnet.py

@@ -0,0 +1,25 @@
+checkpoint_config = dict(interval=10)
+
+log_config = dict(
+    interval=50,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        dict(type='WandbLoggerHook',
+             init_kwargs={
+                 'project': 'HPE modesl',
+                 'entity' : 'somang',
+                 'name' : 'CID_hrnet_prun'},
+            )
+        # dict(type='TensorboardLoggerHook')
+        # dict(type='PaviLoggerHook') # for internal services
+    ])
+
+log_level = 'INFO'
+load_from = None
+dist_params = dict(backend='nccl')
+workflow = [('train', 1)]
+
+# disable opencv multithreading to avoid system being overloaded
+opencv_num_threads = 0
+# set multi-process start method as `fork` to speed up the training
+mp_start_method = 'fork'

AI/pose-estimation/mmpose/configs/_base_/default_runtime_hrnet_lite1.py

@@ -0,0 +1,25 @@
+checkpoint_config = dict(interval=1)
+
+log_config = dict(
+    interval=50,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        dict(type='WandbLoggerHook',
+             init_kwargs={
+                 'project': 'HPE modesl',
+                 'entity' : 'somang',
+                 'name' : 'CID_hrnet_lite1'},
+            )
+        # dict(type='TensorboardLoggerHook')
+        # dict(type='PaviLoggerHook') # for internal services
+    ])
+
+log_level = 'INFO'
+load_from = None
+dist_params = dict(backend='nccl')
+workflow = [('train', 1)]
+
+# disable opencv multithreading to avoid system being overloaded
+opencv_num_threads = 0
+# set multi-process start method as `fork` to speed up the training
+mp_start_method = 'fork'

AI/pose-estimation/mmpose/configs/myconfigs/cid_mobile_coco_512x512.py

@@ -0,0 +1,148 @@
+_base_ = [
+    '../_base_/default_runtime_cid_mobile.py',
+    '../_base_/datasets/coco.py'
+]
+checkpoint_config = dict(interval=20)
+evaluation = dict(interval=20, metric='mAP', save_best='AP')
+
+optimizer = dict(
+    type='Adam',
+    lr=0.001,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(policy='step', step=[90, 120])
+total_epochs = 140
+channel_cfg = dict(
+    dataset_joints=17,
+    dataset_channel=[
+        [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16
+    ])
+
+data_cfg = dict(
+    image_size=512,
+    base_size=256,
+    base_sigma=2,
+    heatmap_size=[128],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'],
+    num_scales=1,
+    scale_aware_sigma=False,
+    with_bbox=True,
+    use_nms=True,
+    soft_nms=False,
+    oks_thr=0.8,
+)
+
+# model settings
+model = dict(
+    type='CID',
+    pretrained='https://download.openmmlab.com/mmpose/'
+    'pretrain_models/hrnet_w32-36af842e.pth',
+    backbone=dict(
+        backbone=dict(type='MobileNetV2', widen_factor=1., out_indices=(7, )),
+    ),
+    keypoint_head=dict(
+        type='CIDHead',
+        in_channels=480,
+        gfd_channels=32,
+        num_joints=17,
+        multi_hm_loss_factor=1.0,
+        single_hm_loss_factor=4.0,
+        contrastive_loss_factor=1.0,
+        max_train_instances=200,
+        prior_prob=0.01),
+    train_cfg=dict(),
+    test_cfg=dict(
+        num_joints=channel_cfg['dataset_joints'],
+        flip_test=True,
+        max_num_people=30,
+        detection_threshold=0.01,
+        center_pool_kernel=3))
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(
+        type='BottomUpRandomAffine',
+        rot_factor=30,
+        scale_factor=[0.75, 1.5],
+        scale_type='short',
+        trans_factor=40),
+    dict(type='BottomUpRandomFlip', flip_prob=0.5),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='CIDGenerateTarget',
+        max_num_people=30,
+    ),
+    dict(
+        type='Collect',
+        keys=[
+            'img', 'multi_heatmap', 'multi_mask', 'instance_coord',
+            'instance_heatmap', 'instance_mask', 'instance_valid'
+        ],
+        meta_keys=[]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='BottomUpGetImgSize', test_scale_factor=[1]),
+    dict(
+        type='BottomUpResizeAlign',
+        transforms=[
+            dict(type='ToTensor'),
+            dict(
+                type='NormalizeTensor',
+                mean=[0.485, 0.456, 0.406],
+                std=[0.229, 0.224, 0.225]),
+        ]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'aug_data', 'test_scale_factor', 'base_size',
+            'center', 'scale', 'flip_index'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = '/shared/home/navermind/coco'
+
+data = dict(
+    workers_per_gpu=2,
+    train_dataloader=dict(samples_per_gpu=20),
+    val_dataloader=dict(samples_per_gpu=1),
+    test_dataloader=dict(samples_per_gpu=1),
+    train=dict(
+        type='BottomUpCocoDataset',
+        ann_file=f'{data_root}/annotations_trainval2017/annotations/person_keypoints_train2017.json',
+        img_prefix=f'{data_root}/train2017/train2017',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}
+    ),
+    val=dict(
+        type='BottomUpCocoDataset',
+        ann_file=f'{data_root}/annotations_trainval2017/annotations/person_keypoints_val2017.json',
+        img_prefix=f'{data_root}/val2017/val2017',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}
+    ),
+    test=dict(
+        type='BottomUpCocoDataset',
+        ann_file=f'{data_root}/annotations_trainval2017/annotations/person_keypoints_val2017.json',
+        img_prefix=f'{data_root}/val2017/val2017',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}
+    )
+)