Training resnet50 backbone on 2 GPUs

[lucasjinreal]

what's your loss will to be? I trained all loss to 1.0 but the result is totally wrong:

Also the category seems totally miss classified. Do u know which problem could be?

[lucasjinreal]

Configs changed for 2 GPUs:

data = dict(
    samples_per_gpu=2,
    workers_per_gpu=2,
    train=dict(
        type=dataset_type,
        ann_file=data_root + 'annotations/instances_train2017.json',
        img_prefix=data_root + 'train2017/',
        pipeline=train_pipeline),
    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
# 2 gpus, 0.01/4
optimizer = dict(type='SGD', lr=0.0025, momentum=0.9, weight_decay=0.0001)
optimizer_config = dict(grad_clip=dict(max_norm=35, norm_type=2))
# learning policy
lr_config = dict(
    policy='step',
    warmup='linear',
    warmup_iters=2000,
    warmup_ratio=1.0 / 3,
    step=[88, 132])
checkpoint_config = dict(interval=1)
# yapf:disable
log_config = dict(
    interval=50,
    hooks=[
        dict(type='TextLoggerHook'),
        dict(type='TensorboardLoggerHook')
    ])
# yapf:enable
# runtime settings
total_epochs = 164
# using python3 tools/train.py ./configs/solov2/solov2_r50_3x.py  --gpu-ids 1 2 to set
device_ids = range(2)
dist_params = dict(backend='nccl')
log_level = 'INFO'
work_dir = './work_dirs/solo_r50_3x'
load_from = None
resume_from = None
workflow = [('train', 1)]