79 lines
2.2 KiB
YAML

Collections:
- Name: SSD
Metadata:
Training Data: COCO
Training Techniques:
- SGD with Momentum
- Weight Decay
Training Resources: 8x V100 GPUs
Architecture:
- VGG
Paper:
URL: https://arxiv.org/abs/1512.02325
Title: 'SSD: Single Shot MultiBox Detector'
README: configs/ssd/README.md
Code:
URL: https://github.com/open-mmlab/mmdetection/blob/v2.14.0/mmdet/models/dense_heads/ssd_head.py#L16
Version: v2.14.0
Models:
- Name: ssd300_coco
In Collection: SSD
Config: configs/ssd/ssd300_coco.py
Metadata:
Training Memory (GB): 9.9
inference time (ms/im):
- value: 22.88
hardware: V100
backend: PyTorch
batch size: 1
mode: FP32
resolution: (300, 300)
Epochs: 120
Results:
- Task: Object Detection
Dataset: COCO
Metrics:
box AP: 25.5
Weights: https://download.openmmlab.com/mmdetection/v2.0/ssd/ssd300_coco/ssd300_coco_20210803_015428-d231a06e.pth
- Name: ssd512_coco
In Collection: SSD
Config: configs/ssd/ssd512_coco.py
Metadata:
Training Memory (GB): 19.4
inference time (ms/im):
- value: 32.57
hardware: V100
backend: PyTorch
batch size: 1
mode: FP32
resolution: (512, 512)
Epochs: 120
Results:
- Task: Object Detection
Dataset: COCO
Metrics:
box AP: 29.5
Weights: https://download.openmmlab.com/mmdetection/v2.0/ssd/ssd512_coco/ssd512_coco_20210803_022849-0a47a1ca.pth
- Name: ssdlite_mobilenetv2_scratch_600e_coco
In Collection: SSD
Config: configs/ssd/ssdlite_mobilenetv2_scratch_600e_coco.py
Metadata:
Training Memory (GB): 4.0
inference time (ms/im):
- value: 14.3
hardware: V100
backend: PyTorch
batch size: 1
mode: FP32
resolution: (320, 320)
Epochs: 600
Results:
- Task: Object Detection
Dataset: COCO
Metrics:
box AP: 21.3
Weights: https://download.openmmlab.com/mmdetection/v2.0/ssd/ssdlite_mobilenetv2_scratch_600e_coco/ssdlite_mobilenetv2_scratch_600e_coco_20210629_110627-974d9307.pth