# ssd_pytorch **Repository Path**: On_theway361/ssd_pytorch ## Basic Information - **Project Name**: ssd_pytorch - **Description**: this is pytorch version for ssd - **Primary Language**: Python - **License**: Not specified - **Default Branch**: master - **Homepage**: None - **GVP Project**: No ## Statistics - **Stars**: 1 - **Forks**: 0 - **Created**: 2019-07-09 - **Last Updated**: 2020-12-19 ## Categories & Tags **Categories**: Uncategorized **Tags**: None ## README # Single Shot MultiBox Detector Training With Pytorch # train parameter : ---------------------------------------------------------------- input_size = 300 | lr = 1e-3 | weight_decay = 5e-4 momentum = 0.9 | lr_gamma = 0.1 | lr_steps = [75, 98] batch_size = 32 | epochs = 108 | dataset = coco ---------------------------------------------------------------- epoch_id | mAP ---------------------------------------------------------------- 32 | 0.3468 ---------------------------------------------------------------- 50 | -- ---------------------------------------------------------------- 60 | 0.3886 ---------------------------------------------------------------- 70 | ---------------------------------------------------------------- 80 | 0.4224 ---------------------------------------------------------------- 90 | 0.4247 ---------------------------------------------------------------- 100 | 0.4278 ---------------------------------------------------------------- ---------------------------------------------------------------- input_size = 512 | lr = 1e-3 | weight_decay = 5e-4 momentum = 0.9 | lr_gamma = 0.1 | lr_steps = [75, 98] batch_size = 32 | epochs = 108 | dataset = coco ---------------------------------------------------------------- epoch_id | mAP ---------------------------------------------------------------- 32 | -- ---------------------------------------------------------------- 50 | -- ---------------------------------------------------------------- 60 | -- ---------------------------------------------------------------- 70 | -- ---------------------------------------------------------------- 80 | -- ---------------------------------------------------------------- 90 | -- ---------------------------------------------------------------- 100 | -- ----------------------------------------------------------------