关于精度损失

[goldlee]

很感谢作者开源这么牛的工作，另外请教下作者这个转换前后的精度损失大概是什么个水位，我转换了一个v3的检测模型，有些小一点的文字就检测不到，是不是我哪里操作的不对导致的呢？

[frotms]

是怎么转换的呢？几乎是没损失的

[goldlee]

我用的converter/ch_ppocr_v3_det_converter.py转换的检测模型，模型是这个：https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_distill_train.tar

[frotms]

发一下命令看看。

[goldlee]

命令：python -u converter/ch_ppocr_v3_det_converter.py

打印的日志信息

paddle: Student2.backbone.conv.conv.weight ---- (8, 3, 3, 3) paddle: Student.backbone.conv.conv.weight ---- (8, 3, 3, 3) paddle: Student2.backbone.conv.bn.weight ---- (8,) paddle: Student.backbone.conv.bn.weight ---- (8,) paddle: Student2.backbone.conv.bn.bias ---- (8,) paddle: Student.backbone.conv.bn.bias ---- (8,) paddle: Student2.backbone.conv.bn._mean ---- (8,) paddle: Student.backbone.conv.bn._mean ---- (8,) paddle: Student2.backbone.conv.bn._variance ---- (8,) paddle: Student.backbone.conv.bn._variance ---- (8,) paddle: Student2.backbone.stage0.0.expand_conv.conv.weight ---- (8, 8, 1, 1) paddle: Student.backbone.stage0.0.expand_conv.conv.weight ---- (8, 8, 1, 1) paddle: Student2.backbone.stage0.0.expand_conv.bn.weight ---- (8,) paddle: Student.backbone.stage0.0.expand_conv.bn.weight ---- (8,) paddle: Student2.backbone.stage0.0.expand_conv.bn.bias ---- (8,) paddle: Student.backbone.stage0.0.expand_conv.bn.bias ---- (8,) paddle: Student2.backbone.stage0.0.expand_conv.bn._mean ---- (8,) paddle: Student.backbone.stage0.0.expand_conv.bn._mean ---- (8,) paddle: Student2.backbone.stage0.0.expand_conv.bn._variance ---- (8,) paddle: Student.backbone.stage0.0.expand_conv.bn._variance ---- (8,) paddle: Student2.backbone.stage0.0.bottleneck_conv.conv.weight ---- (8, 1, 3, 3) paddle: Student.backbone.stage0.0.bottleneck_conv.conv.weight ---- (8, 1, 3, 3) paddle: Student2.backbone.stage0.0.bottleneck_conv.bn.weight ---- (8,) paddle: Student.backbone.stage0.0.bottleneck_conv.bn.weight ---- (8,) paddle: Student2.backbone.stage0.0.bottleneck_conv.bn.bias ---- (8,) paddle: Student.backbone.stage0.0.bottleneck_conv.bn.bias ---- (8,) paddle: Student2.backbone.stage0.0.bottleneck_conv.bn._mean ---- (8,) paddle: Student.backbone.stage0.0.bottleneck_conv.bn._mean ---- (8,) paddle: Student2.backbone.stage0.0.bottleneck_conv.bn._variance ---- (8,) paddle: Student.backbone.stage0.0.bottleneck_conv.bn._variance ---- (8,) paddle: Student2.backbone.stage0.0.linear_conv.conv.weight ---- (8, 8, 1, 1) paddle: Student.backbone.stage0.0.linear_conv.conv.weight ---- (8, 8, 1, 1) paddle: Student2.backbone.stage0.0.linear_conv.bn.weight ---- (8,) paddle: Student.backbone.stage0.0.linear_conv.bn.weight ---- (8,) paddle: Student2.backbone.stage0.0.linear_conv.bn.bias ---- (8,) paddle: Student.backbone.stage0.0.linear_conv.bn.bias ---- (8,) paddle: Student2.backbone.stage0.0.linear_conv.bn._mean ---- (8,) paddle: Student.backbone.stage0.0.linear_conv.bn._mean ---- (8,) paddle: Student2.backbone.stage0.0.linear_conv.bn._variance ---- (8,) paddle: Student.backbone.stage0.0.linear_conv.bn._variance ---- (8,) paddle: Student2.backbone.stage0.1.expand_conv.conv.weight ---- (32, 8, 1, 1) paddle: Student.backbone.stage0.1.expand_conv.conv.weight ---- (32, 8, 1, 1) paddle: Student2.backbone.stage0.1.expand_conv.bn.weight ---- (32,) paddle: Student.backbone.stage0.1.expand_conv.bn.weight ---- (32,) paddle: Student2.backbone.stage0.1.expand_conv.bn.bias ---- (32,) paddle: Student.backbone.stage0.1.expand_conv.bn.bias ---- (32,) paddle: Student2.backbone.stage0.1.expand_conv.bn._mean ---- (32,) paddle: Student.backbone.stage0.1.expand_conv.bn._mean ---- (32,) paddle: Student2.backbone.stage0.1.expand_conv.bn._variance ---- (32,) paddle: Student.backbone.stage0.1.expand_conv.bn._variance ---- (32,) paddle: Student2.backbone.stage0.1.bottleneck_conv.conv.weight ---- (32, 1, 3, 3) paddle: Student.backbone.stage0.1.bottleneck_conv.conv.weight ---- (32, 1, 3, 3) paddle: Student2.backbone.stage0.1.bottleneck_conv.bn.weight ---- (32,) paddle: Student.backbone.stage0.1.bottleneck_conv.bn.weight ---- (32,) paddle: Student2.backbone.stage0.1.bottleneck_conv.bn.bias ---- (32,) paddle: Student.backbone.stage0.1.bottleneck_conv.bn.bias ---- (32,) paddle: Student2.backbone.stage0.1.bottleneck_conv.bn._mean ---- (32,) paddle: Student.backbone.stage0.1.bottleneck_conv.bn._mean ---- (32,) paddle: Student2.backbone.stage0.1.bottleneck_conv.bn._variance ---- (32,) paddle: Student.backbone.stage0.1.bottleneck_conv.bn._variance ---- (32,) paddle: Student2.backbone.stage0.1.linear_conv.conv.weight ---- (16, 32, 1, 1) paddle: Student.backbone.stage0.1.linear_conv.conv.weight ---- (16, 32, 1, 1) paddle: Student2.backbone.stage0.1.linear_conv.bn.weight ---- (16,) paddle: Student.backbone.stage0.1.linear_conv.bn.weight ---- (16,) paddle: Student2.backbone.stage0.1.linear_conv.bn.bias ---- (16,) paddle: Student.backbone.stage0.1.linear_conv.bn.bias ---- (16,) paddle: Student2.backbone.stage0.1.linear_conv.bn._mean ---- (16,) paddle: Student.backbone.stage0.1.linear_conv.bn._mean ---- (16,) paddle: Student2.backbone.stage0.1.linear_conv.bn._variance ---- (16,) paddle: Student.backbone.stage0.1.linear_conv.bn._variance ---- (16,) paddle: Student2.backbone.stage0.2.expand_conv.conv.weight ---- (40, 16, 1, 1) paddle: Student.backbone.stage0.2.expand_conv.conv.weight ---- (40, 16, 1, 1) paddle: Student2.backbone.stage0.2.expand_conv.bn.weight ---- (40,) paddle: Student.backbone.stage0.2.expand_conv.bn.weight ---- (40,) paddle: Student2.backbone.stage0.2.expand_conv.bn.bias ---- (40,) paddle: Student.backbone.stage0.2.expand_conv.bn.bias ---- (40,) paddle: Student2.backbone.stage0.2.expand_conv.bn._mean ---- (40,) paddle: Student.backbone.stage0.2.expand_conv.bn._mean ---- (40,) paddle: Student2.backbone.stage0.2.expand_conv.bn._variance ---- (40,) paddle: Student.backbone.stage0.2.expand_conv.bn._variance ---- (40,) paddle: Student2.backbone.stage0.2.bottleneck_conv.conv.weight ---- (40, 1, 3, 3) paddle: Student.backbone.stage0.2.bottleneck_conv.conv.weight ---- (40, 1, 3, 3) paddle: Student2.backbone.stage0.2.bottleneck_conv.bn.weight ---- (40,) paddle: Student.backbone.stage0.2.bottleneck_conv.bn.weight ---- (40,) paddle: Student2.backbone.stage0.2.bottleneck_conv.bn.bias ---- (40,) paddle: Student.backbone.stage0.2.bottleneck_conv.bn.bias ---- (40,) paddle: Student2.backbone.stage0.2.bottleneck_conv.bn._mean ---- (40,) paddle: Student.backbone.stage0.2.bottleneck_conv.bn._mean ---- (40,) paddle: Student2.backbone.stage0.2.bottleneck_conv.bn._variance ---- (40,) paddle: Student.backbone.stage0.2.bottleneck_conv.bn._variance ---- (40,) paddle: Student2.backbone.stage0.2.linear_conv.conv.weight ---- (16, 40, 1, 1) paddle: Student.backbone.stage0.2.linear_conv.conv.weight ---- (16, 40, 1, 1) paddle: Student2.backbone.stage0.2.linear_conv.bn.weight ---- (16,) paddle: Student.backbone.stage0.2.linear_conv.bn.weight ---- (16,) paddle: Student2.backbone.stage0.2.linear_conv.bn.bias ---- (16,) paddle: Student.backbone.stage0.2.linear_conv.bn.bias ---- (16,) paddle: Student2.backbone.stage0.2.linear_conv.bn._mean ---- (16,) paddle: Student.backbone.stage0.2.linear_conv.bn._mean ---- (16,) paddle: Student2.backbone.stage0.2.linear_conv.bn._variance ---- (16,) paddle: Student.backbone.stage0.2.linear_conv.bn._variance ---- (16,) paddle: Student2.backbone.stage1.0.expand_conv.conv.weight ---- (40, 16, 1, 1) paddle: Student.backbone.stage1.0.expand_conv.conv.weight ---- (40, 16, 1, 1) paddle: Student2.backbone.stage1.0.expand_conv.bn.weight ---- (40,) paddle: Student.backbone.stage1.0.expand_conv.bn.weight ---- (40,) paddle: Student2.backbone.stage1.0.expand_conv.bn.bias ---- (40,) paddle: Student.backbone.stage1.0.expand_conv.bn.bias ---- (40,) paddle: Student2.backbone.stage1.0.expand_conv.bn._mean ---- (40,) paddle: Student.backbone.stage1.0.expand_conv.bn._mean ---- (40,) paddle: Student2.backbone.stage1.0.expand_conv.bn._variance ---- (40,) paddle: Student.backbone.stage1.0.expand_conv.bn._variance ---- (40,) paddle: Student2.backbone.stage1.0.bottleneck_conv.conv.weight ---- (40, 1, 5, 5) paddle: Student.backbone.stage1.0.bottleneck_conv.conv.weight ---- (40, 1, 5, 5) paddle: Student2.backbone.stage1.0.bottleneck_conv.bn.weight ---- (40,) paddle: Student.backbone.stage1.0.bottleneck_conv.bn.weight ---- (40,) paddle: Student2.backbone.stage1.0.bottleneck_conv.bn.bias ---- (40,) paddle: Student.backbone.stage1.0.bottleneck_conv.bn.bias ---- (40,) paddle: Student2.backbone.stage1.0.bottleneck_conv.bn._mean ---- (40,) paddle: Student.backbone.stage1.0.bottleneck_conv.bn._mean ---- (40,) paddle: Student2.backbone.stage1.0.bottleneck_conv.bn._variance ---- (40,) paddle: Student.backbone.stage1.0.bottleneck_conv.bn._variance ---- (40,) paddle: Student2.backbone.stage1.0.linear_conv.conv.weight ---- (24, 40, 1, 1) paddle: Student.backbone.stage1.0.linear_conv.conv.weight ---- (24, 40, 1, 1) paddle: Student2.backbone.stage1.0.linear_conv.bn.weight ---- (24,) paddle: Student.backbone.stage1.0.linear_conv.bn.weight ---- (24,) paddle: Student2.backbone.stage1.0.linear_conv.bn.bias ---- (24,) paddle: Student.backbone.stage1.0.linear_conv.bn.bias ---- (24,) paddle: Student2.backbone.stage1.0.linear_conv.bn._mean ---- (24,) paddle: Student.backbone.stage1.0.linear_conv.bn._mean ---- (24,) paddle: Student2.backbone.stage1.0.linear_conv.bn._variance ---- (24,) paddle: Student.backbone.stage1.0.linear_conv.bn._variance ---- (24,) paddle: Student2.backbone.stage1.1.expand_conv.conv.weight ---- (64, 24, 1, 1) paddle: Student.backbone.stage1.1.expand_conv.conv.weight ---- (64, 24, 1, 1) paddle: Student2.backbone.stage1.1.expand_conv.bn.weight ---- (64,) paddle: Student.backbone.stage1.1.expand_conv.bn.weight ---- (64,) paddle: Student2.backbone.stage1.1.expand_conv.bn.bias ---- (64,) paddle: Student.backbone.stage1.1.expand_conv.bn.bias ---- (64,) paddle: Student2.backbone.stage1.1.expand_conv.bn._mean ---- (64,) paddle: Student.backbone.stage1.1.expand_conv.bn._mean ---- (64,) paddle: Student2.backbone.stage1.1.expand_conv.bn._variance ---- (64,) paddle: Student.backbone.stage1.1.expand_conv.bn._variance ---- (64,) paddle: Student2.backbone.stage1.1.bottleneck_conv.conv.weight ---- (64, 1, 5, 5) paddle: Student.backbone.stage1.1.bottleneck_conv.conv.weight ---- (64, 1, 5, 5) paddle: Student2.backbone.stage1.1.bottleneck_conv.bn.weight ---- (64,) paddle: Student.backbone.stage1.1.bottleneck_conv.bn.weight ---- (64,) paddle: Student2.backbone.stage1.1.bottleneck_conv.bn.bias ---- (64,) paddle: Student.backbone.stage1.1.bottleneck_conv.bn.bias ---- (64,) paddle: Student2.backbone.stage1.1.bottleneck_conv.bn._mean ---- (64,) paddle: Student.backbone.stage1.1.bottleneck_conv.bn._mean ---- (64,) paddle: Student2.backbone.stage1.1.bottleneck_conv.bn._variance ---- (64,) paddle: Student.backbone.stage1.1.bottleneck_conv.bn._variance ---- (64,) paddle: Student2.backbone.stage1.1.linear_conv.conv.weight ---- (24, 64, 1, 1) paddle: Student.backbone.stage1.1.linear_conv.conv.weight ---- (24, 64, 1, 1) paddle: Student2.backbone.stage1.1.linear_conv.bn.weight ---- (24,) paddle: Student.backbone.stage1.1.linear_conv.bn.weight ---- (24,) paddle: Student2.backbone.stage1.1.linear_conv.bn.bias ---- (24,) paddle: Student.backbone.stage1.1.linear_conv.bn.bias ---- (24,) paddle: Student2.backbone.stage1.1.linear_conv.bn._mean ---- (24,) paddle: Student.backbone.stage1.1.linear_conv.bn._mean ---- (24,) paddle: Student2.backbone.stage1.1.linear_conv.bn._variance ---- (24,) paddle: Student.backbone.stage1.1.linear_conv.bn._variance ---- (24,) paddle: Student2.backbone.stage1.2.expand_conv.conv.weight ---- (64, 24, 1, 1) paddle: Student.backbone.stage1.2.expand_conv.conv.weight ---- (64, 24, 1, 1) paddle: Student2.backbone.stage1.2.expand_conv.bn.weight ---- (64,) paddle: Student.backbone.stage1.2.expand_conv.bn.weight ---- (64,) paddle: Student2.backbone.stage1.2.expand_conv.bn.bias ---- (64,) paddle: Student.backbone.stage1.2.expand_conv.bn.bias ---- (64,) paddle: Student2.backbone.stage1.2.expand_conv.bn._mean ---- (64,) paddle: Student.backbone.stage1.2.expand_conv.bn._mean ---- (64,) paddle: Student2.backbone.stage1.2.expand_conv.bn._variance ---- (64,) paddle: Student.backbone.stage1.2.expand_conv.bn._variance ---- (64,) paddle: Student2.backbone.stage1.2.bottleneck_conv.conv.weight ---- (64, 1, 5, 5) paddle: Student.backbone.stage1.2.bottleneck_conv.conv.weight ---- (64, 1, 5, 5) paddle: Student2.backbone.stage1.2.bottleneck_conv.bn.weight ---- (64,) paddle: Student.backbone.stage1.2.bottleneck_conv.bn.weight ---- (64,) paddle: Student2.backbone.stage1.2.bottleneck_conv.bn.bias ---- (64,) paddle: Student.backbone.stage1.2.bottleneck_conv.bn.bias ---- (64,) paddle: Student2.backbone.stage1.2.bottleneck_conv.bn._mean ---- (64,) paddle: Student.backbone.stage1.2.bottleneck_conv.bn._mean ---- (64,) paddle: Student2.backbone.stage1.2.bottleneck_conv.bn._variance ---- (64,) paddle: Student.backbone.stage1.2.bottleneck_conv.bn._variance ---- (64,) paddle: Student2.backbone.stage1.2.linear_conv.conv.weight ---- (24, 64, 1, 1) paddle: Student.backbone.stage1.2.linear_conv.conv.weight ---- (24, 64, 1, 1) paddle: Student2.backbone.stage1.2.linear_conv.bn.weight ---- (24,) paddle: Student.backbone.stage1.2.linear_conv.bn.weight ---- (24,) paddle: Student2.backbone.stage1.2.linear_conv.bn.bias ---- (24,) paddle: Student.backbone.stage1.2.linear_conv.bn.bias ---- (24,) paddle: Student2.backbone.stage1.2.linear_conv.bn._mean ---- (24,) paddle: Student.backbone.stage1.2.linear_conv.bn._mean ---- (24,) paddle: Student2.backbone.stage1.2.linear_conv.bn._variance ---- (24,) paddle: Student.backbone.stage1.2.linear_conv.bn._variance ---- (24,) paddle: Student2.backbone.stage2.0.expand_conv.conv.weight ---- (120, 24, 1, 1) paddle: Student.backbone.stage2.0.expand_conv.conv.weight ---- (120, 24, 1, 1) paddle: Student2.backbone.stage2.0.expand_conv.bn.weight ---- (120,) paddle: Student.backbone.stage2.0.expand_conv.bn.weight ---- (120,) paddle: Student2.backbone.stage2.0.expand_conv.bn.bias ---- (120,) paddle: Student.backbone.stage2.0.expand_conv.bn.bias ---- (120,) paddle: Student2.backbone.stage2.0.expand_conv.bn._mean ---- (120,) paddle: Student.backbone.stage2.0.expand_conv.bn._mean ---- (120,) paddle: Student2.backbone.stage2.0.expand_conv.bn._variance ---- (120,) paddle: Student.backbone.stage2.0.expand_conv.bn._variance ---- (120,) paddle: Student2.backbone.stage2.0.bottleneck_conv.conv.weight ---- (120, 1, 3, 3) paddle: Student.backbone.stage2.0.bottleneck_conv.conv.weight ---- (120, 1, 3, 3) paddle: Student2.backbone.stage2.0.bottleneck_conv.bn.weight ---- (120,) paddle: Student.backbone.stage2.0.bottleneck_conv.bn.weight ---- (120,) paddle: Student2.backbone.stage2.0.bottleneck_conv.bn.bias ---- (120,) paddle: Student.backbone.stage2.0.bottleneck_conv.bn.bias ---- (120,) paddle: Student2.backbone.stage2.0.bottleneck_conv.bn._mean ---- (120,) paddle: Student.backbone.stage2.0.bottleneck_conv.bn._mean ---- (120,) paddle: Student2.backbone.stage2.0.bottleneck_conv.bn._variance ---- (120,) paddle: Student.backbone.stage2.0.bottleneck_conv.bn._variance ---- (120,) paddle: Student2.backbone.stage2.0.linear_conv.conv.weight ---- (40, 120, 1, 1) paddle: Student.backbone.stage2.0.linear_conv.conv.weight ---- (40, 120, 1, 1) paddle: Student2.backbone.stage2.0.linear_conv.bn.weight ---- (40,) paddle: Student.backbone.stage2.0.linear_conv.bn.weight ---- (40,) paddle: Student2.backbone.stage2.0.linear_conv.bn.bias ---- (40,) paddle: Student.backbone.stage2.0.linear_conv.bn.bias ---- (40,) paddle: Student2.backbone.stage2.0.linear_conv.bn._mean ---- (40,) paddle: Student.backbone.stage2.0.linear_conv.bn._mean ---- (40,) paddle: Student2.backbone.stage2.0.linear_conv.bn._variance ---- (40,) paddle: Student.backbone.stage2.0.linear_conv.bn._variance ---- (40,) paddle: Student2.backbone.stage2.1.expand_conv.conv.weight ---- (104, 40, 1, 1) paddle: Student.backbone.stage2.1.expand_conv.conv.weight ---- (104, 40, 1, 1) paddle: Student2.backbone.stage2.1.expand_conv.bn.weight ---- (104,) paddle: Student.backbone.stage2.1.expand_conv.bn.weight ---- (104,) paddle: Student2.backbone.stage2.1.expand_conv.bn.bias ---- (104,) paddle: Student.backbone.stage2.1.expand_conv.bn.bias ---- (104,) paddle: Student2.backbone.stage2.1.expand_conv.bn._mean ---- (104,) paddle: Student.backbone.stage2.1.expand_conv.bn._mean ---- (104,) paddle: Student2.backbone.stage2.1.expand_conv.bn._variance ---- (104,) paddle: Student.backbone.stage2.1.expand_conv.bn._variance ---- (104,) paddle: Student2.backbone.stage2.1.bottleneck_conv.conv.weight ---- (104, 1, 3, 3) paddle: Student.backbone.stage2.1.bottleneck_conv.conv.weight ---- (104, 1, 3, 3) paddle: Student2.backbone.stage2.1.bottleneck_conv.bn.weight ---- (104,) paddle: Student.backbone.stage2.1.bottleneck_conv.bn.weight ---- (104,) paddle: Student2.backbone.stage2.1.bottleneck_conv.bn.bias ---- (104,) paddle: Student.backbone.stage2.1.bottleneck_conv.bn.bias ---- (104,) paddle: Student2.backbone.stage2.1.bottleneck_conv.bn._mean ---- (104,) paddle: Student.backbone.stage2.1.bottleneck_conv.bn._mean ---- (104,) paddle: Student2.backbone.stage2.1.bottleneck_conv.bn._variance ---- (104,) paddle: Student.backbone.stage2.1.bottleneck_conv.bn._variance ---- (104,) paddle: Student2.backbone.stage2.1.linear_conv.conv.weight ---- (40, 104, 1, 1) paddle: Student.backbone.stage2.1.linear_conv.conv.weight ---- (40, 104, 1, 1) paddle: Student2.backbone.stage2.1.linear_conv.bn.weight ---- (40,) paddle: Student.backbone.stage2.1.linear_conv.bn.weight ---- (40,) paddle: Student2.backbone.stage2.1.linear_conv.bn.bias ---- (40,) paddle: Student.backbone.stage2.1.linear_conv.bn.bias ---- (40,) paddle: Student2.backbone.stage2.1.linear_conv.bn._mean ---- (40,) paddle: Student.backbone.stage2.1.linear_conv.bn._mean ---- (40,) paddle: Student2.backbone.stage2.1.linear_conv.bn._variance ---- (40,) paddle: Student.backbone.stage2.1.linear_conv.bn._variance ---- (40,) paddle: Student2.backbone.stage2.2.expand_conv.conv.weight ---- (96, 40, 1, 1) paddle: Student.backbone.stage2.2.expand_conv.conv.weight ---- (96, 40, 1, 1) paddle: Student2.backbone.stage2.2.expand_conv.bn.weight ---- (96,) paddle: Student.backbone.stage2.2.expand_conv.bn.weight ---- (96,) paddle: Student2.backbone.stage2.2.expand_conv.bn.bias ---- (96,) paddle: Student.backbone.stage2.2.expand_conv.bn.bias ---- (96,) paddle: Student2.backbone.stage2.2.expand_conv.bn._mean ---- (96,) paddle: Student.backbone.stage2.2.expand_conv.bn._mean ---- (96,) paddle: Student2.backbone.stage2.2.expand_conv.bn._variance ---- (96,) paddle: Student.backbone.stage2.2.expand_conv.bn._variance ---- (96,) paddle: Student2.backbone.stage2.2.bottleneck_conv.conv.weight ---- (96, 1, 3, 3) paddle: Student.backbone.stage2.2.bottleneck_conv.conv.weight ---- (96, 1, 3, 3) paddle: Student2.backbone.stage2.2.bottleneck_conv.bn.weight ---- (96,) paddle: Student.backbone.stage2.2.bottleneck_conv.bn.weight ---- (96,) paddle: Student2.backbone.stage2.2.bottleneck_conv.bn.bias ---- (96,) paddle: Student.backbone.stage2.2.bottleneck_conv.bn.bias ---- (96,) paddle: Student2.backbone.stage2.2.bottleneck_conv.bn._mean ---- (96,) paddle: Student.backbone.stage2.2.bottleneck_conv.bn._mean ---- (96,) paddle: Student2.backbone.stage2.2.bottleneck_conv.bn._variance ---- (96,) paddle: Student.backbone.stage2.2.bottleneck_conv.bn._variance ---- (96,) paddle: Student2.backbone.stage2.2.linear_conv.conv.weight ---- (40, 96, 1, 1) paddle: Student.backbone.stage2.2.linear_conv.conv.weight ---- (40, 96, 1, 1) paddle: Student2.backbone.stage2.2.linear_conv.bn.weight ---- (40,) paddle: Student.backbone.stage2.2.linear_conv.bn.weight ---- (40,) paddle: Student2.backbone.stage2.2.linear_conv.bn.bias ---- (40,) paddle: Student.backbone.stage2.2.linear_conv.bn.bias ---- (40,) paddle: Student2.backbone.stage2.2.linear_conv.bn._mean ---- (40,) paddle: Student.backbone.stage2.2.linear_conv.bn._mean ---- (40,) paddle: Student2.backbone.stage2.2.linear_conv.bn._variance ---- (40,) paddle: Student.backbone.stage2.2.linear_conv.bn._variance ---- (40,) paddle: Student2.backbone.stage2.3.expand_conv.conv.weight ---- (96, 40, 1, 1) paddle: Student.backbone.stage2.3.expand_conv.conv.weight ---- (96, 40, 1, 1) paddle: Student2.backbone.stage2.3.expand_conv.bn.weight ---- (96,) paddle: Student.backbone.stage2.3.expand_conv.bn.weight ---- (96,) paddle: Student2.backbone.stage2.3.expand_conv.bn.bias ---- (96,) paddle: Student.backbone.stage2.3.expand_conv.bn.bias ---- (96,) paddle: Student2.backbone.stage2.3.expand_conv.bn._mean ---- (96,) paddle: Student.backbone.stage2.3.expand_conv.bn._mean ---- (96,) paddle: Student2.backbone.stage2.3.expand_conv.bn._variance ---- (96,) paddle: Student.backbone.stage2.3.expand_conv.bn._variance ---- (96,) paddle: Student2.backbone.stage2.3.bottleneck_conv.conv.weight ---- (96, 1, 3, 3) paddle: Student.backbone.stage2.3.bottleneck_conv.conv.weight ---- (96, 1, 3, 3) paddle: Student2.backbone.stage2.3.bottleneck_conv.bn.weight ---- (96,) paddle: Student.backbone.stage2.3.bottleneck_conv.bn.weight ---- (96,) paddle: Student2.backbone.stage2.3.bottleneck_conv.bn.bias ---- (96,) paddle: Student.backbone.stage2.3.bottleneck_conv.bn.bias ---- (96,) paddle: Student2.backbone.stage2.3.bottleneck_conv.bn._mean ---- (96,) paddle: Student.backbone.stage2.3.bottleneck_conv.bn._mean ---- (96,) paddle: Student2.backbone.stage2.3.bottleneck_conv.bn._variance ---- (96,) paddle: Student.backbone.stage2.3.bottleneck_conv.bn._variance ---- (96,) paddle: Student2.backbone.stage2.3.linear_conv.conv.weight ---- (40, 96, 1, 1) paddle: Student.backbone.stage2.3.linear_conv.conv.weight ---- (40, 96, 1, 1) paddle: Student2.backbone.stage2.3.linear_conv.bn.weight ---- (40,) paddle: Student.backbone.stage2.3.linear_conv.bn.weight ---- (40,) paddle: Student2.backbone.stage2.3.linear_conv.bn.bias ---- (40,) paddle: Student.backbone.stage2.3.linear_conv.bn.bias ---- (40,) paddle: Student2.backbone.stage2.3.linear_conv.bn._mean ---- (40,) paddle: Student.backbone.stage2.3.linear_conv.bn._mean ---- (40,) paddle: Student2.backbone.stage2.3.linear_conv.bn._variance ---- (40,) paddle: Student.backbone.stage2.3.linear_conv.bn._variance ---- (40,) paddle: Student2.backbone.stage2.4.expand_conv.conv.weight ---- (240, 40, 1, 1) paddle: Student.backbone.stage2.4.expand_conv.conv.weight ---- (240, 40, 1, 1) paddle: Student2.backbone.stage2.4.expand_conv.bn.weight ---- (240,) paddle: Student.backbone.stage2.4.expand_conv.bn.weight ---- (240,) paddle: Student2.backbone.stage2.4.expand_conv.bn.bias ---- (240,) paddle: Student.backbone.stage2.4.expand_conv.bn.bias ---- (240,) paddle: Student2.backbone.stage2.4.expand_conv.bn._mean ---- (240,) paddle: Student.backbone.stage2.4.expand_conv.bn._mean ---- (240,) paddle: Student2.backbone.stage2.4.expand_conv.bn._variance ---- (240,) paddle: Student.backbone.stage2.4.expand_conv.bn._variance ---- (240,) paddle: Student2.backbone.stage2.4.bottleneck_conv.conv.weight ---- (240, 1, 3, 3) paddle: Student.backbone.stage2.4.bottleneck_conv.conv.weight ---- (240, 1, 3, 3) paddle: Student2.backbone.stage2.4.bottleneck_conv.bn.weight ---- (240,) paddle: Student.backbone.stage2.4.bottleneck_conv.bn.weight ---- (240,) paddle: Student2.backbone.stage2.4.bottleneck_conv.bn.bias ---- (240,) paddle: Student.backbone.stage2.4.bottleneck_conv.bn.bias ---- (240,) paddle: Student2.backbone.stage2.4.bottleneck_conv.bn._mean ---- (240,) paddle: Student.backbone.stage2.4.bottleneck_conv.bn._mean ---- (240,) paddle: Student2.backbone.stage2.4.bottleneck_conv.bn._variance ---- (240,) paddle: Student.backbone.stage2.4.bottleneck_conv.bn._variance ---- (240,) paddle: Student2.backbone.stage2.4.linear_conv.conv.weight ---- (56, 240, 1, 1) paddle: Student.backbone.stage2.4.linear_conv.conv.weight ---- (56, 240, 1, 1) paddle: Student2.backbone.stage2.4.linear_conv.bn.weight ---- (56,) paddle: Student.backbone.stage2.4.linear_conv.bn.weight ---- (56,) paddle: Student2.backbone.stage2.4.linear_conv.bn.bias ---- (56,) paddle: Student.backbone.stage2.4.linear_conv.bn.bias ---- (56,) paddle: Student2.backbone.stage2.4.linear_conv.bn._mean ---- (56,) paddle: Student.backbone.stage2.4.linear_conv.bn._mean ---- (56,) paddle: Student2.backbone.stage2.4.linear_conv.bn._variance ---- (56,) paddle: Student.backbone.stage2.4.linear_conv.bn._variance ---- (56,) paddle: Student2.backbone.stage2.5.expand_conv.conv.weight ---- (336, 56, 1, 1) paddle: Student.backbone.stage2.5.expand_conv.conv.weight ---- (336, 56, 1, 1) paddle: Student2.backbone.stage2.5.expand_conv.bn.weight ---- (336,) paddle: Student.backbone.stage2.5.expand_conv.bn.weight ---- (336,) paddle: Student2.backbone.stage2.5.expand_conv.bn.bias ---- (336,) paddle: Student.backbone.stage2.5.expand_conv.bn.bias ---- (336,) paddle: Student2.backbone.stage2.5.expand_conv.bn._mean ---- (336,) paddle: Student.backbone.stage2.5.expand_conv.bn._mean ---- (336,) paddle: Student2.backbone.stage2.5.expand_conv.bn._variance ---- (336,) paddle: Student.backbone.stage2.5.expand_conv.bn._variance ---- (336,) paddle: Student2.backbone.stage2.5.bottleneck_conv.conv.weight ---- (336, 1, 3, 3) paddle: Student.backbone.stage2.5.bottleneck_conv.conv.weight ---- (336, 1, 3, 3) paddle: Student2.backbone.stage2.5.bottleneck_conv.bn.weight ---- (336,) paddle: Student.backbone.stage2.5.bottleneck_conv.bn.weight ---- (336,) paddle: Student2.backbone.stage2.5.bottleneck_conv.bn.bias ---- (336,) paddle: Student.backbone.stage2.5.bottleneck_conv.bn.bias ---- (336,) paddle: Student2.backbone.stage2.5.bottleneck_conv.bn._mean ---- (336,) paddle: Student.backbone.stage2.5.bottleneck_conv.bn._mean ---- (336,) paddle: Student2.backbone.stage2.5.bottleneck_conv.bn._variance ---- (336,) paddle: Student.backbone.stage2.5.bottleneck_conv.bn._variance ---- (336,) paddle: Student2.backbone.stage2.5.linear_conv.conv.weight ---- (56, 336, 1, 1) paddle: Student.backbone.stage2.5.linear_conv.conv.weight ---- (56, 336, 1, 1) paddle: Student2.backbone.stage2.5.linear_conv.bn.weight ---- (56,) paddle: Student.backbone.stage2.5.linear_conv.bn.weight ---- (56,) paddle: Student2.backbone.stage2.5.linear_conv.bn.bias ---- (56,) paddle: Student.backbone.stage2.5.linear_conv.bn.bias ---- (56,) paddle: Student2.backbone.stage2.5.linear_conv.bn._mean ---- (56,) paddle: Student.backbone.stage2.5.linear_conv.bn._mean ---- (56,) paddle: Student2.backbone.stage2.5.linear_conv.bn._variance ---- (56,) paddle: Student.backbone.stage2.5.linear_conv.bn._variance ---- (56,) paddle: Student2.backbone.stage3.0.expand_conv.conv.weight ---- (336, 56, 1, 1) paddle: Student.backbone.stage3.0.expand_conv.conv.weight ---- (336, 56, 1, 1) paddle: Student2.backbone.stage3.0.expand_conv.bn.weight ---- (336,) paddle: Student.backbone.stage3.0.expand_conv.bn.weight ---- (336,) paddle: Student2.backbone.stage3.0.expand_conv.bn.bias ---- (336,) paddle: Student.backbone.stage3.0.expand_conv.bn.bias ---- (336,) paddle: Student2.backbone.stage3.0.expand_conv.bn._mean ---- (336,) paddle: Student.backbone.stage3.0.expand_conv.bn._mean ---- (336,) paddle: Student2.backbone.stage3.0.expand_conv.bn._variance ---- (336,) paddle: Student.backbone.stage3.0.expand_conv.bn._variance ---- (336,) paddle: Student2.backbone.stage3.0.bottleneck_conv.conv.weight ---- (336, 1, 5, 5) paddle: Student.backbone.stage3.0.bottleneck_conv.conv.weight ---- (336, 1, 5, 5) paddle: Student2.backbone.stage3.0.bottleneck_conv.bn.weight ---- (336,) paddle: Student.backbone.stage3.0.bottleneck_conv.bn.weight ---- (336,) paddle: Student2.backbone.stage3.0.bottleneck_conv.bn.bias ---- (336,) paddle: Student.backbone.stage3.0.bottleneck_conv.bn.bias ---- (336,) paddle: Student2.backbone.stage3.0.bottleneck_conv.bn._mean ---- (336,) paddle: Student.backbone.stage3.0.bottleneck_conv.bn._mean ---- (336,) paddle: Student2.backbone.stage3.0.bottleneck_conv.bn._variance ---- (336,) paddle: Student.backbone.stage3.0.bottleneck_conv.bn._variance ---- (336,) paddle: Student2.backbone.stage3.0.linear_conv.conv.weight ---- (80, 336, 1, 1) paddle: Student.backbone.stage3.0.linear_conv.conv.weight ---- (80, 336, 1, 1) paddle: Student2.backbone.stage3.0.linear_conv.bn.weight ---- (80,) paddle: Student.backbone.stage3.0.linear_conv.bn.weight ---- (80,) paddle: Student2.backbone.stage3.0.linear_conv.bn.bias ---- (80,) paddle: Student.backbone.stage3.0.linear_conv.bn.bias ---- (80,) paddle: Student2.backbone.stage3.0.linear_conv.bn._mean ---- (80,) paddle: Student.backbone.stage3.0.linear_conv.bn._mean ---- (80,) paddle: Student2.backbone.stage3.0.linear_conv.bn._variance ---- (80,) paddle: Student.backbone.stage3.0.linear_conv.bn._variance ---- (80,) paddle: Student2.backbone.stage3.1.expand_conv.conv.weight ---- (480, 80, 1, 1) paddle: Student.backbone.stage3.1.expand_conv.conv.weight ---- (480, 80, 1, 1) paddle: Student2.backbone.stage3.1.expand_conv.bn.weight ---- (480,) paddle: Student.backbone.stage3.1.expand_conv.bn.weight ---- (480,) paddle: Student2.backbone.stage3.1.expand_conv.bn.bias ---- (480,) paddle: Student.backbone.stage3.1.expand_conv.bn.bias ---- (480,) paddle: Student2.backbone.stage3.1.expand_conv.bn._mean ---- (480,) paddle: Student.backbone.stage3.1.expand_conv.bn._mean ---- (480,) paddle: Student2.backbone.stage3.1.expand_conv.bn._variance ---- (480,) paddle: Student.backbone.stage3.1.expand_conv.bn._variance ---- (480,) paddle: Student2.backbone.stage3.1.bottleneck_conv.conv.weight ---- (480, 1, 5, 5) paddle: Student.backbone.stage3.1.bottleneck_conv.conv.weight ---- (480, 1, 5, 5) paddle: Student2.backbone.stage3.1.bottleneck_conv.bn.weight ---- (480,) paddle: Student.backbone.stage3.1.bottleneck_conv.bn.weight ---- (480,) paddle: Student2.backbone.stage3.1.bottleneck_conv.bn.bias ---- (480,) paddle: Student.backbone.stage3.1.bottleneck_conv.bn.bias ---- (480,) paddle: Student2.backbone.stage3.1.bottleneck_conv.bn._mean ---- (480,) paddle: Student.backbone.stage3.1.bottleneck_conv.bn._mean ---- (480,) paddle: Student2.backbone.stage3.1.bottleneck_conv.bn._variance ---- (480,) paddle: Student.backbone.stage3.1.bottleneck_conv.bn._variance ---- (480,) paddle: Student2.backbone.stage3.1.linear_conv.conv.weight ---- (80, 480, 1, 1) paddle: Student.backbone.stage3.1.linear_conv.conv.weight ---- (80, 480, 1, 1) paddle: Student2.backbone.stage3.1.linear_conv.bn.weight ---- (80,) paddle: Student.backbone.stage3.1.linear_conv.bn.weight ---- (80,) paddle: Student2.backbone.stage3.1.linear_conv.bn.bias ---- (80,) paddle: Student.backbone.stage3.1.linear_conv.bn.bias ---- (80,) paddle: Student2.backbone.stage3.1.linear_conv.bn._mean ---- (80,) paddle: Student.backbone.stage3.1.linear_conv.bn._mean ---- (80,) paddle: Student2.backbone.stage3.1.linear_conv.bn._variance ---- (80,) paddle: Student.backbone.stage3.1.linear_conv.bn._variance ---- (80,) paddle: Student2.backbone.stage3.2.expand_conv.conv.weight ---- (480, 80, 1, 1) paddle: Student.backbone.stage3.2.expand_conv.conv.weight ---- (480, 80, 1, 1) paddle: Student2.backbone.stage3.2.expand_conv.bn.weight ---- (480,) paddle: Student.backbone.stage3.2.expand_conv.bn.weight ---- (480,) paddle: Student2.backbone.stage3.2.expand_conv.bn.bias ---- (480,) paddle: Student.backbone.stage3.2.expand_conv.bn.bias ---- (480,) paddle: Student2.backbone.stage3.2.expand_conv.bn._mean ---- (480,) paddle: Student.backbone.stage3.2.expand_conv.bn._mean ---- (480,) paddle: Student2.backbone.stage3.2.expand_conv.bn._variance ---- (480,) paddle: Student.backbone.stage3.2.expand_conv.bn._variance ---- (480,) paddle: Student2.backbone.stage3.2.bottleneck_conv.conv.weight ---- (480, 1, 5, 5) paddle: Student.backbone.stage3.2.bottleneck_conv.conv.weight ---- (480, 1, 5, 5) paddle: Student2.backbone.stage3.2.bottleneck_conv.bn.weight ---- (480,) paddle: Student.backbone.stage3.2.bottleneck_conv.bn.weight ---- (480,) paddle: Student2.backbone.stage3.2.bottleneck_conv.bn.bias ---- (480,) paddle: Student.backbone.stage3.2.bottleneck_conv.bn.bias ---- (480,) paddle: Student2.backbone.stage3.2.bottleneck_conv.bn._mean ---- (480,) paddle: Student.backbone.stage3.2.bottleneck_conv.bn._mean ---- (480,) paddle: Student2.backbone.stage3.2.bottleneck_conv.bn._variance ---- (480,) paddle: Student.backbone.stage3.2.bottleneck_conv.bn._variance ---- (480,) paddle: Student2.backbone.stage3.2.linear_conv.conv.weight ---- (80, 480, 1, 1) paddle: Student.backbone.stage3.2.linear_conv.conv.weight ---- (80, 480, 1, 1) paddle: Student2.backbone.stage3.2.linear_conv.bn.weight ---- (80,) paddle: Student.backbone.stage3.2.linear_conv.bn.weight ---- (80,) paddle: Student2.backbone.stage3.2.linear_conv.bn.bias ---- (80,) paddle: Student.backbone.stage3.2.linear_conv.bn.bias ---- (80,) paddle: Student2.backbone.stage3.2.linear_conv.bn._mean ---- (80,) paddle: Student.backbone.stage3.2.linear_conv.bn._mean ---- (80,) paddle: Student2.backbone.stage3.2.linear_conv.bn._variance ---- (80,) paddle: Student.backbone.stage3.2.linear_conv.bn._variance ---- (80,) paddle: Student2.backbone.stage3.3.conv.weight ---- (480, 80, 1, 1) paddle: Student.backbone.stage3.3.conv.weight ---- (480, 80, 1, 1) paddle: Student2.backbone.stage3.3.bn.weight ---- (480,) paddle: Student.backbone.stage3.3.bn.weight ---- (480,) paddle: Student2.backbone.stage3.3.bn.bias ---- (480,) paddle: Student.backbone.stage3.3.bn.bias ---- (480,) paddle: Student2.backbone.stage3.3.bn._mean ---- (480,) paddle: Student.backbone.stage3.3.bn._mean ---- (480,) paddle: Student2.backbone.stage3.3.bn._variance ---- (480,) paddle: Student.backbone.stage3.3.bn._variance ---- (480,) paddle: Student2.neck.ins_conv.0.in_conv.weight ---- (96, 16, 1, 1) paddle: Student.neck.ins_conv.0.in_conv.weight ---- (96, 16, 1, 1) paddle: Student2.neck.ins_conv.0.se_block.conv1.weight ---- (24, 96, 1, 1) paddle: Student.neck.ins_conv.0.se_block.conv1.weight ---- (24, 96, 1, 1) paddle: Student2.neck.ins_conv.0.se_block.conv1.bias ---- (24,) paddle: Student.neck.ins_conv.0.se_block.conv1.bias ---- (24,) paddle: Student2.neck.ins_conv.0.se_block.conv2.weight ---- (96, 24, 1, 1) paddle: Student.neck.ins_conv.0.se_block.conv2.weight ---- (96, 24, 1, 1) paddle: Student2.neck.ins_conv.0.se_block.conv2.bias ---- (96,) paddle: Student.neck.ins_conv.0.se_block.conv2.bias ---- (96,) paddle: Student2.neck.ins_conv.1.in_conv.weight ---- (96, 24, 1, 1) paddle: Student.neck.ins_conv.1.in_conv.weight ---- (96, 24, 1, 1) paddle: Student2.neck.ins_conv.1.se_block.conv1.weight ---- (24, 96, 1, 1) paddle: Student.neck.ins_conv.1.se_block.conv1.weight ---- (24, 96, 1, 1) paddle: Student2.neck.ins_conv.1.se_block.conv1.bias ---- (24,) paddle: Student.neck.ins_conv.1.se_block.conv1.bias ---- (24,) paddle: Student2.neck.ins_conv.1.se_block.conv2.weight ---- (96, 24, 1, 1) paddle: Student.neck.ins_conv.1.se_block.conv2.weight ---- (96, 24, 1, 1) paddle: Student2.neck.ins_conv.1.se_block.conv2.bias ---- (96,) paddle: Student.neck.ins_conv.1.se_block.conv2.bias ---- (96,) paddle: Student2.neck.ins_conv.2.in_conv.weight ---- (96, 56, 1, 1) paddle: Student.neck.ins_conv.2.in_conv.weight ---- (96, 56, 1, 1) paddle: Student2.neck.ins_conv.2.se_block.conv1.weight ---- (24, 96, 1, 1) paddle: Student.neck.ins_conv.2.se_block.conv1.weight ---- (24, 96, 1, 1) paddle: Student2.neck.ins_conv.2.se_block.conv1.bias ---- (24,) paddle: Student.neck.ins_conv.2.se_block.conv1.bias ---- (24,) paddle: Student2.neck.ins_conv.2.se_block.conv2.weight ---- (96, 24, 1, 1) paddle: Student.neck.ins_conv.2.se_block.conv2.weight ---- (96, 24, 1, 1) paddle: Student2.neck.ins_conv.2.se_block.conv2.bias ---- (96,) paddle: Student.neck.ins_conv.2.se_block.conv2.bias ---- (96,) paddle: Student2.neck.ins_conv.3.in_conv.weight ---- (96, 480, 1, 1) paddle: Student.neck.ins_conv.3.in_conv.weight ---- (96, 480, 1, 1) paddle: Student2.neck.ins_conv.3.se_block.conv1.weight ---- (24, 96, 1, 1) paddle: Student.neck.ins_conv.3.se_block.conv1.weight ---- (24, 96, 1, 1) paddle: Student2.neck.ins_conv.3.se_block.conv1.bias ---- (24,) paddle: Student.neck.ins_conv.3.se_block.conv1.bias ---- (24,) paddle: Student2.neck.ins_conv.3.se_block.conv2.weight ---- (96, 24, 1, 1) paddle: Student.neck.ins_conv.3.se_block.conv2.weight ---- (96, 24, 1, 1) paddle: Student2.neck.ins_conv.3.se_block.conv2.bias ---- (96,) paddle: Student.neck.ins_conv.3.se_block.conv2.bias ---- (96,) paddle: Student2.neck.inp_conv.0.in_conv.weight ---- (24, 96, 3, 3) paddle: Student.neck.inp_conv.0.in_conv.weight ---- (24, 96, 3, 3) paddle: Student2.neck.inp_conv.0.se_block.conv1.weight ---- (6, 24, 1, 1) paddle: Student.neck.inp_conv.0.se_block.conv1.weight ---- (6, 24, 1, 1) paddle: Student2.neck.inp_conv.0.se_block.conv1.bias ---- (6,) paddle: Student.neck.inp_conv.0.se_block.conv1.bias ---- (6,) paddle: Student2.neck.inp_conv.0.se_block.conv2.weight ---- (24, 6, 1, 1) paddle: Student.neck.inp_conv.0.se_block.conv2.weight ---- (24, 6, 1, 1) paddle: Student2.neck.inp_conv.0.se_block.conv2.bias ---- (24,) paddle: Student.neck.inp_conv.0.se_block.conv2.bias ---- (24,) paddle: Student2.neck.inp_conv.1.in_conv.weight ---- (24, 96, 3, 3) paddle: Student.neck.inp_conv.1.in_conv.weight ---- (24, 96, 3, 3) paddle: Student2.neck.inp_conv.1.se_block.conv1.weight ---- (6, 24, 1, 1) paddle: Student.neck.inp_conv.1.se_block.conv1.weight ---- (6, 24, 1, 1) paddle: Student2.neck.inp_conv.1.se_block.conv1.bias ---- (6,) paddle: Student.neck.inp_conv.1.se_block.conv1.bias ---- (6,) paddle: Student2.neck.inp_conv.1.se_block.conv2.weight ---- (24, 6, 1, 1) paddle: Student.neck.inp_conv.1.se_block.conv2.weight ---- (24, 6, 1, 1) paddle: Student2.neck.inp_conv.1.se_block.conv2.bias ---- (24,) paddle: Student.neck.inp_conv.1.se_block.conv2.bias ---- (24,) paddle: Student2.neck.inp_conv.2.in_conv.weight ---- (24, 96, 3, 3) paddle: Student.neck.inp_conv.2.in_conv.weight ---- (24, 96, 3, 3) paddle: Student2.neck.inp_conv.2.se_block.conv1.weight ---- (6, 24, 1, 1) paddle: Student.neck.inp_conv.2.se_block.conv1.weight ---- (6, 24, 1, 1) paddle: Student2.neck.inp_conv.2.se_block.conv1.bias ---- (6,) paddle: Student.neck.inp_conv.2.se_block.conv1.bias ---- (6,) paddle: Student2.neck.inp_conv.2.se_block.conv2.weight ---- (24, 6, 1, 1) paddle: Student.neck.inp_conv.2.se_block.conv2.weight ---- (24, 6, 1, 1) paddle: Student2.neck.inp_conv.2.se_block.conv2.bias ---- (24,) paddle: Student.neck.inp_conv.2.se_block.conv2.bias ---- (24,) paddle: Student2.neck.inp_conv.3.in_conv.weight ---- (24, 96, 3, 3) paddle: Student.neck.inp_conv.3.in_conv.weight ---- (24, 96, 3, 3) paddle: Student2.neck.inp_conv.3.se_block.conv1.weight ---- (6, 24, 1, 1) paddle: Student.neck.inp_conv.3.se_block.conv1.weight ---- (6, 24, 1, 1) paddle: Student2.neck.inp_conv.3.se_block.conv1.bias ---- (6,) paddle: Student.neck.inp_conv.3.se_block.conv1.bias ---- (6,) paddle: Student2.neck.inp_conv.3.se_block.conv2.weight ---- (24, 6, 1, 1) paddle: Student.neck.inp_conv.3.se_block.conv2.weight ---- (24, 6, 1, 1) paddle: Student2.neck.inp_conv.3.se_block.conv2.bias ---- (24,) paddle: Student.neck.inp_conv.3.se_block.conv2.bias ---- (24,) paddle: Student2.head.binarize.conv1.weight ---- (24, 96, 3, 3) paddle: Student.head.binarize.conv1.weight ---- (24, 96, 3, 3) paddle: Student2.head.binarize.conv_bn1.weight ---- (24,) paddle: Student.head.binarize.conv_bn1.weight ---- (24,) paddle: Student2.head.binarize.conv_bn1.bias ---- (24,) paddle: Student.head.binarize.conv_bn1.bias ---- (24,) paddle: Student2.head.binarize.conv_bn1._mean ---- (24,) paddle: Student.head.binarize.conv_bn1._mean ---- (24,) paddle: Student2.head.binarize.conv_bn1._variance ---- (24,) paddle: Student.head.binarize.conv_bn1._variance ---- (24,) paddle: Student2.head.binarize.conv2.weight ---- (24, 24, 2, 2) paddle: Student.head.binarize.conv2.weight ---- (24, 24, 2, 2) paddle: Student2.head.binarize.conv2.bias ---- (24,) paddle: Student.head.binarize.conv2.bias ---- (24,) paddle: Student2.head.binarize.conv_bn2.weight ---- (24,) paddle: Student.head.binarize.conv_bn2.weight ---- (24,) paddle: Student2.head.binarize.conv_bn2.bias ---- (24,) paddle: Student.head.binarize.conv_bn2.bias ---- (24,) paddle: Student2.head.binarize.conv_bn2._mean ---- (24,) paddle: Student.head.binarize.conv_bn2._mean ---- (24,) paddle: Student2.head.binarize.conv_bn2._variance ---- (24,) paddle: Student.head.binarize.conv_bn2._variance ---- (24,) paddle: Student2.head.binarize.conv3.weight ---- (24, 1, 2, 2) paddle: Student.head.binarize.conv3.weight ---- (24, 1, 2, 2) paddle: Student2.head.binarize.conv3.bias ---- (1,) paddle: Student.head.binarize.conv3.bias ---- (1,) paddle: Student2.head.thresh.conv1.weight ---- (24, 96, 3, 3) paddle: Student.head.thresh.conv1.weight ---- (24, 96, 3, 3) paddle: Student2.head.thresh.conv_bn1.weight ---- (24,) paddle: Student.head.thresh.conv_bn1.weight ---- (24,) paddle: Student2.head.thresh.conv_bn1.bias ---- (24,) paddle: Student.head.thresh.conv_bn1.bias ---- (24,) paddle: Student2.head.thresh.conv_bn1._mean ---- (24,) paddle: Student.head.thresh.conv_bn1._mean ---- (24,) paddle: Student2.head.thresh.conv_bn1._variance ---- (24,) paddle: Student.head.thresh.conv_bn1._variance ---- (24,) paddle: Student2.head.thresh.conv2.weight ---- (24, 24, 2, 2) paddle: Student.head.thresh.conv2.weight ---- (24, 24, 2, 2) paddle: Student2.head.thresh.conv2.bias ---- (24,) paddle: Student.head.thresh.conv2.bias ---- (24,) paddle: Student2.head.thresh.conv_bn2.weight ---- (24,) paddle: Student.head.thresh.conv_bn2.weight ---- (24,) paddle: Student2.head.thresh.conv_bn2.bias ---- (24,) paddle: Student.head.thresh.conv_bn2.bias ---- (24,) paddle: Student2.head.thresh.conv_bn2._mean ---- (24,) paddle: Student.head.thresh.conv_bn2._mean ---- (24,) paddle: Student2.head.thresh.conv_bn2._variance ---- (24,) paddle: Student.head.thresh.conv_bn2._variance ---- (24,) paddle: Student2.head.thresh.conv3.weight ---- (24, 1, 2, 2) paddle: Student.head.thresh.conv3.weight ---- (24, 1, 2, 2) paddle: Student2.head.thresh.conv3.bias ---- (1,) paddle: Student.head.thresh.conv3.bias ---- (1,) paddle: Teacher.backbone.conv1_1._conv.weight ---- (32, 3, 3, 3) paddle: Teacher.backbone.conv1_1._batch_norm.weight ---- (32,) paddle: Teacher.backbone.conv1_1._batch_norm.bias ---- (32,) paddle: Teacher.backbone.conv1_1._batch_norm._mean ---- (32,) paddle: Teacher.backbone.conv1_1._batch_norm._variance ---- (32,) paddle: Teacher.backbone.conv1_2._conv.weight ---- (32, 32, 3, 3) paddle: Teacher.backbone.conv1_2._batch_norm.weight ---- (32,) paddle: Teacher.backbone.conv1_2._batch_norm.bias ---- (32,) paddle: Teacher.backbone.conv1_2._batch_norm._mean ---- (32,) paddle: Teacher.backbone.conv1_2._batch_norm._variance ---- (32,) paddle: Teacher.backbone.conv1_3._conv.weight ---- (64, 32, 3, 3) paddle: Teacher.backbone.conv1_3._batch_norm.weight ---- (64,) paddle: Teacher.backbone.conv1_3._batch_norm.bias ---- (64,) paddle: Teacher.backbone.conv1_3._batch_norm._mean ---- (64,) paddle: Teacher.backbone.conv1_3._batch_norm._variance ---- (64,) paddle: Teacher.backbone.bb_0_0.conv0._conv.weight ---- (64, 64, 1, 1) paddle: Teacher.backbone.bb_0_0.conv0._batch_norm.weight ---- (64,) paddle: Teacher.backbone.bb_0_0.conv0._batch_norm.bias ---- (64,) paddle: Teacher.backbone.bb_0_0.conv0._batch_norm._mean ---- (64,) paddle: Teacher.backbone.bb_0_0.conv0._batch_norm._variance ---- (64,) paddle: Teacher.backbone.bb_0_0.conv1._conv.weight ---- (64, 64, 3, 3) paddle: Teacher.backbone.bb_0_0.conv1._batch_norm.weight ---- (64,) paddle: Teacher.backbone.bb_0_0.conv1._batch_norm.bias ---- (64,) paddle: Teacher.backbone.bb_0_0.conv1._batch_norm._mean ---- (64,) paddle: Teacher.backbone.bb_0_0.conv1._batch_norm._variance ---- (64,) paddle: Teacher.backbone.bb_0_0.conv2._conv.weight ---- (256, 64, 1, 1) paddle: Teacher.backbone.bb_0_0.conv2._batch_norm.weight ---- (256,) paddle: Teacher.backbone.bb_0_0.conv2._batch_norm.bias ---- (256,) paddle: Teacher.backbone.bb_0_0.conv2._batch_norm._mean ---- (256,) paddle: Teacher.backbone.bb_0_0.conv2._batch_norm._variance ---- (256,) paddle: Teacher.backbone.bb_0_0.short._conv.weight ---- (256, 64, 1, 1) paddle: Teacher.backbone.bb_0_0.short._batch_norm.weight ---- (256,) paddle: Teacher.backbone.bb_0_0.short._batch_norm.bias ---- (256,) paddle: Teacher.backbone.bb_0_0.short._batch_norm._mean ---- (256,) paddle: Teacher.backbone.bb_0_0.short._batch_norm._variance ---- (256,) paddle: Teacher.backbone.bb_0_1.conv0._conv.weight ---- (64, 256, 1, 1) paddle: Teacher.backbone.bb_0_1.conv0._batch_norm.weight ---- (64,) paddle: Teacher.backbone.bb_0_1.conv0._batch_norm.bias ---- (64,) paddle: Teacher.backbone.bb_0_1.conv0._batch_norm._mean ---- (64,) paddle: Teacher.backbone.bb_0_1.conv0._batch_norm._variance ---- (64,) paddle: Teacher.backbone.bb_0_1.conv1._conv.weight ---- (64, 64, 3, 3) paddle: Teacher.backbone.bb_0_1.conv1._batch_norm.weight ---- (64,) paddle: Teacher.backbone.bb_0_1.conv1._batch_norm.bias ---- (64,) paddle: Teacher.backbone.bb_0_1.conv1._batch_norm._mean ---- (64,) paddle: Teacher.backbone.bb_0_1.conv1._batch_norm._variance ---- (64,) paddle: Teacher.backbone.bb_0_1.conv2._conv.weight ---- (256, 64, 1, 1) paddle: Teacher.backbone.bb_0_1.conv2._batch_norm.weight ---- (256,) paddle: Teacher.backbone.bb_0_1.conv2._batch_norm.bias ---- (256,) paddle: Teacher.backbone.bb_0_1.conv2._batch_norm._mean ---- (256,) paddle: Teacher.backbone.bb_0_1.conv2._batch_norm._variance ---- (256,) paddle: Teacher.backbone.bb_0_2.conv0._conv.weight ---- (64, 256, 1, 1) paddle: Teacher.backbone.bb_0_2.conv0._batch_norm.weight ---- (64,) paddle: Teacher.backbone.bb_0_2.conv0._batch_norm.bias ---- (64,) paddle: Teacher.backbone.bb_0_2.conv0._batch_norm._mean ---- (64,) paddle: Teacher.backbone.bb_0_2.conv0._batch_norm._variance ---- (64,) paddle: Teacher.backbone.bb_0_2.conv1._conv.weight ---- (64, 64, 3, 3) paddle: Teacher.backbone.bb_0_2.conv1._batch_norm.weight ---- (64,) paddle: Teacher.backbone.bb_0_2.conv1._batch_norm.bias ---- (64,) paddle: Teacher.backbone.bb_0_2.conv1._batch_norm._mean ---- (64,) paddle: Teacher.backbone.bb_0_2.conv1._batch_norm._variance ---- (64,) paddle: Teacher.backbone.bb_0_2.conv2._conv.weight ---- (256, 64, 1, 1) paddle: Teacher.backbone.bb_0_2.conv2._batch_norm.weight ---- (256,) paddle: Teacher.backbone.bb_0_2.conv2._batch_norm.bias ---- (256,) paddle: Teacher.backbone.bb_0_2.conv2._batch_norm._mean ---- (256,) paddle: Teacher.backbone.bb_0_2.conv2._batch_norm._variance ---- (256,) paddle: Teacher.backbone.bb_1_0.conv0._conv.weight ---- (128, 256, 1, 1) paddle: Teacher.backbone.bb_1_0.conv0._batch_norm.weight ---- (128,) paddle: Teacher.backbone.bb_1_0.conv0._batch_norm.bias ---- (128,) paddle: Teacher.backbone.bb_1_0.conv0._batch_norm._mean ---- (128,) paddle: Teacher.backbone.bb_1_0.conv0._batch_norm._variance ---- (128,) paddle: Teacher.backbone.bb_1_0.conv1._conv.weight ---- (128, 128, 3, 3) paddle: Teacher.backbone.bb_1_0.conv1._batch_norm.weight ---- (128,) paddle: Teacher.backbone.bb_1_0.conv1._batch_norm.bias ---- (128,) paddle: Teacher.backbone.bb_1_0.conv1._batch_norm._mean ---- (128,) paddle: Teacher.backbone.bb_1_0.conv1._batch_norm._variance ---- (128,) paddle: Teacher.backbone.bb_1_0.conv2._conv.weight ---- (512, 128, 1, 1) paddle: Teacher.backbone.bb_1_0.conv2._batch_norm.weight ---- (512,) paddle: Teacher.backbone.bb_1_0.conv2._batch_norm.bias ---- (512,) paddle: Teacher.backbone.bb_1_0.conv2._batch_norm._mean ---- (512,) paddle: Teacher.backbone.bb_1_0.conv2._batch_norm._variance ---- (512,) paddle: Teacher.backbone.bb_1_0.short._conv.weight ---- (512, 256, 1, 1) paddle: Teacher.backbone.bb_1_0.short._batch_norm.weight ---- (512,) paddle: Teacher.backbone.bb_1_0.short._batch_norm.bias ---- (512,) paddle: Teacher.backbone.bb_1_0.short._batch_norm._mean ---- (512,) paddle: Teacher.backbone.bb_1_0.short._batch_norm._variance ---- (512,) paddle: Teacher.backbone.bb_1_1.conv0._conv.weight ---- (128, 512, 1, 1) paddle: Teacher.backbone.bb_1_1.conv0._batch_norm.weight ---- (128,) paddle: Teacher.backbone.bb_1_1.conv0._batch_norm.bias ---- (128,) paddle: Teacher.backbone.bb_1_1.conv0._batch_norm._mean ---- (128,) paddle: Teacher.backbone.bb_1_1.conv0._batch_norm._variance ---- (128,) paddle: Teacher.backbone.bb_1_1.conv1._conv.weight ---- (128, 128, 3, 3) paddle: Teacher.backbone.bb_1_1.conv1._batch_norm.weight ---- (128,) paddle: Teacher.backbone.bb_1_1.conv1._batch_norm.bias ---- (128,) paddle: Teacher.backbone.bb_1_1.conv1._batch_norm._mean ---- (128,) paddle: Teacher.backbone.bb_1_1.conv1._batch_norm._variance ---- (128,) paddle: Teacher.backbone.bb_1_1.conv2._conv.weight ---- (512, 128, 1, 1) paddle: Teacher.backbone.bb_1_1.conv2._batch_norm.weight ---- (512,) paddle: Teacher.backbone.bb_1_1.conv2._batch_norm.bias ---- (512,) paddle: Teacher.backbone.bb_1_1.conv2._batch_norm._mean ---- (512,) paddle: Teacher.backbone.bb_1_1.conv2._batch_norm._variance ---- (512,) paddle: Teacher.backbone.bb_1_2.conv0._conv.weight ---- (128, 512, 1, 1) paddle: Teacher.backbone.bb_1_2.conv0._batch_norm.weight ---- (128,) paddle: Teacher.backbone.bb_1_2.conv0._batch_norm.bias ---- (128,) paddle: Teacher.backbone.bb_1_2.conv0._batch_norm._mean ---- (128,) paddle: Teacher.backbone.bb_1_2.conv0._batch_norm._variance ---- (128,) paddle: Teacher.backbone.bb_1_2.conv1._conv.weight ---- (128, 128, 3, 3) paddle: Teacher.backbone.bb_1_2.conv1._batch_norm.weight ---- (128,) paddle: Teacher.backbone.bb_1_2.conv1._batch_norm.bias ---- (128,) paddle: Teacher.backbone.bb_1_2.conv1._batch_norm._mean ---- (128,) paddle: Teacher.backbone.bb_1_2.conv1._batch_norm._variance ---- (128,) paddle: Teacher.backbone.bb_1_2.conv2._conv.weight ---- (512, 128, 1, 1) paddle: Teacher.backbone.bb_1_2.conv2._batch_norm.weight ---- (512,) paddle: Teacher.backbone.bb_1_2.conv2._batch_norm.bias ---- (512,) paddle: Teacher.backbone.bb_1_2.conv2._batch_norm._mean ---- (512,) paddle: Teacher.backbone.bb_1_2.conv2._batch_norm._variance ---- (512,) paddle: Teacher.backbone.bb_1_3.conv0._conv.weight ---- (128, 512, 1, 1) paddle: Teacher.backbone.bb_1_3.conv0._batch_norm.weight ---- (128,) paddle: Teacher.backbone.bb_1_3.conv0._batch_norm.bias ---- (128,) paddle: Teacher.backbone.bb_1_3.conv0._batch_norm._mean ---- (128,) paddle: Teacher.backbone.bb_1_3.conv0._batch_norm._variance ---- (128,) paddle: Teacher.backbone.bb_1_3.conv1._conv.weight ---- (128, 128, 3, 3) paddle: Teacher.backbone.bb_1_3.conv1._batch_norm.weight ---- (128,) paddle: Teacher.backbone.bb_1_3.conv1._batch_norm.bias ---- (128,) paddle: Teacher.backbone.bb_1_3.conv1._batch_norm._mean ---- (128,) paddle: Teacher.backbone.bb_1_3.conv1._batch_norm._variance ---- (128,) paddle: Teacher.backbone.bb_1_3.conv2._conv.weight ---- (512, 128, 1, 1) paddle: Teacher.backbone.bb_1_3.conv2._batch_norm.weight ---- (512,) paddle: Teacher.backbone.bb_1_3.conv2._batch_norm.bias ---- (512,) paddle: Teacher.backbone.bb_1_3.conv2._batch_norm._mean ---- (512,) paddle: Teacher.backbone.bb_1_3.conv2._batch_norm._variance ---- (512,) paddle: Teacher.backbone.bb_2_0.conv0._conv.weight ---- (256, 512, 1, 1) paddle: Teacher.backbone.bb_2_0.conv0._batch_norm.weight ---- (256,) paddle: Teacher.backbone.bb_2_0.conv0._batch_norm.bias ---- (256,) paddle: Teacher.backbone.bb_2_0.conv0._batch_norm._mean ---- (256,) paddle: Teacher.backbone.bb_2_0.conv0._batch_norm._variance ---- (256,) paddle: Teacher.backbone.bb_2_0.conv1._conv.weight ---- (256, 256, 3, 3) paddle: Teacher.backbone.bb_2_0.conv1._batch_norm.weight ---- (256,) paddle: Teacher.backbone.bb_2_0.conv1._batch_norm.bias ---- (256,) paddle: Teacher.backbone.bb_2_0.conv1._batch_norm._mean ---- (256,) paddle: Teacher.backbone.bb_2_0.conv1._batch_norm._variance ---- (256,) paddle: Teacher.backbone.bb_2_0.conv2._conv.weight ---- (1024, 256, 1, 1) paddle: Teacher.backbone.bb_2_0.conv2._batch_norm.weight ---- (1024,) paddle: Teacher.backbone.bb_2_0.conv2._batch_norm.bias ---- (1024,) paddle: Teacher.backbone.bb_2_0.conv2._batch_norm._mean ---- (1024,) paddle: Teacher.backbone.bb_2_0.conv2._batch_norm._variance ---- (1024,) paddle: Teacher.backbone.bb_2_0.short._conv.weight ---- (1024, 512, 1, 1) paddle: Teacher.backbone.bb_2_0.short._batch_norm.weight ---- (1024,) paddle: Teacher.backbone.bb_2_0.short._batch_norm.bias ---- (1024,) paddle: Teacher.backbone.bb_2_0.short._batch_norm._mean ---- (1024,) paddle: Teacher.backbone.bb_2_0.short._batch_norm._variance ---- (1024,) paddle: Teacher.backbone.bb_2_1.conv0._conv.weight ---- (256, 1024, 1, 1) paddle: Teacher.backbone.bb_2_1.conv0._batch_norm.weight ---- (256,) paddle: Teacher.backbone.bb_2_1.conv0._batch_norm.bias ---- (256,) paddle: Teacher.backbone.bb_2_1.conv0._batch_norm._mean ---- (256,) paddle: Teacher.backbone.bb_2_1.conv0._batch_norm._variance ---- (256,) paddle: Teacher.backbone.bb_2_1.conv1._conv.weight ---- (256, 256, 3, 3) paddle: Teacher.backbone.bb_2_1.conv1._batch_norm.weight ---- (256,) paddle: Teacher.backbone.bb_2_1.conv1._batch_norm.bias ---- (256,) paddle: Teacher.backbone.bb_2_1.conv1._batch_norm._mean ---- (256,) paddle: Teacher.backbone.bb_2_1.conv1._batch_norm._variance ---- (256,) paddle: Teacher.backbone.bb_2_1.conv2._conv.weight ---- (1024, 256, 1, 1) paddle: Teacher.backbone.bb_2_1.conv2._batch_norm.weight ---- (1024,) paddle: Teacher.backbone.bb_2_1.conv2._batch_norm.bias ---- (1024,) paddle: Teacher.backbone.bb_2_1.conv2._batch_norm._mean ---- (1024,) paddle: Teacher.backbone.bb_2_1.conv2._batch_norm._variance ---- (1024,) paddle: Teacher.backbone.bb_2_2.conv0._conv.weight ---- (256, 1024, 1, 1) paddle: Teacher.backbone.bb_2_2.conv0._batch_norm.weight ---- (256,) paddle: Teacher.backbone.bb_2_2.conv0._batch_norm.bias ---- (256,) paddle: Teacher.backbone.bb_2_2.conv0._batch_norm._mean ---- (256,) paddle: Teacher.backbone.bb_2_2.conv0._batch_norm._variance ---- (256,) paddle: Teacher.backbone.bb_2_2.conv1._conv.weight ---- (256, 256, 3, 3) paddle: Teacher.backbone.bb_2_2.conv1._batch_norm.weight ---- (256,) paddle: Teacher.backbone.bb_2_2.conv1._batch_norm.bias ---- (256,) paddle: Teacher.backbone.bb_2_2.conv1._batch_norm._mean ---- (256,) paddle: Teacher.backbone.bb_2_2.conv1._batch_norm._variance ---- (256,) paddle: Teacher.backbone.bb_2_2.conv2._conv.weight ---- (1024, 256, 1, 1) paddle: Teacher.backbone.bb_2_2.conv2._batch_norm.weight ---- (1024,) paddle: Teacher.backbone.bb_2_2.conv2._batch_norm.bias ---- (1024,) paddle: Teacher.backbone.bb_2_2.conv2._batch_norm._mean ---- (1024,) paddle: Teacher.backbone.bb_2_2.conv2._batch_norm._variance ---- (1024,) paddle: Teacher.backbone.bb_2_3.conv0._conv.weight ---- (256, 1024, 1, 1) paddle: Teacher.backbone.bb_2_3.conv0._batch_norm.weight ---- (256,) paddle: Teacher.backbone.bb_2_3.conv0._batch_norm.bias ---- (256,) paddle: Teacher.backbone.bb_2_3.conv0._batch_norm._mean ---- (256,) paddle: Teacher.backbone.bb_2_3.conv0._batch_norm._variance ---- (256,) paddle: Teacher.backbone.bb_2_3.conv1._conv.weight ---- (256, 256, 3, 3) paddle: Teacher.backbone.bb_2_3.conv1._batch_norm.weight ---- (256,) paddle: Teacher.backbone.bb_2_3.conv1._batch_norm.bias ---- (256,) paddle: Teacher.backbone.bb_2_3.conv1._batch_norm._mean ---- (256,) paddle: Teacher.backbone.bb_2_3.conv1._batch_norm._variance ---- (256,) paddle: Teacher.backbone.bb_2_3.conv2._conv.weight ---- (1024, 256, 1, 1) paddle: Teacher.backbone.bb_2_3.conv2._batch_norm.weight ---- (1024,) paddle: Teacher.backbone.bb_2_3.conv2._batch_norm.bias ---- (1024,) paddle: Teacher.backbone.bb_2_3.conv2._batch_norm._mean ---- (1024,) paddle: Teacher.backbone.bb_2_3.conv2._batch_norm._variance ---- (1024,) paddle: Teacher.backbone.bb_2_4.conv0._conv.weight ---- (256, 1024, 1, 1) paddle: Teacher.backbone.bb_2_4.conv0._batch_norm.weight ---- (256,) paddle: Teacher.backbone.bb_2_4.conv0._batch_norm.bias ---- (256,) paddle: Teacher.backbone.bb_2_4.conv0._batch_norm._mean ---- (256,) paddle: Teacher.backbone.bb_2_4.conv0._batch_norm._variance ---- (256,) paddle: Teacher.backbone.bb_2_4.conv1._conv.weight ---- (256, 256, 3, 3) paddle: Teacher.backbone.bb_2_4.conv1._batch_norm.weight ---- (256,) paddle: Teacher.backbone.bb_2_4.conv1._batch_norm.bias ---- (256,) paddle: Teacher.backbone.bb_2_4.conv1._batch_norm._mean ---- (256,) paddle: Teacher.backbone.bb_2_4.conv1._batch_norm._variance ---- (256,) paddle: Teacher.backbone.bb_2_4.conv2._conv.weight ---- (1024, 256, 1, 1) paddle: Teacher.backbone.bb_2_4.conv2._batch_norm.weight ---- (1024,) paddle: Teacher.backbone.bb_2_4.conv2._batch_norm.bias ---- (1024,) paddle: Teacher.backbone.bb_2_4.conv2._batch_norm._mean ---- (1024,) paddle: Teacher.backbone.bb_2_4.conv2._batch_norm._variance ---- (1024,) paddle: Teacher.backbone.bb_2_5.conv0._conv.weight ---- (256, 1024, 1, 1) paddle: Teacher.backbone.bb_2_5.conv0._batch_norm.weight ---- (256,) paddle: Teacher.backbone.bb_2_5.conv0._batch_norm.bias ---- (256,) paddle: Teacher.backbone.bb_2_5.conv0._batch_norm._mean ---- (256,) paddle: Teacher.backbone.bb_2_5.conv0._batch_norm._variance ---- (256,) paddle: Teacher.backbone.bb_2_5.conv1._conv.weight ---- (256, 256, 3, 3) paddle: Teacher.backbone.bb_2_5.conv1._batch_norm.weight ---- (256,) paddle: Teacher.backbone.bb_2_5.conv1._batch_norm.bias ---- (256,) paddle: Teacher.backbone.bb_2_5.conv1._batch_norm._mean ---- (256,) paddle: Teacher.backbone.bb_2_5.conv1._batch_norm._variance ---- (256,) paddle: Teacher.backbone.bb_2_5.conv2._conv.weight ---- (1024, 256, 1, 1) paddle: Teacher.backbone.bb_2_5.conv2._batch_norm.weight ---- (1024,) paddle: Teacher.backbone.bb_2_5.conv2._batch_norm.bias ---- (1024,) paddle: Teacher.backbone.bb_2_5.conv2._batch_norm._mean ---- (1024,) paddle: Teacher.backbone.bb_2_5.conv2._batch_norm._variance ---- (1024,) paddle: Teacher.backbone.bb_3_0.conv0._conv.weight ---- (512, 1024, 1, 1) paddle: Teacher.backbone.bb_3_0.conv0._batch_norm.weight ---- (512,) paddle: Teacher.backbone.bb_3_0.conv0._batch_norm.bias ---- (512,) paddle: Teacher.backbone.bb_3_0.conv0._batch_norm._mean ---- (512,) paddle: Teacher.backbone.bb_3_0.conv0._batch_norm._variance ---- (512,) paddle: Teacher.backbone.bb_3_0.conv1._conv.weight ---- (512, 512, 3, 3) paddle: Teacher.backbone.bb_3_0.conv1._batch_norm.weight ---- (512,) paddle: Teacher.backbone.bb_3_0.conv1._batch_norm.bias ---- (512,) paddle: Teacher.backbone.bb_3_0.conv1._batch_norm._mean ---- (512,) paddle: Teacher.backbone.bb_3_0.conv1._batch_norm._variance ---- (512,) paddle: Teacher.backbone.bb_3_0.conv2._conv.weight ---- (2048, 512, 1, 1) paddle: Teacher.backbone.bb_3_0.conv2._batch_norm.weight ---- (2048,) paddle: Teacher.backbone.bb_3_0.conv2._batch_norm.bias ---- (2048,) paddle: Teacher.backbone.bb_3_0.conv2._batch_norm._mean ---- (2048,) paddle: Teacher.backbone.bb_3_0.conv2._batch_norm._variance ---- (2048,) paddle: Teacher.backbone.bb_3_0.short._conv.weight ---- (2048, 1024, 1, 1) paddle: Teacher.backbone.bb_3_0.short._batch_norm.weight ---- (2048,) paddle: Teacher.backbone.bb_3_0.short._batch_norm.bias ---- (2048,) paddle: Teacher.backbone.bb_3_0.short._batch_norm._mean ---- (2048,) paddle: Teacher.backbone.bb_3_0.short._batch_norm._variance ---- (2048,) paddle: Teacher.backbone.bb_3_1.conv0._conv.weight ---- (512, 2048, 1, 1) paddle: Teacher.backbone.bb_3_1.conv0._batch_norm.weight ---- (512,) paddle: Teacher.backbone.bb_3_1.conv0._batch_norm.bias ---- (512,) paddle: Teacher.backbone.bb_3_1.conv0._batch_norm._mean ---- (512,) paddle: Teacher.backbone.bb_3_1.conv0._batch_norm._variance ---- (512,) paddle: Teacher.backbone.bb_3_1.conv1._conv.weight ---- (512, 512, 3, 3) paddle: Teacher.backbone.bb_3_1.conv1._batch_norm.weight ---- (512,) paddle: Teacher.backbone.bb_3_1.conv1._batch_norm.bias ---- (512,) paddle: Teacher.backbone.bb_3_1.conv1._batch_norm._mean ---- (512,) paddle: Teacher.backbone.bb_3_1.conv1._batch_norm._variance ---- (512,) paddle: Teacher.backbone.bb_3_1.conv2._conv.weight ---- (2048, 512, 1, 1) paddle: Teacher.backbone.bb_3_1.conv2._batch_norm.weight ---- (2048,) paddle: Teacher.backbone.bb_3_1.conv2._batch_norm.bias ---- (2048,) paddle: Teacher.backbone.bb_3_1.conv2._batch_norm._mean ---- (2048,) paddle: Teacher.backbone.bb_3_1.conv2._batch_norm._variance ---- (2048,) paddle: Teacher.backbone.bb_3_2.conv0._conv.weight ---- (512, 2048, 1, 1) paddle: Teacher.backbone.bb_3_2.conv0._batch_norm.weight ---- (512,) paddle: Teacher.backbone.bb_3_2.conv0._batch_norm.bias ---- (512,) paddle: Teacher.backbone.bb_3_2.conv0._batch_norm._mean ---- (512,) paddle: Teacher.backbone.bb_3_2.conv0._batch_norm._variance ---- (512,) paddle: Teacher.backbone.bb_3_2.conv1._conv.weight ---- (512, 512, 3, 3) paddle: Teacher.backbone.bb_3_2.conv1._batch_norm.weight ---- (512,) paddle: Teacher.backbone.bb_3_2.conv1._batch_norm.bias ---- (512,) paddle: Teacher.backbone.bb_3_2.conv1._batch_norm._mean ---- (512,) paddle: Teacher.backbone.bb_3_2.conv1._batch_norm._variance ---- (512,) paddle: Teacher.backbone.bb_3_2.conv2._conv.weight ---- (2048, 512, 1, 1) paddle: Teacher.backbone.bb_3_2.conv2._batch_norm.weight ---- (2048,) paddle: Teacher.backbone.bb_3_2.conv2._batch_norm.bias ---- (2048,) paddle: Teacher.backbone.bb_3_2.conv2._batch_norm._mean ---- (2048,) paddle: Teacher.backbone.bb_3_2.conv2._batch_norm._variance ---- (2048,) paddle: Teacher.neck.ins_conv.0.weight ---- (256, 256, 1, 1) paddle: Teacher.neck.ins_conv.1.weight ---- (256, 512, 1, 1) paddle: Teacher.neck.ins_conv.2.weight ---- (256, 1024, 1, 1) paddle: Teacher.neck.ins_conv.3.weight ---- (256, 2048, 1, 1) paddle: Teacher.neck.pan_head_conv.2.weight ---- (64, 64, 3, 3) paddle: Teacher.neck.pan_head_conv.1.weight ---- (64, 64, 3, 3) paddle: Teacher.neck.pan_head_conv.0.weight ---- (64, 64, 3, 3) paddle: Teacher.neck.inp_conv.3.weight ---- (64, 256, 9, 9) paddle: Teacher.neck.inp_conv.2.weight ---- (64, 256, 9, 9) paddle: Teacher.neck.inp_conv.1.weight ---- (64, 256, 9, 9) paddle: Teacher.neck.inp_conv.0.weight ---- (64, 256, 9, 9) paddle: Teacher.neck.pan_lat_conv.2.weight ---- (64, 64, 9, 9) paddle: Teacher.neck.pan_lat_conv.1.weight ---- (64, 64, 9, 9) paddle: Teacher.neck.pan_lat_conv.0.weight ---- (64, 64, 9, 9) paddle: Teacher.neck.pan_lat_conv.3.weight ---- (64, 64, 9, 9) paddle: Teacher.head.binarize.conv1.weight ---- (64, 256, 7, 7) paddle: Teacher.head.binarize.conv_bn1.weight ---- (64,) paddle: Teacher.head.binarize.conv_bn1.bias ---- (64,) paddle: Teacher.head.binarize.conv_bn1._mean ---- (64,) paddle: Teacher.head.binarize.conv_bn1._variance ---- (64,) paddle: Teacher.head.binarize.conv2.weight ---- (64, 64, 2, 2) paddle: Teacher.head.binarize.conv2.bias ---- (64,) paddle: Teacher.head.binarize.conv_bn2.weight ---- (64,) paddle: Teacher.head.binarize.conv_bn2.bias ---- (64,) paddle: Teacher.head.binarize.conv_bn2._mean ---- (64,) paddle: Teacher.head.binarize.conv_bn2._variance ---- (64,) paddle: Teacher.head.binarize.conv3.weight ---- (64, 1, 2, 2) paddle: Teacher.head.binarize.conv3.bias ---- (1,) paddle: Teacher.head.thresh.conv1.weight ---- (64, 256, 7, 7) paddle: Teacher.head.thresh.conv_bn1.weight ---- (64,) paddle: Teacher.head.thresh.conv_bn1.bias ---- (64,) paddle: Teacher.head.thresh.conv_bn1._mean ---- (64,) paddle: Teacher.head.thresh.conv_bn1._variance ---- (64,) paddle: Teacher.head.thresh.conv2.weight ---- (64, 64, 2, 2) paddle: Teacher.head.thresh.conv2.bias ---- (64,) paddle: Teacher.head.thresh.conv_bn2.weight ---- (64,) paddle: Teacher.head.thresh.conv_bn2.bias ---- (64,) paddle: Teacher.head.thresh.conv_bn2._mean ---- (64,) paddle: Teacher.head.thresh.conv_bn2._variance ---- (64,) paddle: Teacher.head.thresh.conv3.weight ---- (64, 1, 2, 2) paddle: Teacher.head.thresh.conv3.bias ---- (1,) pytorch: backbone.conv.conv.weight ---- torch.Size([8, 3, 3, 3]) pytorch: backbone.conv.bn.weight ---- torch.Size([8]) pytorch: backbone.conv.bn.bias ---- torch.Size([8]) pytorch: backbone.conv.bn.running_mean ---- torch.Size([8]) pytorch: backbone.conv.bn.running_var ---- torch.Size([8]) pytorch: backbone.conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.0.0.expand_conv.conv.weight ---- torch.Size([8, 8, 1, 1]) pytorch: backbone.stages.0.0.expand_conv.bn.weight ---- torch.Size([8]) pytorch: backbone.stages.0.0.expand_conv.bn.bias ---- torch.Size([8]) pytorch: backbone.stages.0.0.expand_conv.bn.running_mean ---- torch.Size([8]) pytorch: backbone.stages.0.0.expand_conv.bn.running_var ---- torch.Size([8]) pytorch: backbone.stages.0.0.expand_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.0.0.bottleneck_conv.conv.weight ---- torch.Size([8, 1, 3, 3]) pytorch: backbone.stages.0.0.bottleneck_conv.bn.weight ---- torch.Size([8]) pytorch: backbone.stages.0.0.bottleneck_conv.bn.bias ---- torch.Size([8]) pytorch: backbone.stages.0.0.bottleneck_conv.bn.running_mean ---- torch.Size([8]) pytorch: backbone.stages.0.0.bottleneck_conv.bn.running_var ---- torch.Size([8]) pytorch: backbone.stages.0.0.bottleneck_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.0.0.linear_conv.conv.weight ---- torch.Size([8, 8, 1, 1]) pytorch: backbone.stages.0.0.linear_conv.bn.weight ---- torch.Size([8]) pytorch: backbone.stages.0.0.linear_conv.bn.bias ---- torch.Size([8]) pytorch: backbone.stages.0.0.linear_conv.bn.running_mean ---- torch.Size([8]) pytorch: backbone.stages.0.0.linear_conv.bn.running_var ---- torch.Size([8]) pytorch: backbone.stages.0.0.linear_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.0.1.expand_conv.conv.weight ---- torch.Size([32, 8, 1, 1]) pytorch: backbone.stages.0.1.expand_conv.bn.weight ---- torch.Size([32]) pytorch: backbone.stages.0.1.expand_conv.bn.bias ---- torch.Size([32]) pytorch: backbone.stages.0.1.expand_conv.bn.running_mean ---- torch.Size([32]) pytorch: backbone.stages.0.1.expand_conv.bn.running_var ---- torch.Size([32]) pytorch: backbone.stages.0.1.expand_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.0.1.bottleneck_conv.conv.weight ---- torch.Size([32, 1, 3, 3]) pytorch: backbone.stages.0.1.bottleneck_conv.bn.weight ---- torch.Size([32]) pytorch: backbone.stages.0.1.bottleneck_conv.bn.bias ---- torch.Size([32]) pytorch: backbone.stages.0.1.bottleneck_conv.bn.running_mean ---- torch.Size([32]) pytorch: backbone.stages.0.1.bottleneck_conv.bn.running_var ---- torch.Size([32]) pytorch: backbone.stages.0.1.bottleneck_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.0.1.linear_conv.conv.weight ---- torch.Size([16, 32, 1, 1]) pytorch: backbone.stages.0.1.linear_conv.bn.weight ---- torch.Size([16]) pytorch: backbone.stages.0.1.linear_conv.bn.bias ---- torch.Size([16]) pytorch: backbone.stages.0.1.linear_conv.bn.running_mean ---- torch.Size([16]) pytorch: backbone.stages.0.1.linear_conv.bn.running_var ---- torch.Size([16]) pytorch: backbone.stages.0.1.linear_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.0.2.expand_conv.conv.weight ---- torch.Size([40, 16, 1, 1]) pytorch: backbone.stages.0.2.expand_conv.bn.weight ---- torch.Size([40]) pytorch: backbone.stages.0.2.expand_conv.bn.bias ---- torch.Size([40]) pytorch: backbone.stages.0.2.expand_conv.bn.running_mean ---- torch.Size([40]) pytorch: backbone.stages.0.2.expand_conv.bn.running_var ---- torch.Size([40]) pytorch: backbone.stages.0.2.expand_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.0.2.bottleneck_conv.conv.weight ---- torch.Size([40, 1, 3, 3]) pytorch: backbone.stages.0.2.bottleneck_conv.bn.weight ---- torch.Size([40]) pytorch: backbone.stages.0.2.bottleneck_conv.bn.bias ---- torch.Size([40]) pytorch: backbone.stages.0.2.bottleneck_conv.bn.running_mean ---- torch.Size([40]) pytorch: backbone.stages.0.2.bottleneck_conv.bn.running_var ---- torch.Size([40]) pytorch: backbone.stages.0.2.bottleneck_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.0.2.linear_conv.conv.weight ---- torch.Size([16, 40, 1, 1]) pytorch: backbone.stages.0.2.linear_conv.bn.weight ---- torch.Size([16]) pytorch: backbone.stages.0.2.linear_conv.bn.bias ---- torch.Size([16]) pytorch: backbone.stages.0.2.linear_conv.bn.running_mean ---- torch.Size([16]) pytorch: backbone.stages.0.2.linear_conv.bn.running_var ---- torch.Size([16]) pytorch: backbone.stages.0.2.linear_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.1.0.expand_conv.conv.weight ---- torch.Size([40, 16, 1, 1]) pytorch: backbone.stages.1.0.expand_conv.bn.weight ---- torch.Size([40]) pytorch: backbone.stages.1.0.expand_conv.bn.bias ---- torch.Size([40]) pytorch: backbone.stages.1.0.expand_conv.bn.running_mean ---- torch.Size([40]) pytorch: backbone.stages.1.0.expand_conv.bn.running_var ---- torch.Size([40]) pytorch: backbone.stages.1.0.expand_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.1.0.bottleneck_conv.conv.weight ---- torch.Size([40, 1, 5, 5]) pytorch: backbone.stages.1.0.bottleneck_conv.bn.weight ---- torch.Size([40]) pytorch: backbone.stages.1.0.bottleneck_conv.bn.bias ---- torch.Size([40]) pytorch: backbone.stages.1.0.bottleneck_conv.bn.running_mean ---- torch.Size([40]) pytorch: backbone.stages.1.0.bottleneck_conv.bn.running_var ---- torch.Size([40]) pytorch: backbone.stages.1.0.bottleneck_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.1.0.linear_conv.conv.weight ---- torch.Size([24, 40, 1, 1]) pytorch: backbone.stages.1.0.linear_conv.bn.weight ---- torch.Size([24]) pytorch: backbone.stages.1.0.linear_conv.bn.bias ---- torch.Size([24]) pytorch: backbone.stages.1.0.linear_conv.bn.running_mean ---- torch.Size([24]) pytorch: backbone.stages.1.0.linear_conv.bn.running_var ---- torch.Size([24]) pytorch: backbone.stages.1.0.linear_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.1.1.expand_conv.conv.weight ---- torch.Size([64, 24, 1, 1]) pytorch: backbone.stages.1.1.expand_conv.bn.weight ---- torch.Size([64]) pytorch: backbone.stages.1.1.expand_conv.bn.bias ---- torch.Size([64]) pytorch: backbone.stages.1.1.expand_conv.bn.running_mean ---- torch.Size([64]) pytorch: backbone.stages.1.1.expand_conv.bn.running_var ---- torch.Size([64]) pytorch: backbone.stages.1.1.expand_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.1.1.bottleneck_conv.conv.weight ---- torch.Size([64, 1, 5, 5]) pytorch: backbone.stages.1.1.bottleneck_conv.bn.weight ---- torch.Size([64]) pytorch: backbone.stages.1.1.bottleneck_conv.bn.bias ---- torch.Size([64]) pytorch: backbone.stages.1.1.bottleneck_conv.bn.running_mean ---- torch.Size([64]) pytorch: backbone.stages.1.1.bottleneck_conv.bn.running_var ---- torch.Size([64]) pytorch: backbone.stages.1.1.bottleneck_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.1.1.linear_conv.conv.weight ---- torch.Size([24, 64, 1, 1]) pytorch: backbone.stages.1.1.linear_conv.bn.weight ---- torch.Size([24]) pytorch: backbone.stages.1.1.linear_conv.bn.bias ---- torch.Size([24]) pytorch: backbone.stages.1.1.linear_conv.bn.running_mean ---- torch.Size([24]) pytorch: backbone.stages.1.1.linear_conv.bn.running_var ---- torch.Size([24]) pytorch: backbone.stages.1.1.linear_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.1.2.expand_conv.conv.weight ---- torch.Size([64, 24, 1, 1]) pytorch: backbone.stages.1.2.expand_conv.bn.weight ---- torch.Size([64]) pytorch: backbone.stages.1.2.expand_conv.bn.bias ---- torch.Size([64]) pytorch: backbone.stages.1.2.expand_conv.bn.running_mean ---- torch.Size([64]) pytorch: backbone.stages.1.2.expand_conv.bn.running_var ---- torch.Size([64]) pytorch: backbone.stages.1.2.expand_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.1.2.bottleneck_conv.conv.weight ---- torch.Size([64, 1, 5, 5]) pytorch: backbone.stages.1.2.bottleneck_conv.bn.weight ---- torch.Size([64]) pytorch: backbone.stages.1.2.bottleneck_conv.bn.bias ---- torch.Size([64]) pytorch: backbone.stages.1.2.bottleneck_conv.bn.running_mean ---- torch.Size([64]) pytorch: backbone.stages.1.2.bottleneck_conv.bn.running_var ---- torch.Size([64]) pytorch: backbone.stages.1.2.bottleneck_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.1.2.linear_conv.conv.weight ---- torch.Size([24, 64, 1, 1]) pytorch: backbone.stages.1.2.linear_conv.bn.weight ---- torch.Size([24]) pytorch: backbone.stages.1.2.linear_conv.bn.bias ---- torch.Size([24]) pytorch: backbone.stages.1.2.linear_conv.bn.running_mean ---- torch.Size([24]) pytorch: backbone.stages.1.2.linear_conv.bn.running_var ---- torch.Size([24]) pytorch: backbone.stages.1.2.linear_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.2.0.expand_conv.conv.weight ---- torch.Size([120, 24, 1, 1]) pytorch: backbone.stages.2.0.expand_conv.bn.weight ---- torch.Size([120]) pytorch: backbone.stages.2.0.expand_conv.bn.bias ---- torch.Size([120]) pytorch: backbone.stages.2.0.expand_conv.bn.running_mean ---- torch.Size([120]) pytorch: backbone.stages.2.0.expand_conv.bn.running_var ---- torch.Size([120]) pytorch: backbone.stages.2.0.expand_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.2.0.bottleneck_conv.conv.weight ---- torch.Size([120, 1, 3, 3]) pytorch: backbone.stages.2.0.bottleneck_conv.bn.weight ---- torch.Size([120]) pytorch: backbone.stages.2.0.bottleneck_conv.bn.bias ---- torch.Size([120]) pytorch: backbone.stages.2.0.bottleneck_conv.bn.running_mean ---- torch.Size([120]) pytorch: backbone.stages.2.0.bottleneck_conv.bn.running_var ---- torch.Size([120]) pytorch: backbone.stages.2.0.bottleneck_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.2.0.linear_conv.conv.weight ---- torch.Size([40, 120, 1, 1]) pytorch: backbone.stages.2.0.linear_conv.bn.weight ---- torch.Size([40]) pytorch: backbone.stages.2.0.linear_conv.bn.bias ---- torch.Size([40]) pytorch: backbone.stages.2.0.linear_conv.bn.running_mean ---- torch.Size([40]) pytorch: backbone.stages.2.0.linear_conv.bn.running_var ---- torch.Size([40]) pytorch: backbone.stages.2.0.linear_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.2.1.expand_conv.conv.weight ---- torch.Size([104, 40, 1, 1]) pytorch: backbone.stages.2.1.expand_conv.bn.weight ---- torch.Size([104]) pytorch: backbone.stages.2.1.expand_conv.bn.bias ---- torch.Size([104]) pytorch: backbone.stages.2.1.expand_conv.bn.running_mean ---- torch.Size([104]) pytorch: backbone.stages.2.1.expand_conv.bn.running_var ---- torch.Size([104]) pytorch: backbone.stages.2.1.expand_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.2.1.bottleneck_conv.conv.weight ---- torch.Size([104, 1, 3, 3]) pytorch: backbone.stages.2.1.bottleneck_conv.bn.weight ---- torch.Size([104]) pytorch: backbone.stages.2.1.bottleneck_conv.bn.bias ---- torch.Size([104]) pytorch: backbone.stages.2.1.bottleneck_conv.bn.running_mean ---- torch.Size([104]) pytorch: backbone.stages.2.1.bottleneck_conv.bn.running_var ---- torch.Size([104]) pytorch: backbone.stages.2.1.bottleneck_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.2.1.linear_conv.conv.weight ---- torch.Size([40, 104, 1, 1]) pytorch: backbone.stages.2.1.linear_conv.bn.weight ---- torch.Size([40]) pytorch: backbone.stages.2.1.linear_conv.bn.bias ---- torch.Size([40]) pytorch: backbone.stages.2.1.linear_conv.bn.running_mean ---- torch.Size([40]) pytorch: backbone.stages.2.1.linear_conv.bn.running_var ---- torch.Size([40]) pytorch: backbone.stages.2.1.linear_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.2.2.expand_conv.conv.weight ---- torch.Size([96, 40, 1, 1]) pytorch: backbone.stages.2.2.expand_conv.bn.weight ---- torch.Size([96]) pytorch: backbone.stages.2.2.expand_conv.bn.bias ---- torch.Size([96]) pytorch: backbone.stages.2.2.expand_conv.bn.running_mean ---- torch.Size([96]) pytorch: backbone.stages.2.2.expand_conv.bn.running_var ---- torch.Size([96]) pytorch: backbone.stages.2.2.expand_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.2.2.bottleneck_conv.conv.weight ---- torch.Size([96, 1, 3, 3]) pytorch: backbone.stages.2.2.bottleneck_conv.bn.weight ---- torch.Size([96]) pytorch: backbone.stages.2.2.bottleneck_conv.bn.bias ---- torch.Size([96]) pytorch: backbone.stages.2.2.bottleneck_conv.bn.running_mean ---- torch.Size([96]) pytorch: backbone.stages.2.2.bottleneck_conv.bn.running_var ---- torch.Size([96]) pytorch: backbone.stages.2.2.bottleneck_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.2.2.linear_conv.conv.weight ---- torch.Size([40, 96, 1, 1]) pytorch: backbone.stages.2.2.linear_conv.bn.weight ---- torch.Size([40]) pytorch: backbone.stages.2.2.linear_conv.bn.bias ---- torch.Size([40]) pytorch: backbone.stages.2.2.linear_conv.bn.running_mean ---- torch.Size([40]) pytorch: backbone.stages.2.2.linear_conv.bn.running_var ---- torch.Size([40]) pytorch: backbone.stages.2.2.linear_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.2.3.expand_conv.conv.weight ---- torch.Size([96, 40, 1, 1]) pytorch: backbone.stages.2.3.expand_conv.bn.weight ---- torch.Size([96]) pytorch: backbone.stages.2.3.expand_conv.bn.bias ---- torch.Size([96]) pytorch: backbone.stages.2.3.expand_conv.bn.running_mean ---- torch.Size([96]) pytorch: backbone.stages.2.3.expand_conv.bn.running_var ---- torch.Size([96]) pytorch: backbone.stages.2.3.expand_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.2.3.bottleneck_conv.conv.weight ---- torch.Size([96, 1, 3, 3]) pytorch: backbone.stages.2.3.bottleneck_conv.bn.weight ---- torch.Size([96]) pytorch: backbone.stages.2.3.bottleneck_conv.bn.bias ---- torch.Size([96]) pytorch: backbone.stages.2.3.bottleneck_conv.bn.running_mean ---- torch.Size([96]) pytorch: backbone.stages.2.3.bottleneck_conv.bn.running_var ---- torch.Size([96]) pytorch: backbone.stages.2.3.bottleneck_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.2.3.linear_conv.conv.weight ---- torch.Size([40, 96, 1, 1]) pytorch: backbone.stages.2.3.linear_conv.bn.weight ---- torch.Size([40]) pytorch: backbone.stages.2.3.linear_conv.bn.bias ---- torch.Size([40]) pytorch: backbone.stages.2.3.linear_conv.bn.running_mean ---- torch.Size([40]) pytorch: backbone.stages.2.3.linear_conv.bn.running_var ---- torch.Size([40]) pytorch: backbone.stages.2.3.linear_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.2.4.expand_conv.conv.weight ---- torch.Size([240, 40, 1, 1]) pytorch: backbone.stages.2.4.expand_conv.bn.weight ---- torch.Size([240]) pytorch: backbone.stages.2.4.expand_conv.bn.bias ---- torch.Size([240]) pytorch: backbone.stages.2.4.expand_conv.bn.running_mean ---- torch.Size([240]) pytorch: backbone.stages.2.4.expand_conv.bn.running_var ---- torch.Size([240]) pytorch: backbone.stages.2.4.expand_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.2.4.bottleneck_conv.conv.weight ---- torch.Size([240, 1, 3, 3]) pytorch: backbone.stages.2.4.bottleneck_conv.bn.weight ---- torch.Size([240]) pytorch: backbone.stages.2.4.bottleneck_conv.bn.bias ---- torch.Size([240]) pytorch: backbone.stages.2.4.bottleneck_conv.bn.running_mean ---- torch.Size([240]) pytorch: backbone.stages.2.4.bottleneck_conv.bn.running_var ---- torch.Size([240]) pytorch: backbone.stages.2.4.bottleneck_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.2.4.linear_conv.conv.weight ---- torch.Size([56, 240, 1, 1]) pytorch: backbone.stages.2.4.linear_conv.bn.weight ---- torch.Size([56]) pytorch: backbone.stages.2.4.linear_conv.bn.bias ---- torch.Size([56]) pytorch: backbone.stages.2.4.linear_conv.bn.running_mean ---- torch.Size([56]) pytorch: backbone.stages.2.4.linear_conv.bn.running_var ---- torch.Size([56]) pytorch: backbone.stages.2.4.linear_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.2.5.expand_conv.conv.weight ---- torch.Size([336, 56, 1, 1]) pytorch: backbone.stages.2.5.expand_conv.bn.weight ---- torch.Size([336]) pytorch: backbone.stages.2.5.expand_conv.bn.bias ---- torch.Size([336]) pytorch: backbone.stages.2.5.expand_conv.bn.running_mean ---- torch.Size([336]) pytorch: backbone.stages.2.5.expand_conv.bn.running_var ---- torch.Size([336]) pytorch: backbone.stages.2.5.expand_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.2.5.bottleneck_conv.conv.weight ---- torch.Size([336, 1, 3, 3]) pytorch: backbone.stages.2.5.bottleneck_conv.bn.weight ---- torch.Size([336]) pytorch: backbone.stages.2.5.bottleneck_conv.bn.bias ---- torch.Size([336]) pytorch: backbone.stages.2.5.bottleneck_conv.bn.running_mean ---- torch.Size([336]) pytorch: backbone.stages.2.5.bottleneck_conv.bn.running_var ---- torch.Size([336]) pytorch: backbone.stages.2.5.bottleneck_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.2.5.linear_conv.conv.weight ---- torch.Size([56, 336, 1, 1]) pytorch: backbone.stages.2.5.linear_conv.bn.weight ---- torch.Size([56]) pytorch: backbone.stages.2.5.linear_conv.bn.bias ---- torch.Size([56]) pytorch: backbone.stages.2.5.linear_conv.bn.running_mean ---- torch.Size([56]) pytorch: backbone.stages.2.5.linear_conv.bn.running_var ---- torch.Size([56]) pytorch: backbone.stages.2.5.linear_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.3.0.expand_conv.conv.weight ---- torch.Size([336, 56, 1, 1]) pytorch: backbone.stages.3.0.expand_conv.bn.weight ---- torch.Size([336]) pytorch: backbone.stages.3.0.expand_conv.bn.bias ---- torch.Size([336]) pytorch: backbone.stages.3.0.expand_conv.bn.running_mean ---- torch.Size([336]) pytorch: backbone.stages.3.0.expand_conv.bn.running_var ---- torch.Size([336]) pytorch: backbone.stages.3.0.expand_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.3.0.bottleneck_conv.conv.weight ---- torch.Size([336, 1, 5, 5]) pytorch: backbone.stages.3.0.bottleneck_conv.bn.weight ---- torch.Size([336]) pytorch: backbone.stages.3.0.bottleneck_conv.bn.bias ---- torch.Size([336]) pytorch: backbone.stages.3.0.bottleneck_conv.bn.running_mean ---- torch.Size([336]) pytorch: backbone.stages.3.0.bottleneck_conv.bn.running_var ---- torch.Size([336]) pytorch: backbone.stages.3.0.bottleneck_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.3.0.linear_conv.conv.weight ---- torch.Size([80, 336, 1, 1]) pytorch: backbone.stages.3.0.linear_conv.bn.weight ---- torch.Size([80]) pytorch: backbone.stages.3.0.linear_conv.bn.bias ---- torch.Size([80]) pytorch: backbone.stages.3.0.linear_conv.bn.running_mean ---- torch.Size([80]) pytorch: backbone.stages.3.0.linear_conv.bn.running_var ---- torch.Size([80]) pytorch: backbone.stages.3.0.linear_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.3.1.expand_conv.conv.weight ---- torch.Size([480, 80, 1, 1]) pytorch: backbone.stages.3.1.expand_conv.bn.weight ---- torch.Size([480]) pytorch: backbone.stages.3.1.expand_conv.bn.bias ---- torch.Size([480]) pytorch: backbone.stages.3.1.expand_conv.bn.running_mean ---- torch.Size([480]) pytorch: backbone.stages.3.1.expand_conv.bn.running_var ---- torch.Size([480]) pytorch: backbone.stages.3.1.expand_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.3.1.bottleneck_conv.conv.weight ---- torch.Size([480, 1, 5, 5]) pytorch: backbone.stages.3.1.bottleneck_conv.bn.weight ---- torch.Size([480]) pytorch: backbone.stages.3.1.bottleneck_conv.bn.bias ---- torch.Size([480]) pytorch: backbone.stages.3.1.bottleneck_conv.bn.running_mean ---- torch.Size([480]) pytorch: backbone.stages.3.1.bottleneck_conv.bn.running_var ---- torch.Size([480]) pytorch: backbone.stages.3.1.bottleneck_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.3.1.linear_conv.conv.weight ---- torch.Size([80, 480, 1, 1]) pytorch: backbone.stages.3.1.linear_conv.bn.weight ---- torch.Size([80]) pytorch: backbone.stages.3.1.linear_conv.bn.bias ---- torch.Size([80]) pytorch: backbone.stages.3.1.linear_conv.bn.running_mean ---- torch.Size([80]) pytorch: backbone.stages.3.1.linear_conv.bn.running_var ---- torch.Size([80]) pytorch: backbone.stages.3.1.linear_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.3.2.expand_conv.conv.weight ---- torch.Size([480, 80, 1, 1]) pytorch: backbone.stages.3.2.expand_conv.bn.weight ---- torch.Size([480]) pytorch: backbone.stages.3.2.expand_conv.bn.bias ---- torch.Size([480]) pytorch: backbone.stages.3.2.expand_conv.bn.running_mean ---- torch.Size([480]) pytorch: backbone.stages.3.2.expand_conv.bn.running_var ---- torch.Size([480]) pytorch: backbone.stages.3.2.expand_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.3.2.bottleneck_conv.conv.weight ---- torch.Size([480, 1, 5, 5]) pytorch: backbone.stages.3.2.bottleneck_conv.bn.weight ---- torch.Size([480]) pytorch: backbone.stages.3.2.bottleneck_conv.bn.bias ---- torch.Size([480]) pytorch: backbone.stages.3.2.bottleneck_conv.bn.running_mean ---- torch.Size([480]) pytorch: backbone.stages.3.2.bottleneck_conv.bn.running_var ---- torch.Size([480]) pytorch: backbone.stages.3.2.bottleneck_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.3.2.linear_conv.conv.weight ---- torch.Size([80, 480, 1, 1]) pytorch: backbone.stages.3.2.linear_conv.bn.weight ---- torch.Size([80]) pytorch: backbone.stages.3.2.linear_conv.bn.bias ---- torch.Size([80]) pytorch: backbone.stages.3.2.linear_conv.bn.running_mean ---- torch.Size([80]) pytorch: backbone.stages.3.2.linear_conv.bn.running_var ---- torch.Size([80]) pytorch: backbone.stages.3.2.linear_conv.bn.num_batches_tracked ---- torch.Size([]) pytorch: backbone.stages.3.3.conv.weight ---- torch.Size([480, 80, 1, 1]) pytorch: backbone.stages.3.3.bn.weight ---- torch.Size([480]) pytorch: backbone.stages.3.3.bn.bias ---- torch.Size([480]) pytorch: backbone.stages.3.3.bn.running_mean ---- torch.Size([480]) pytorch: backbone.stages.3.3.bn.running_var ---- torch.Size([480]) pytorch: backbone.stages.3.3.bn.num_batches_tracked ---- torch.Size([]) pytorch: neck.ins_conv.0.in_conv.weight ---- torch.Size([96, 16, 1, 1]) pytorch: neck.ins_conv.0.se_block.conv1.weight ---- torch.Size([24, 96, 1, 1]) pytorch: neck.ins_conv.0.se_block.conv1.bias ---- torch.Size([24]) pytorch: neck.ins_conv.0.se_block.conv2.weight ---- torch.Size([96, 24, 1, 1]) pytorch: neck.ins_conv.0.se_block.conv2.bias ---- torch.Size([96]) pytorch: neck.ins_conv.1.in_conv.weight ---- torch.Size([96, 24, 1, 1]) pytorch: neck.ins_conv.1.se_block.conv1.weight ---- torch.Size([24, 96, 1, 1]) pytorch: neck.ins_conv.1.se_block.conv1.bias ---- torch.Size([24]) pytorch: neck.ins_conv.1.se_block.conv2.weight ---- torch.Size([96, 24, 1, 1]) pytorch: neck.ins_conv.1.se_block.conv2.bias ---- torch.Size([96]) pytorch: neck.ins_conv.2.in_conv.weight ---- torch.Size([96, 56, 1, 1]) pytorch: neck.ins_conv.2.se_block.conv1.weight ---- torch.Size([24, 96, 1, 1]) pytorch: neck.ins_conv.2.se_block.conv1.bias ---- torch.Size([24]) pytorch: neck.ins_conv.2.se_block.conv2.weight ---- torch.Size([96, 24, 1, 1]) pytorch: neck.ins_conv.2.se_block.conv2.bias ---- torch.Size([96]) pytorch: neck.ins_conv.3.in_conv.weight ---- torch.Size([96, 480, 1, 1]) pytorch: neck.ins_conv.3.se_block.conv1.weight ---- torch.Size([24, 96, 1, 1]) pytorch: neck.ins_conv.3.se_block.conv1.bias ---- torch.Size([24]) pytorch: neck.ins_conv.3.se_block.conv2.weight ---- torch.Size([96, 24, 1, 1]) pytorch: neck.ins_conv.3.se_block.conv2.bias ---- torch.Size([96]) pytorch: neck.inp_conv.0.in_conv.weight ---- torch.Size([24, 96, 3, 3]) pytorch: neck.inp_conv.0.se_block.conv1.weight ---- torch.Size([6, 24, 1, 1]) pytorch: neck.inp_conv.0.se_block.conv1.bias ---- torch.Size([6]) pytorch: neck.inp_conv.0.se_block.conv2.weight ---- torch.Size([24, 6, 1, 1]) pytorch: neck.inp_conv.0.se_block.conv2.bias ---- torch.Size([24]) pytorch: neck.inp_conv.1.in_conv.weight ---- torch.Size([24, 96, 3, 3]) pytorch: neck.inp_conv.1.se_block.conv1.weight ---- torch.Size([6, 24, 1, 1]) pytorch: neck.inp_conv.1.se_block.conv1.bias ---- torch.Size([6]) pytorch: neck.inp_conv.1.se_block.conv2.weight ---- torch.Size([24, 6, 1, 1]) pytorch: neck.inp_conv.1.se_block.conv2.bias ---- torch.Size([24]) pytorch: neck.inp_conv.2.in_conv.weight ---- torch.Size([24, 96, 3, 3]) pytorch: neck.inp_conv.2.se_block.conv1.weight ---- torch.Size([6, 24, 1, 1]) pytorch: neck.inp_conv.2.se_block.conv1.bias ---- torch.Size([6]) pytorch: neck.inp_conv.2.se_block.conv2.weight ---- torch.Size([24, 6, 1, 1]) pytorch: neck.inp_conv.2.se_block.conv2.bias ---- torch.Size([24]) pytorch: neck.inp_conv.3.in_conv.weight ---- torch.Size([24, 96, 3, 3]) pytorch: neck.inp_conv.3.se_block.conv1.weight ---- torch.Size([6, 24, 1, 1]) pytorch: neck.inp_conv.3.se_block.conv1.bias ---- torch.Size([6]) pytorch: neck.inp_conv.3.se_block.conv2.weight ---- torch.Size([24, 6, 1, 1]) pytorch: neck.inp_conv.3.se_block.conv2.bias ---- torch.Size([24]) pytorch: head.binarize.conv1.weight ---- torch.Size([24, 96, 3, 3]) pytorch: head.binarize.conv_bn1.weight ---- torch.Size([24]) pytorch: head.binarize.conv_bn1.bias ---- torch.Size([24]) pytorch: head.binarize.conv_bn1.running_mean ---- torch.Size([24]) pytorch: head.binarize.conv_bn1.running_var ---- torch.Size([24]) pytorch: head.binarize.conv_bn1.num_batches_tracked ---- torch.Size([]) pytorch: head.binarize.conv2.weight ---- torch.Size([24, 24, 2, 2]) pytorch: head.binarize.conv2.bias ---- torch.Size([24]) pytorch: head.binarize.conv_bn2.weight ---- torch.Size([24]) pytorch: head.binarize.conv_bn2.bias ---- torch.Size([24]) pytorch: head.binarize.conv_bn2.running_mean ---- torch.Size([24]) pytorch: head.binarize.conv_bn2.running_var ---- torch.Size([24]) pytorch: head.binarize.conv_bn2.num_batches_tracked ---- torch.Size([]) pytorch: head.binarize.conv3.weight ---- torch.Size([24, 1, 2, 2]) pytorch: head.binarize.conv3.bias ---- torch.Size([1]) pytorch: head.thresh.conv1.weight ---- torch.Size([24, 96, 3, 3]) pytorch: head.thresh.conv_bn1.weight ---- torch.Size([24]) pytorch: head.thresh.conv_bn1.bias ---- torch.Size([24]) pytorch: head.thresh.conv_bn1.running_mean ---- torch.Size([24]) pytorch: head.thresh.conv_bn1.running_var ---- torch.Size([24]) pytorch: head.thresh.conv_bn1.num_batches_tracked ---- torch.Size([]) pytorch: head.thresh.conv2.weight ---- torch.Size([24, 24, 2, 2]) pytorch: head.thresh.conv2.bias ---- torch.Size([24]) pytorch: head.thresh.conv_bn2.weight ---- torch.Size([24]) pytorch: head.thresh.conv_bn2.bias ---- torch.Size([24]) pytorch: head.thresh.conv_bn2.running_mean ---- torch.Size([24]) pytorch: head.thresh.conv_bn2.running_var ---- torch.Size([24]) pytorch: head.thresh.conv_bn2.num_batches_tracked ---- torch.Size([]) pytorch: head.thresh.conv3.weight ---- torch.Size([24, 1, 2, 2]) pytorch: head.thresh.conv3.bias ---- torch.Size([1]) model is loaded: pdocr/ch_PP-OCRv3_det_distill_train/best_accuracy todo out: 270.15472 0.0006595574 0.9999943 1.5668587e-32 model is saved: ch_ptocr_v3_det_infer.pth done.

[goldlee]

这个是我稍微改了之后的代码，就改了传参那儿，填了个默认值。

改后的代码

``` import os, sys sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__)))) from collections import OrderedDict import numpy as np import cv2 import torch from pytorchocr.base_ocr_v20 import BaseOCRV20 class PPOCRv3DetConverter(BaseOCRV20): def __init__(self, config, paddle_pretrained_model_path, **kwargs): super(PPOCRv3DetConverter, self).__init__(config, **kwargs) self.load_paddle_weights(paddle_pretrained_model_path) self.net.eval() def load_paddle_weights(self, weights_path): print('paddle weights loading...') import paddle.fluid as fluid with fluid.dygraph.guard(): para_state_dict, opti_state_dict = fluid.load_dygraph(weights_path) [print('paddle: {} ---- {}'.format(k, v.shape)) for k, v in para_state_dict.items()] [print('pytorch: {} ---- {}'.format(k, v.shape)) for k, v in self.net.state_dict().items()] for k,v in self.net.state_dict().items(): if k.endswith('num_batches_tracked'): continue ppname = k ppname = ppname.replace('.running_mean', '._mean') ppname = ppname.replace('.running_var', '._variance') if k.startswith('backbone.'): ppname = ppname.replace('backbone.', 'Student2.backbone.') ppname = ppname.replace('.stages.', '.stage') elif k.startswith('neck.'): ppname = ppname.replace('neck.', 'Student2.neck.') elif k.startswith('head.'): ppname = ppname.replace('head.', 'Student2.head.') else: print('Redundance:') print(k) raise ValueError try: self.net.state_dict()[k].copy_(torch.Tensor(para_state_dict[ppname])) except Exception as e: print('pytorch: {}, {}'.format(k, v.size())) print('paddle: {}, {}'.format(ppname, para_state_dict[ppname].shape)) raise e print('model is loaded: {}'.format(weights_path)) if __name__ == '__main__': import argparse, json, textwrap, sys, os parser = argparse.ArgumentParser() parser.add_argument("--src_model_path", type=str, default='pdocr/ch_PP-OCRv3_det_distill_train') args = parser.parse_args() cfg = {'model_type':'det', 'algorithm':'DB', 'Transform':None, 'Backbone':{'name':'MobileNetV3', 'model_name':'large', 'scale':0.5, 'disable_se':True}, 'Neck':{'name':'RSEFPN', 'out_channels':96, 'shortcut': True}, 'Head':{'name':'DBHead', 'k':50}} paddle_pretrained_model_path = os.path.join(os.path.abspath(args.src_model_path), 'best_accuracy') converter = PPOCRv3DetConverter(cfg, paddle_pretrained_model_path) print('todo') np.random.seed(666) inputs = np.random.randn(1, 3, 640, 640).astype(np.float32) inp = torch.from_numpy(inputs) out = converter.net(inp) out = out['maps'].data.numpy() print('out:', np.sum(out), np.mean(out), np.max(out), np.min(out)) # save save_basename = os.path.basename(os.path.abspath(args.src_model_path)) if save_basename.lower()[:3] == 'en_': save_name = 'en_ptocr_v3_det_infer.pth' elif save_basename.lower()[:3] == 'ch_': save_name = 'ch_ptocr_v3_det_infer.pth' else: save_name = 'multilingual_ptocr_v3_det_infer.pth' converter.save_pytorch_weights(save_name) print('done.') ```

[frotms]

是自己训练的模型吗？如果是paddle官方提供的模型是验证过的

[goldlee]

不是的，paddle官方的。

[frotms]

试下这里的模型：PyTorch模型下载链接：https://pan.baidu.com/s/1r1DELT8BlgxeOP2RqREJEg 提取码：6clx

[goldlee]

很感谢，用你网盘里的转是没问题的。