LORE 训练模型推理

[cqray1990]

trafficstars

训练命令： python main.py --task ctdet
--dataset_name table
--exp_id training_wtw
--dataset_name WTW
--image_dir /train/images
--wiz_4ps
--wiz_stacking
--wiz_pairloss
--tsfm_layers 3
--stacking_layers 3
--batch_size 32
--master_batch 6
--arch dla_34
--lr 1e-4
--K 500
--MK 1000
--num_epochs 100
--lr_step '70, 90'
--gpus 0,1,2,3
--num_workers 16
--val_intervals 10

得到训练模型推理命令： python demo.py --task ctdet
--dataset table
--demo /home/user/Docllm/docprj/AdvancedLiterateMachinery/DocumentUnderstanding/LORE-TSR/input_images/wired
--demo_name demo_wired
--debug 1
--arch dla_34
--K 3000
--MK 5000
--tsfm_layers 4
--stacking_layers 4
--gpus 0
--wiz_4ps
--wiz_detect
--wiz_rev
--wiz_stacking
--convert_onnx 1
--vis_thresh_corner 0.3
--vis_thresh 0.20
--scores_thresh 0.2
--nms
--demo_dir ../visualization_wired/
--load_model /exp/ctdet/training_wtw/model_last.pth
--load_processor /exp/ctdet/training_wtw/processor_last.pth 为什么会提示以错误呢：用仓库的给的模型没有提示这种参数找不到的

No param stacker.tsfm.encoder.layers.3.norm_1.alpha. No param stacker.tsfm.encoder.layers.3.norm_1.bias. No param stacker.tsfm.encoder.layers.3.norm_2.alpha. No param stacker.tsfm.encoder.layers.3.norm_2.bias. No param stacker.tsfm.encoder.layers.3.attn.q_linear.weight. No param stacker.tsfm.encoder.layers.3.attn.q_linear.bias. No param stacker.tsfm.encoder.layers.3.attn.v_linear.weight. No param stacker.tsfm.encoder.layers.3.attn.v_linear.bias. No param stacker.tsfm.encoder.layers.3.attn.k_linear.weight. No param stacker.tsfm.encoder.layers.3.attn.k_linear.bias. No param stacker.tsfm.encoder.layers.3.attn.out.weight. No param stacker.tsfm.encoder.layers.3.attn.out.bias. No param stacker.tsfm.encoder.layers.3.ff.linear_1.weight. No param stacker.tsfm.encoder.layers.3.ff.linear_1.bias. No param stacker.tsfm.encoder.layers.3.ff.linear_2.weight. No param stacker.tsfm.encoder.layers.3.ff.linear_2.bias. No param tsfm_axis.encoder.layers.3.norm_1.alpha. No param tsfm_axis.encoder.layers.3.norm_1.bias. No param tsfm_axis.encoder.layers.3.norm_2.alpha. No param tsfm_axis.encoder.layers.3.norm_2.bias. No param tsfm_axis.encoder.layers.3.attn.q_linear.weight. No param tsfm_axis.encoder.layers.3.attn.q_linear.bias. No param tsfm_axis.encoder.layers.3.attn.v_linear.weight. No param tsfm_axis.encoder.layers.3.attn.v_linear.bias. No param tsfm_axis.encoder.layers.3.attn.k_linear.weight. No param tsfm_axis.encoder.layers.3.attn.k_linear.bias. No param tsfm_axis.encoder.layers.3.attn.out.weight. No param tsfm_axis.encoder.layers.3.attn.out.bias. No param tsfm_axis.encoder.layers.3.ff.linear_1.weight. No param tsfm_axis.encoder.layers.3.ff.linear_1.bias. No param tsfm_axis.encoder.layers.3.ff.linear_2.weight. No param tsfm_axis.encoder.layers.3.ff.linear_2.bias.