kkpssr
kkpssr
现在图神经网络挺火的,mmlab有计划开一个包含各种图神经网络模型的项目么?
训练loss越来越高,eval准确率召回率越来越低,前10个epoch结果还好(eval结果miou 0.6),再跑10个epoch miou直接归0
如题,有相关的模型或预训练权重么?
由于原模型提取特征时输出只有特征无法和图片对应,如果将keeporder改为True则预测速度大幅下降,目前尝试了下改动reader中的processImg处理方法,增加了图片地址返回值,但报错(显示reader读取越界),同时也不知道对于str的返回值fluid.data应该选什么dtype类型
hi, i did not find pretrained model for FAN. how can i get the pretrained model? thanks
希望可视化的是ResNEt50的pdmodel文件,输入指令完全按照图示,一直卡在载入中页面
### Is there an existing issue for this? - [X] I have searched the existing issues ### Environment ```markdown - Milvus version:2.1 - Deployment mode(standalone or cluster): - SDK version(e.g....
hi, cause there is no training code , i try pretrain model on my own dataset(data from real world may experience resample, resize and muti-compression) , however i have tried...
我通过如下命令运行 1 import cv2 2 import fastdeploy as fd 3 import os 4 option=fd.RuntimeOption() 5 option.use_gpu(0) 6 option.use_trt_backend() 7 option.enable_paddle_to_trt() 8 option.enable_paddle_trt_collect_shape() 10 option.set_trt_input_shape("image", [1, 3, 256, 256], [1,3,512,512],[1, 3,...