MTP

MTP copied to clipboard

Question about the environment setting

2

Dear author, I have a question about the environment setting. It seems to have some conflicts in your given Python lib setting. For example, Pytorch 0.10.0 can not match Torchvision...

cassiaaaaaa

Multichannel images

3

Hello, It seems that the examples focus on 3 channel images. Is semantic segmentation expected to work on multichannel images, and is there an example on that? I know that...

sebbelese

测试

3

请问应该如何用你训练好的模型做测试呢，没有找到tool/test.py，在windows系统下可以跑吗

bzanniupoiiu

运用

2

你好，我想请问一下我可以直接用你们已发布的模型，来估计自己数据集图像中目标数量吗，怎样做更加方便呢？谢谢！

chris3chris

我想请问一下你们tools/train.py可以公开一下吗

2

freja-zy

ChenHongruixuan

关于旋转目标检测

1

您好作者，如果我想用您给的预训练模型训练我自己的数据集的话，有两个问题想要请教您 1.因为我的数据集是细粒度类别的目标检测，所以想用您FAIR1M训练的config进行训练，是不是用 `pretrained ='/work/share/achk2o1zg1/diwang22/work_dir/multitask_pretrain/pretrain/avg/with_background/vit_b_rvsa_224_mae_samrs_mtp_three/last_vit_b_rvsa_ss_is_rd_pretrn_model_encoder.pth'` 把这个预训练模型下载下来然后训练就可以了？ 2.看您论文里没提多尺度训练的事，需要进行多尺度的切割吗比如[0.5,1.0,1.5]，还是就单尺度切割就可以了呢？ 希望您能回复，谢谢！

lzJune

Data Question

1

How can I learn from a tif file in Change detection? The directory configuration is the same, but it is a tif file, not a png file.

chartgod

多任务预训练大概需要多少显存

1

happyywq

请问DIOR-R上的mAP50指标是在验证集还是测试集上评估得到的呢

8

Fairydetail