PromptCC icon indicating copy to clipboard operation
PromptCC copied to clipboard

PyTorch implementation of 'A Decoupling Paradigm With Prompt Learning for Remote Sensing Image Change Captioning'

Results 6 PromptCC issues
Sort by recently updated
recently updated
newest added

Thanks for your interesting work. May you share the codes for classification part. Thanks.

作者,你好。 我在复现你的网络,train.py跑到1000iter进行验证的时候出现这个问题。 Traceback (most recent call last): File "F:/0-code/image-description/Chg2Cap-main/test.py", line 220, in main(args) File "F:/0-code/image-description/Chg2Cap-main/test.py", line 139, in main nochange_metric = get_eval_score(nochange_references, nochange_hypotheses) File "F:\0-code\image-description\Chg2Cap-main\utils.py", line 75, in get_eval_score score_i,...

作者您好,我在推理和训练时都遇到了这个问题: `File "/PromptCC/models_CC.py", line 264, in forward clip_emb_A, img_feat_A = self.clip_model.encode_image(img_A) ValueError: too many values to unpack (expected 2, got 1)` 看起来是因为原始的CLIP输出的是一整张图的特征,而这里的img_feat_A似乎是(N, h*w, 512)的特征 请问您可以分享下是怎么修改CLIP让它输出patch-level的图像特征吗? 感谢!

Good job,I am very interested in change caption task and your job promptcc.And I want to ask when to release the code

您好老师,我是一个初学者。在按照readme中put cls_model.pth.tar in ./checkpoints/classification_model/ and put BEST_checkpoint_ViT-B_32.pth.tar in ./checkpoints/cap_model/.这一步后,运行caption_beams.py出现了如下错误: 请问运行代码是否必须连接国外网站,由于条件限制,我现在的地方无法连接国外网站,有什么替代方法吗,恳请老师指导

您好,我尝试训练分类网络,但训练的分类网络在训练集与测试集上性能差距较大。请问您有遇到这种问题吗?