{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":685554848,"defaultBranch":"master","name":"mmsegmentation","ownerLogin":"HeYun1999","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2023-08-31T13:46:37.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/117512267?v=4","public":true,"private":false,"isOrgOwned":false},"refInfo":{"name":"","listCacheKey":"v0:1695367159.0","currentOid":""},"activityList":{"items":[{"before":"803e3ff775c88b4979e852926ae7124667e74b59","after":"7a46bdfc4fb1661df1508ca97d31fb145c147f6c","ref":"refs/heads/master","pushedAt":"2023-10-27T06:45:52.000Z","pushType":"push","commitsCount":6,"pusher":{"login":"HeYun1999","name":null,"path":"/HeYun1999","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/117512267?s=80&v=4"},"commit":{"message":"更改net 与head ，在net中增加融合模块，使两block相加后在cat，但通道数过多，之后再减","shortMessageHtmlLink":"更改net 与head ，在net中增加融合模块，使两block相加后在cat，但通道数过多，之后再减"}},{"before":"04affbfd845276fd1f76be576b0adcd183d688d6","after":"803e3ff775c88b4979e852926ae7124667e74b59","ref":"refs/heads/master","pushedAt":"2023-10-12T06:29:32.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"HeYun1999","name":null,"path":"/HeYun1999","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/117512267?s=80&v=4"},"commit":{"message":"加入transresnet 模型","shortMessageHtmlLink":"加入transresnet 模型"}},{"before":"b2554d02cebb37f3c01f2be398635058e0bf2873","after":"04affbfd845276fd1f76be576b0adcd183d688d6","ref":"refs/heads/master","pushedAt":"2023-10-08T09:58:43.000Z","pushType":"push","commitsCount":15,"pusher":{"login":"HeYun1999","name":null,"path":"/HeYun1999","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/117512267?s=80&v=4"},"commit":{"message":"Merge branch 'decoupling_no_atten'\n\n# Conflicts:\n#\tmmseg/models/decode_heads/segformer_syatten_head.py","shortMessageHtmlLink":"Merge branch 'decoupling_no_atten'"}},{"before":"e5148937e3250b4176a1acabbdb028613d49e2a8","after":"4480b4b6e14987b6b6d2146c361f861e0150675c","ref":"refs/heads/decoupling_no_atten","pushedAt":"2023-10-08T06:13:12.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"HeYun1999","name":null,"path":"/HeYun1999","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/117512267?s=80&v=4"},"commit":{"message":"加入aux decoder 加入vaihingen数据集的代码","shortMessageHtmlLink":"加入aux decoder 加入vaihingen数据集的代码"}},{"before":null,"after":"e5148937e3250b4176a1acabbdb028613d49e2a8","ref":"refs/heads/decoupling_no_atten","pushedAt":"2023-09-22T07:19:19.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"HeYun1999","name":null,"path":"/HeYun1999","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/117512267?s=80&v=4"},"commit":{"message":"去除注意力，修改decoder学习率，由0.00006变为0.0006","shortMessageHtmlLink":"去除注意力，修改decoder学习率，由0.00006变为0.0006"}},{"before":"9cd32cb06b77c3400bd142582c3499be2200d03e","after":"b8d2d9aedc6d3a667917a265fa27d32e5f9f1b9b","ref":"refs/heads/deecoupling","pushedAt":"2023-09-21T10:22:44.000Z","pushType":"push","commitsCount":4,"pusher":{"login":"HeYun1999","name":null,"path":"/HeYun1999","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/117512267?s=80&v=4"},"commit":{"message":"增加segformerv1模型，进行新的特征解耦尝试，decoupling部分解耦边缘信息，因此对label进行边缘信息的提取工作process_label。成功跑通","shortMessageHtmlLink":"增加segformerv1模型，进行新的特征解耦尝试，decoupling部分解耦边缘信息，因此对label进行边缘信息的提取工作proc…"}},{"before":"a4baf3b6a9a6125e9a42923774d70b94f0341063","after":"9cd32cb06b77c3400bd142582c3499be2200d03e","ref":"refs/heads/deecoupling","pushedAt":"2023-09-18T07:59:05.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"HeYun1999","name":null,"path":"/HeYun1999","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/117512267?s=80&v=4"},"commit":{"message":"改lovazloss函数","shortMessageHtmlLink":"改lovazloss函数"}},{"before":"f7de58c8115e1fc04391801179d81471d413e82e","after":"a4baf3b6a9a6125e9a42923774d70b94f0341063","ref":"refs/heads/deecoupling","pushedAt":"2023-09-16T14:26:27.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"HeYun1999","name":null,"path":"/HeYun1999","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/117512267?s=80&v=4"},"commit":{"message":"给部分模块增加开关","shortMessageHtmlLink":"给部分模块增加开关"}},{"before":"6a6b98226e242bcf0f9315971fe116d28e948c39","after":"b2554d02cebb37f3c01f2be398635058e0bf2873","ref":"refs/heads/master","pushedAt":"2023-09-16T12:13:19.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"HeYun1999","name":null,"path":"/HeYun1999","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/117512267?s=80&v=4"},"commit":{"message":"取消syatten，以及从8000改为2000代输出一次指标","shortMessageHtmlLink":"取消syatten，以及从8000改为2000代输出一次指标"}},{"before":"6d7bb48687f248b240c8cd5a0bc002b0d1f97ab7","after":"f7de58c8115e1fc04391801179d81471d413e82e","ref":"refs/heads/deecoupling","pushedAt":"2023-09-16T11:54:12.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"HeYun1999","name":null,"path":"/HeYun1999","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/117512267?s=80&v=4"},"commit":{"message":"decoupling结构会使指标出现nan与0，模型暂定","shortMessageHtmlLink":"decoupling结构会使指标出现nan与0，模型暂定"}},{"before":"e9c375416eec77cf54e7de73297a9ee5c874cd4c","after":"6d7bb48687f248b240c8cd5a0bc002b0d1f97ab7","ref":"refs/heads/deecoupling","pushedAt":"2023-09-15T12:58:57.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"HeYun1999","name":null,"path":"/HeYun1999","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/117512267?s=80&v=4"},"commit":{"message":"修改decoupling模型，在提取部分特征且用第二loss计算损失函数时，在主路支路的交点，禁止回传，以及做标注，得知在mmseg中dataset中可设置loss的ignore_index，在config中dataset中的val_evaluator中可设置评估指标的ignore_index","shortMessageHtmlLink":"修改decoupling模型，在提取部分特征且用第二loss计算损失函数时，在主路支路的交点，禁止回传，以及做标注，得知在mmseg中da…"}},{"before":null,"after":"e9c375416eec77cf54e7de73297a9ee5c874cd4c","ref":"refs/heads/deecoupling","pushedAt":"2023-09-14T02:26:40.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"HeYun1999","name":null,"path":"/HeYun1999","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/117512267?s=80&v=4"},"commit":{"message":" 优化decoupling结构，消减参数量，将segformer的mlp层的输出通道从256变到128，用于消减通道数","shortMessageHtmlLink":" 优化decoupling结构，消减参数量，将segformer的mlp层的输出通道从256变到128，用于消减通道数"}},{"before":"d46bcf052468d11369ba4709fb1bf83be9d8470f","after":"7c367a277d9f62fac16f1c03f6a7e3ea132bf1e5","ref":"refs/heads/error","pushedAt":"2023-09-13T02:57:55.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"HeYun1999","name":null,"path":"/HeYun1999","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/117512267?s=80&v=4"},"commit":{"message":"改命名syatten为decoupling，以及去除syatten，改用cotatten，以及第二损失函数改为lovaloss","shortMessageHtmlLink":"改命名syatten为decoupling，以及去除syatten，改用cotatten，以及第二损失函数改为lovaloss"}},{"before":"7f346d8d6c369c291ba76826e24a1110a9664bf6","after":"6a6b98226e242bcf0f9315971fe116d28e948c39","ref":"refs/heads/master","pushedAt":"2023-09-07T13:51:52.000Z","pushType":"push","commitsCount":3,"pusher":{"login":"HeYun1999","name":null,"path":"/HeYun1999","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/117512267?s=80&v=4"},"commit":{"message":"加入se注意力机制","shortMessageHtmlLink":"加入se注意力机制"}},{"before":null,"after":"d46bcf052468d11369ba4709fb1bf83be9d8470f","ref":"refs/heads/error","pushedAt":"2023-09-07T13:51:52.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"HeYun1999","name":null,"path":"/HeYun1999","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/117512267?s=80&v=4"},"commit":{"message":"加入解耦网络，设置两个损失函数，输出预测结果为个列表","shortMessageHtmlLink":"加入解耦网络，设置两个损失函数，输出预测结果为个列表"}},{"before":"7764297e5603098356d2ff9eea315d575dba7d06","after":"7f346d8d6c369c291ba76826e24a1110a9664bf6","ref":"refs/heads/master","pushedAt":"2023-09-04T09:16:43.000Z","pushType":"push","commitsCount":2,"pusher":{"login":"HeYun1999","name":null,"path":"/HeYun1999","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/117512267?s=80&v=4"},"commit":{"message":"对上次修改效果的补充","shortMessageHtmlLink":"对上次修改效果的补充"}},{"before":"f35801fca3f040948010bf849bd205089be3b3af","after":"7764297e5603098356d2ff9eea315d575dba7d06","ref":"refs/heads/master","pushedAt":"2023-09-02T09:34:30.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"HeYun1999","name":null,"path":"/HeYun1999","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/117512267?s=80&v=4"},"commit":{"message":"加入taiyuan——city数据集，并将G:\\project\\mmsegmentation\\mmseg\\evaluation\\metrics\\iou_metric.py中的ignire_index从255改为9，并加入其他评估指标","shortMessageHtmlLink":"加入taiyuan——city数据集，并将G:\\project\\mmsegmentation\\mmseg\\evaluation\\metri…"}},{"before":null,"after":"f35801fca3f040948010bf849bd205089be3b3af","ref":"refs/heads/master","pushedAt":"2023-08-31T13:47:27.000Z","pushType":"branch_creation","commitsCount":0,"pusher":{"login":"HeYun1999","name":null,"path":"/HeYun1999","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/117512267?s=80&v=4"},"commit":{"message":"创建sourcetree，同时对isprs.py中的ISPRSDataset类，修改了图像扩展名png-》tif","shortMessageHtmlLink":"创建sourcetree，同时对isprs.py中的ISPRSDataset类，修改了图像扩展名png-》tif"}}],"hasNextPage":false,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"Y3Vyc29yOnYyOpK7MjAyMy0xMC0yN1QwNjo0NTo1Mi4wMDAwMDBazwAAAAOhFtz1","startCursor":"Y3Vyc29yOnYyOpK7MjAyMy0xMC0yN1QwNjo0NTo1Mi4wMDAwMDBazwAAAAOhFtz1","endCursor":"Y3Vyc29yOnYyOpK7MjAyMy0wOC0zMVQxMzo0NzoyNy4wMDAwMDBazwAAAAN2qpc_"}},"title":"Activity · HeYun1999/mmsegmentation"}