赫さん

贴一个好理解的数组解法 ```java class Solution { public int minFallingPathSum(int[][] matrix) { int m = matrix.length; int n = matrix[0].length; int[][] dp = new int[m][n]; // base case for(int i = 0;...

遇见了同样的问题，哪个大大解决了能告诉一下原因或者解决方案嘛？ 应该不是显卡性能和内存的问题

@evilidol @Stephenfang51 非常感谢两位大神，pin_memory改成False之后完美运行不再卡死啦 总结一下遇到训练前几个epoch时卡死的问题： 1.根目录 ---> utils ---> config.py 将文件中相应变量改为 num_workers = 0，test_num_workers = 0 2.根目录 ---> train.py 在train()函数中的test_dataloader(pin_memory=False), 或直接删除pin_memory参数即可 应该是电脑的GPU显存或者内存不能同时处理多线程等问题引起的错误，更改参数减小电脑压力即可 For people who doesn't speak chinese: Thanks so much...

@goldentimecoolk @goldentimecoolk Hi, I also tested on object detection task in dataset Pascal voc, and it largely decrease the performance compared to the default resnet-50, have you solved this problem?

@xmyqsh So do you mean you actually didn't fuse them together in one branch but with each branch goes to a individual RPN and RCNN header even without scale-aware training...

Thanks for the comments. It seems TridentNet split the original resnet into 3 branches and each branch connects to a RPN and RCNN header respectively which means there are 3...

@Leotju Do you mind to share the train file in the github. I think this code is just for testing instead of training cause some of the file for training...