monody1
monody1
check this one id glfwGetCocoaWindow ( GLFWwindow *window )
> hello, sorry to hear this. > > > The agents seem unable to maintain stable connections post-crash, with repeated failures to upgrade connections and repeated logs of agents needing...
I forgot to save the logs before restarting, here is the log after restart [det-agent-server5_partial.log](https://github.com/determined-ai/determined/files/15146510/det-agent-server5_partial.log) [determined_determined-master_partial.log](https://github.com/determined-ai/determined/files/15146511/determined_determined-master_partial.log)
I have been trying the ViT-B/16 architecture on 8 datasets using V100s GPUs (32G) with the main_task_wise_adamerging method. However, I've observed some issues at data loading. Additionally, regarding the loss...
I am using the default batch size of 16, but the code crashes during the data loading phase. The issue seems to occur at ```python x = data['images'].to(args.device) y =...
It can run by making this change, but the memory usage is still inefficient. ```python for epoch in #range(epochs): losses = 0. for dataset_name in exam_datasets: dataset = get_dataset(dataset_name, pretrained_model.val_preprocess,...
你好 我还是没有复现出ViT-B/16的效果, 每次进入这个循环 ```python for dataset_name in exam_datasets: ``` dataloader重新初始化 然后获得一个batch 对吗？ 是不是等价于每个epoch都是从8个数据集里重新采样的，也就是在epoch_i 中 batch from dataset_A 和 epoch_j 中 batch from dataset_A 中的样本是可以重复的对吗? 目前λ在500 epoch下[[1.0000, 0.1601, 0.0774, 0.0510, 0.0546,...
I'm pretty sure i used newest commit and newest version of comfyui, the checkpoint download from readme url. The only error i found in log is clip_l miss some weight....