flowGMM

1. conda create -n flow python=3.7
2. conda activate flow
3. conda install pytorch=1.0.1 torchvision=0.2.1 -c pytorch
4. pip install tensorboardX
5. 创建日志和检查点目录：
   mkdir -p /root/flowgmm/logs mkdir -p /root/flowgmm/checkpoints
6. python3 experiments/train_flows/train_semisup_cons.py --dataset=CIFAR10 --data_path=data/images/cifar/cifar10/by-image/ --label_path=data/labels/cifar10/4000_balanced_labels/10.txt --logdir=/root/flowgmm/logs --ckptdir=/root/flowgmm/checkpoints --save_freq=500 --num_epochs=1501 --label_weight=3 --consistency_weight=1. --consistency_rampup=100 --lr=1e-4 --eval_freq=50
7. 报错：
   ImportError: /lib/x86_64-linux-gnu/libstdc++.so.6: version `GLIBCXX_3.4.29' not found

这个错误是由于你的系统上缺少特定版本的 libstdc++ 库。你可以尝试更新你的 libstdc++ 来解决这个问题：
sudo apt-get update sudo apt-get install libstdc++6
8. ModuleNotFoundError: No module named 'seaborn'
pip install seaborn
9. ModuleNotFoundError: No module named 'tensorflow'
pip install tensorflow==1.13.1
10. TypeError: Descriptors cannot not be created directly.
pip install protobuf==3.20.0
11. ImportError: cannot import name 'container_abcs' from 'torch._six' (/root/miniconda3/envs/CircleNet/lib/python3.7/site-packages/torch/_six.py)
改具体内容
12. File "/root/flowgmm/flow_ssl/invertible/iresnet_trash/mixed_lipschitz.py", line 1, in
from torch._six import container_abcs
ImportError: cannot import name 'container_abcs' from 'torch._six' (/root/miniconda3/envs/CircleNet/lib/python3.7/site-packages/torch/_six.py)
同上12
13. File "/root/flowgmm/flow_ssl/data/ssl_data_utils.py", line 39, in make_ssl_data_loaders
*args, **kwargs)
TypeError: make_ssl_cifar_data_loaders() got an unexpected keyword argument 'return_all_labels'
该具体内容。
14. File "/root/flowgmm/flow_ssl/realnvp/coupling_layer.py", line 143, in init
init_zeros=init_zeros)
TypeError: init() got an unexpected keyword argument 'init_zeros'
该具体内容