【CondInst】 の 填坑指南

教程目录

问题1：gnutls_handshake() failed: A TLS packet with unexpected length was received.

解决方案：参考链接： 问题2：debug模式报错importError: dlopen: cannot load any more object with static TLS

解决方案：参考链接： 问题3：Error in `xxx/bin/python': double free or corruption (!prev): 0x0000561e68638bb0

解决方案：参考链接：

---

问题1：gnutls_handshake() failed: A TLS packet with unexpected length was received.

fatal: unable to access ‘https://github.com/facebookresearch/detectron2.git/’: gnutls_handshake() failed: A TLS packet with unexpected length was received.

解决方案：

执行以下代码：

git config --global http.postBuffer 2000000000

参考链接：

gnutls_handshake() failed: A TLS packet with unexpected length was received

---

问题2：debug模式报错importError: dlopen: cannot load any more object with static TLS

File “/data3/zhanghuiyong/anaconda3/envs/CondInst/lib/python3.6/site-packages/cv2/init.py”, line 8, in
from .cv2 import *
importError: dlopen: cannot load any more object with static TLS

解决方案：

importError: dlopen:cannot load any more object with static TLS 的解决办法

参考链接：

【YOLOX】の 填坑指南

---

问题3：Error in `xxx/bin/python’: double free or corruption (!prev): 0x0000561e68638bb0

[01/09 12:30:14 adet.trainer]: Starting training from iteration 0
Error in /data3/zhanghuiyong/anaconda3/envs/CondInst/bin/python’: double free or corruption (!prev): 0x0000561e68638bb0
Error in `/data3/zhanghuiyong/anaconda3/envs/CondInst/bin/python’: double free or corruption (!prev): 0x0000561e68638bb0
Traceback (most recent call last):
File “/data3/zhanghuiyong/anaconda3/envs/CondInst/lib/python3.6/site-packages/torch/utils/data/dataloader.py”, line 986, in _try_get_data
data = self._data_queue.get(timeout=timeout)
File “/data3/zhanghuiyong/anaconda3/envs/CondInst/lib/python3.6/multiprocessing/queues.py”, line 104, in get
if not self._poll(timeout):
File “/data3/zhanghuiyong/anaconda3/envs/CondInst/lib/python3.6/site-packages/torch/utils/data/_utils/signal_handling.py”, line 66, in handler
_error_if_any_worker_fails()
RuntimeError: DataLoader worker (pid 10736) is killed by signal: Aborted.

解决方案：

修改NUM_WORKERS的值，在CondInst中可通过修改 ./detectron2/detectron2/config/defaults.py 文件中的值。其他网络模型可在基本框架的dataloader处查找。在Debug调试时，将原本的值 4 改为0，训练时需要改回原值。

参考链接：

tensorflow 运行时候遇到 Error in `python’: double free or corruption (fasttop）使用PyTorch版的yolov5训练时，出现Error in `python’: double free or corruption (!prev)错误的解决方式Error in `python’: double free or corruption (!prev)