JDet
JDet copied to clipboard
JDet is an object detection benchmark based on Jittor. Mainly focus on aerial image object detection (oriented object detection).
Loading config from: configs/s2anet_r50_fpn_1x_dota.py [w 0828 08:21:51.503818 84 __init__.py:980] load parameter fc.weight failed ... [w 0828 08:21:51.503950 84 __init__.py:980] load parameter fc.bias failed ... [w 0828 08:21:51.504010 84 __init__.py:998] load...
ssl.SSLCertVerificationError: [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: self signed certificate (_ssl.c:1091) 在运行## Data Preprocessing时遇到上述错误,请问是什么原因呢
Wed Aug 25 15:01:57 2021 Validating.... 0%| | 33/41429 [01:43
使用docker安装的jittor,version是1.2.2.59。 不仅nn.Conv3d没有,而且nn.BatchNorm3d,nn.InstanceNorm3d也都没有。
错误如下 Traceback (most recent call last): File "preprocess.py", line 4, in from jdet.config import init_cfg, get_cfg File "/home/vr1/anaconda3/envs/env_ybgu/lib/python3.7/site-packages/jdet-0.1-py3.7.egg/jdet/__init__.py", line 1, in ImportError: cannot import name 'models' from 'jdet' (/home/vr1/anaconda3/envs/env_ybgu/lib/python3.7/site-packages/jdet-0.1-py3.7.egg/jdet/__init__.py) 搜索并未找到解决办法,还请作者指点。
add gwd, bcd, kld loss code
执行 `python tools/preprocess.py --config-file configs/preprocess/dota1_5_preprocess_config.py` 时, 内存占用量在命令开始执行之后若干秒便开始激增, 直到吃满所有系统内存. 系统环境: * Windows 专业版 21H1 19043.1706 * Python 3.10.7 * 内存 32G * 显卡 3070 开始部分日志: (数据集目录名是 `dota-1.0`, 里面的数据使用的是 1.5 版本) ```text...
Extracting jdet-0.2.0.0-py3.7.egg to /environment/miniconda3/lib/python3.7/site-packages File "/environment/miniconda3/lib/python3.7/site-packages/jdet-0.2.0.0-py3.7.egg/jdet/models/roi_heads/rretina_head.py", line 964 outs = def execute(self,features): outs =
你好,服务器上有两张卡。一张3090,一张v100,在测试同一个程序时,v100报错但是3090没报错,请问这个是什么问题呢?我尝试了调小batchsize不行,也试了使用其他框架在这张卡是训练是正常的 发生异常: RuntimeError (note: full exception trace is shown but execution is paused at: _run_module_as_main) [38;5;1m[f 0919 16:27:01.178698 16 executor.cc:665] Execute fused operator(41/49) failed. [JIT Source]: /home/lwb/.cache/jittor/jt1.3.5/g++7.5.0/py3.7.13/Linux-5.15.0-4xe4/IntelRXeonRGolxda/default/cu11.1.74_sm_70_86/jit/cutt_transpose__T_1__JIT_1__JIT_cuda_1__index_t_int32_hash_e6e42f8f6f3e9195_op.cc [OP TYPE]: cutt_transpose...
请问jittor支持FP16混合精度训练吗?有没有相关的示例代码和文档?感谢