mindyolo
mindyolo copied to clipboard
MindSpore YOLO series toolbox and benchmark
这是 2024-03-29 17:11:44,155 [INFO] Dataset Cache file hash/version check success. 2024-03-29 17:11:44,155 [INFO] Load dataset cache from [coco\val\val2017.cache.npy] success. 2024-03-29 17:11:44,155 [INFO] Dataloader num parallel workers: [1] 2024-03-29 17:11:44,254 [INFO]...
* 从下面这个函数来看, 这个学习率使用的是固定学习率? https://github.com/mindspore-lab/mindyolo/blob/8228636c35b52d571d63713af7f2eb199e2698c1/mindyolo/optim/scheduler.py#L7 * 在log 中的学习率,是一个动态变化的学习率  * 在gpu上 visdrone yolov5 loss 一直不下降 
训练自定义数据集
自定义一个车辆检测的训练数据集,为yolo格式,在yolov5-6.2框架上训练无问题,但今日使用mindyolo进行训练出现一定问题。 本机平台平台ubuntu22.04,显卡3080ti,cuda12.0,采用docker方式拉取cuda11.6版本的mindspore镜像,验证GPU版本安装成功,而后安装mindyolo。 训练指令: `python train.py --config ./config/yolov5/yolov5s.yaml --device_target GPU` 1.ValueError: invalid literal for int() with base 10:"xxxxxx" 对应错误位置:/mindyolo/mindyolo/data/dataset.py Line 198 `self.imgIds = [int(Path(im_file).stem) for im_file in self.img_files]` 由于采用自定义数据集,图片并非按照 int value.jpg...
mindspore:2.2.10 cann:7.1.0.3 /device Ascend 910B /device CPU kunpeng920 执行步骤:  关联用例 代码链接:https://github.com/mindspore-lab/mindyolo/tree/master/configs/yolov8  
运行demo/predict.py 报错 self._context_handle.set_param(param, value) RuntimeError: Unsupported device target Ascend. This process only supports one of the ['CPU']. Please check whether the Ascend environment is installed and configured correctly, and check...
版本信息: CANN 7.0.0 mindspore 2.2.10 存在overflow问题,性能存在劣化,前45步时间正常0.7s左右,后面每step时间基本在30s左右
1.环境配置:modelart(mindspore_1.10.0-cann_6.0.1-py_3.7-euler_2.8.3镜像) ,EulerOS 2.0 (SP8), CANN-6.0.1,mindspore1.10, mindyolo r0.1。 2.数据集制作及训练过程文档:https://github.com/mindspore-lab/mindyolo/blob/master/examples/finetune_SHWD/README.md 3.训练过程中出现报错: **RuntimeError: For 'load_param_into_net', model.model.77.m.0.weight in the argument 'net' should have the same shape as model.model.77.m.0.weight in the argument 'parameter_dict'. But got...
**If this is your first time, please read our contributor guidelines:** 如果这是您第一次提交请先阅读我们提供的贡献指南 https://gitee.com/mindspore/mindspore/blob/master/CONTRIBUTING.md **Is your feature request related to a problem? Please describe.** 您的需求和某个报错有关吗?请描述您的问题 A clear and concise description of...
【附:单卡可以跑通,npu使用正常】 版本信息: 驱动固件:7.0.0 cann:7.0.rc1.3 mindspore:2.2.1
环境信息: CANN-6.3.RC2 Mindspore 2.1.1 Mindspore-lite 2.0.0a0 操作参考文档: https://github.com/mindspore-lab/mindyolo/tree/master/deploy 执行单张图片推理和COCO数据集推理都报错,报错信息如下: Traceback (most recent call last): File "deploy/predict.py", line 217, in [infer(args)](url) File "deploy/predict.py", line 174, in infer network = LiteModel(args.model_path) File...