随笔分类 - Pytorch
摘要：(wind_2021) J:\PytorchProject\yolov5_train_car_2024060501> (wind_2021) J:\PytorchProject\yolov5_train_car_2024060501>python train_20230320.py --img-si
阅读全文
摘要：(wind_2021) J:\PytorchProject\yolov5_train_car_2024051001> (wind_2021) J:\PytorchProject\yolov5_train_car_2024051001> (wind_2021) J:\PytorchProject\yo
阅读全文
摘要：(base) C:\Users\luo> (base) C:\Users\luo>nvidia-smi Thu Mar 21 10:54:51 2024 + + | NVIDIA-SMI 516.94 Driver Version: 516.94 CUDA Version: 11.7 | | + +
阅读全文
摘要：import torch # 加载模型参数 model_path = 'model.pth' model_state_dict = torch.load(model_path) # 创建模型实例 input_size = 10 # 输入层大小，根据实际情况调整 hidden_size1 = 32 #
阅读全文
摘要：import torch import torch.nn as nn import torch.optim as optim # 定义三层神经网络 class ThreeLayerNN(nn.Module): def __init__(self, input_size, hidden_size1,
阅读全文
摘要：1、 (base) C:\Users\Administrator> (base) C:\Users\Administrator> (base) C:\Users\Administrator>conda create -n wind_2023 python==3.7 Collecting packag
阅读全文
摘要：CJJC项目实验环境配置 (base) D:\> (base) D:\>conda create -n wind_2022 python==3.7 Solving environment: done ## Package Plan ## environment location: D:\Anacon
阅读全文
摘要：PaddlePaddle、PyTorch和TensorFlow都是非常流行的深度学习框架，各自有着自己的优势和适用场景。下面是它们的对比： 功能和易用性：PaddlePaddle是由百度开发的开源深度学习框架，它的功能比较全面，支持分布式训练、模型压缩、自动求导等高级功能，同时也提供了较为简单易用的
阅读全文
摘要：yolov5训练栏杆检测模型日志记录 (wind_2021) H:\PytorchProject\yolov5_train_xinru_2023050901> (wind_2021) H:\PytorchProject\yolov5_train_xinru_2023050901> (wind_202
阅读全文
摘要：训练yolov5 模型 (wind_2021) H:\PytorchProject\yolov5_train_xiaotian_2023031401> (wind_2021) H:\PytorchProject\yolov5_train_xiaotian_2023031401>python trai
阅读全文
摘要：alphapose win10运行日志 (base) H:\PytorchProject\AlphaPose-master> (base) H:\PytorchProject\AlphaPose-master> (base) H:\PytorchProject\AlphaPose-master>co
阅读全文
摘要：YOLOv7_OBB 运行异常记录 (wind_2021) H:\PytorchProject\YOLOv7_OBB-main> (wind_2021) H:\PytorchProject\YOLOv7_OBB-main>python detect.py --weights 'models/yolo
阅读全文
摘要：yolov5训练洞口检测模型 1、标注数据 2、整理数据 3、训练：修改：myvoc.yaml 4、开始训练 (wind_2021) H:\PytorchProject\yolov5_train_hole> (wind_2021) H:\PytorchProject\yolov5_train_hol
阅读全文
摘要：1、创建环境 conda create -n wind_2022 python==3.7 (base) D:\> (base) D:\> (base) D:\> (base) D:\>conda create -n wind_2022 python==3.7 Collecting package m
阅读全文
摘要：error log (wind_torch_2022) bim@bim-PowerEdge-R730:~/Pytorch_Project/yolov5_train_warning$ python train_20220602.py --batch-size 1 --epochs 300 --data
阅读全文
摘要：1、标注数据 2、整理数据 3、训练：修改：myvoc.yaml train: VOC_2022063001/train.txt val: VOC_2022063001/val.txt # number of classes nc: 1 # class names names: ["rectz_ci
阅读全文
摘要：yolov5训练识别人、反光衣、安全带、安全绳、安全帽 1、标注数据 2、整理数据 3、训练：修改：myvoc.yaml train: VOC_2022061401/train.txt val: VOC_2022061401/val.txt # number of classes nc: 5 # c
阅读全文
摘要：yolov5训练日志 python train_20220613.py --batch-size 2 --epochs 220 --data ./data/myvoc.yaml --cfg ./models/yolov5m.yaml --workers 0 (wind_2021) F:\Pytorc
阅读全文
摘要：yolov5训练表示识别模型日志记录 1、标注数据 2、整理数据 3、训练：修改：myvoc.yaml train: VOC_2022060901/train.txt val: VOC_2022060901/val.txt # number of classes nc: 1 # class name
阅读全文
摘要：yolov5模型训练过程中显卡使用率查询——记录 vidia-smi (base) F:\> (base) F:\>nvidia-smi Thu Jun 9 14:02:16 2022 + + | NVIDIA-SMI 497.17 Driver Version: 497.17 CUDA Versi
阅读全文
