From 02957cb8bf02a26440ca4b867072795a60a00e2e Mon Sep 17 00:00:00 2001 From: liuzx Date: Mon, 1 Jul 2024 11:12:41 +0800 Subject: [PATCH 1/3] =?UTF-8?q?=E6=9B=B4=E6=96=B0=20'gcu=5Fmnist=5Fexample?= =?UTF-8?q?/train.py'?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- gcu_mnist_example/train.py | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/gcu_mnist_example/train.py b/gcu_mnist_example/train.py index 44766a0..1aa91e6 100644 --- a/gcu_mnist_example/train.py +++ b/gcu_mnist_example/train.py @@ -52,7 +52,7 @@ if __name__ == '__main__': #获取数据集路径 MnistDataset_torch_path = c2net_context.dataset_path+"/"+"MnistDataset_torch" #获取预训练模型路径 - Torch_MNIST_Example_Model_path = c2net_context.pretrain_model_path+"/"+"Torch_MNIST_Example_Model" + Torch_MNIST_Example_Model_path = c2net_context.pretrain_model_path+"/"+"GCU_MNIST_Example_Model" #获取输出路径 output_path = c2net_context.output_path # load DPU envs-xx.sh @@ -88,8 +88,8 @@ if __name__ == '__main__': print('epoch_size is:{}'.format(epochs)) # 如果有保存的模型，则加载模型，并在其基础上继续训练 - if os.path.exists(os.path.join(Torch_MNIST_Example_Model_path, "mnist_epoch1.pkl")): - checkpoint = torch.load(os.path.join(Torch_MNIST_Example_Model_path, "mnist_epoch1.pkl")) + if os.path.exists(os.path.join(Torch_MNIST_Example_Model_path, "mnist_epoch1_0.81.pkl")): + checkpoint = torch.load(os.path.join(Torch_MNIST_Example_Model_path, "mnist_epoch1_0.81.pkl")) model.load_state_dict(checkpoint['model']) optimizer.load_state_dict(checkpoint['optimizer']) start_epoch = checkpoint['epoch'] From daceae587ffa8bee6e6f22f33b906e160a971e06 Mon Sep 17 00:00:00 2001 From: liuzx Date: Mon, 1 Jul 2024 11:23:42 +0800 Subject: [PATCH 2/3] =?UTF-8?q?=E6=9B=B4=E6=96=B0=20'gcu=5Fmnist=5Fexample?= =?UTF-8?q?/README.md'?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- gcu_mnist_example/README.md | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/gcu_mnist_example/README.md b/gcu_mnist_example/README.md index a3d81ae..1317f07 100644 --- a/gcu_mnist_example/README.md +++ b/gcu_mnist_example/README.md @@ -18,12 +18,12 @@ ##### 2，预训练模型说明： -Torch_MNIST_Example_Model模型可从本项目的模型目录中引用，[预训练模型引用](https://openi.pcl.ac.cn/OpenIOSSG/OpenI_Cloudbrain_Example/modelmanage/show_model) +GCU_MNIST_Example_Model模型可从本项目的模型目录中引用，[预训练模型引用](https://openi.pcl.ac.cn/OpenIOSSG/OpenI_Cloudbrain_Example/modelmanage/show_model) -* Torch_MNIST_Example_Model模型的目录结构如下： +* GCU_MNIST_Example_Model模型的目录结构如下： - > Torch_MNIST_Example_Model - > ├── mnist_epoch1.pkl + > GCU_MNIST_Example_Model + > ├── mnist_epoch1_0.81.pkl > ## 二. 如何在云脑上获取代码路径，数据集路径，预训练模型路径，输出路径 @@ -95,7 +95,7 @@ upload_output() | 数据集 | 数据集选择MnistDataset_torch.zip | | 运行参数 | 选择增加运行参数可以向脚本中其他参数传值，如epoch_size,需要在代码里定义增加的超参数 | | 资源规格 | 规格选择[GCU] | -| 模型 | 模型选择Torch_MNIST_Example_Model | +| 模型 | 模型选择GCU_MNIST_Example_Model | 启动调试任务后，先执行prepare()进行数据准备； @@ -113,7 +113,7 @@ upload_output() | 数据集 | 数据集选择MnistDataset_torch.zip | | 运行参数 | 选择增加运行参数可以向脚本中其他参数传值，如epoch_size,需要在代码里定义增加的超参数 | | 资源规格 | 规格选择[GCU] | -| 模型 | 模型选择Torch_MNIST_Example_Model | +| 模型 | 模型选择GCU_MNIST_Example_Model | 启动训练任务后，训练结束会在任务的结果下载页提供输出结果下载 From 4e899a17b43bd8e63815c123bb93b62ee288323f Mon Sep 17 00:00:00 2001 From: liuzx Date: Mon, 1 Jul 2024 11:24:03 +0800 Subject: [PATCH 3/3] =?UTF-8?q?=E6=9B=B4=E6=96=B0=20'.openi-aitemplate'?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- .openi-aitemplate | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/.openi-aitemplate b/.openi-aitemplate index efc19b2..a2e571e 100644 --- a/.openi-aitemplate +++ b/.openi-aitemplate @@ -310,8 +310,8 @@ }, "PretrainModelList": [ { - "ID": "f1711e48-65d6-436c-b34b-9698061979e7", - "ModelName": "Torch_MNIST_Example_Model" + "ID": "bbd88733-ebeb-4009-b44b-9797480bde8e", + "ModelName": "GCU_MNIST_Example_Model" } ], "DatasetList": [