在cpu上运行官方文档上的lenet demo分布式训练Parameter Server模式失败

Environment

Hardware Environment(`Ascend`/`GPU`/`CPU`):

/device cpu

Software Environment:

MindSpore version (source or binary): 1.1.0
Python version (e.g., Python 3.7.5): 3.7
OS platform and distribution (e.g., Linux Ubuntu 16.04): windows 10

Describe the current behavior

我从gitee上clone了mindspore项目，然后想根据官方文档将lenet demo修改成Parameter Server模式失败了

Describe the expected behavior

我可以正常运行起来

Steps to reproduce the issue

通过git clone 下载代码
用pycharm打开 model_zoo/official/cv/lenet 目录 3.修改model_zoo/official/cv/lenet/train.py 文件 4.下载mnist训练数据 5.运行代码，启动参数和环境变量如下

Related log / screenshot

D:\workspace\mindspore-test\venv\Scripts\python.exe D:/workspace/github/mindspore/model_zoo/official/cv/lenet/train.py --device_target=CPU --data_path=mnist_data --ckpt_path=checkpoint ============== Starting Training ============== Traceback (most recent call last): File "D:/workspace/github/mindspore/model_zoo/official/cv/lenet/train.py", line 70, in model.train(cfg['epoch_size'], ds_train, callbacks=[time_cb, ckpoint_cb, LossMonitor()], dataset_sink_mode=False) File "D:\workspace\mindspore-test\venv\lib\site-packages\mindspore\train\model.py", line 592, in train sink_size=sink_size) File "D:\workspace\mindspore-test\venv\lib\site-packages\mindspore\train\model.py", line 385, in _train self._train_process(epoch, train_dataset, list_callback, cb_params) File "D:\workspace\mindspore-test\venv\lib\site-packages\mindspore\train\model.py", line 513, in _train_process outputs = self._train_network(next_element) File "D:\workspace\mindspore-test\venv\lib\site-packages\mindspore\nn\cell.py", line 331, in call out = self.compile_and_run(inputs) File "D:\workspace\mindspore-test\venv\lib\site-packages\mindspore\nn\cell.py", line 588, in compile_and_run self.compile(inputs) File "D:\workspace\mindspore-test\venv\lib\site-packages\mindspore\nn\cell.py", line 575, in compile _executor.compile(self, inputs, phase=self.phase, auto_parallel_mode=self._auto_parallel_mode) File "D:\workspace\mindspore-test\venv\lib\site-packages\mindspore\common\api.py", line 502, in compile result = self._executor.compile(obj, args_list, phase, use_vm) RuntimeError: mindspore\ccsrc\runtime\device\cpu\kernel_select_cpu.cc:299 SetKernelInfo] Operator[Push] is not support. Trace: In file D:\workspace\mindspore-test\venv\lib\site-packages\mindspore\nn\optim\momentum.py(36)/ success = F.depend(success, _ps_pull(_ps_push((learning_rate, gradient, momentum), shapes), weight))/ In file D:\workspace\mindspore-test\venv\lib\site-packages\mindspore\nn\optim\momentum.py(151)/ success = self.hyper_map(F.partial(_momentum_opt, self.opt, self.momentum, lr), gradients, params, moments,/ In file D:\workspace\mindspore-test\venv\lib\site-packages\mindspore\nn\wrap\cell_wrapper.py(251)/ return F.depend(loss, self.optimizer(grads))/

WARNING: Logging before InitGoogleLogging() is written to STDERR [WARNING] KERNEL(17656,?):2021-2-5 10:12:20 [mindspore\ccsrc\backend\kernel_compiler\cpu\cpu_kernel_factory.cc:92] GetSupportedKernelAttrList] Not registered CPU kernel: op[Push]! [ERROR] DEVICE(17656,?):2021-2-5 10:12:20 [mindspore\ccsrc\runtime\device\cpu\kernel_select_cpu.cc:299] SetKernelInfo] Operator[Push] is not support. Trace: In file D:\workspace\mindspore-test\venv\lib\site-packages\mindspore\nn\optim\momentum.py(36)/ success = F.depend(success, _ps_pull(_ps_push((learning_rate, gradient, momentum), shapes), weight))/ In file D:\workspace\mindspore-test\venv\lib\site-packages\mindspore\nn\optim\momentum.py(151)/ success = self.hyper_map(F.partial(_momentum_opt, self.opt, self.momentum, lr), gradients, params, moments,/ In file D:\workspace\mindspore-test\venv\lib\site-packages\mindspore\nn\wrap\cell_wrapper.py(251)/ return F.depend(loss, self.optimizer(grads))/

mindspore-ai / mindspore