- 072805d929 update ut
- 5f741533fb 更新 '.drone.yml' (#920) 更新 '.drone.yml' Reviewed-on: https://openi.pcl.ac.cn/OpenI/MSAdapter/pulls/920
- a8452a4428 update ut
- Compare 3 commits »
Erpim pushed to erpim_0409 at OpenI/MSAdapter
20 hours ago
Erpim pushed to erpim_0409 at OpenI/MSAdapter
1 day ago
Erpim closed issue OpenI/MSAdapter#915
Cannot find reference 'mstorch_enable' in '__init__.py'1 day ago
Erpim closed issue OpenI/MSAdapter#869
ModuleNotFoundError: No module named 'mindtorch.torch.distributed.algorithms'1 day ago
Erpim closed issue OpenI/MSAdapter#863
ModuleNotFoundError: No,module,named 'mindtorch,torch.utils,tensorboard"1 day ago
Erpim commented on pull request OpenI/MSAdapter#918
[WIP]load/save fix> > import torch > > a = torch.tensor(2.) > > > > a.kkk = 3 > > > > pth = torch.save(a, './pt.pth') > > > > tensor = torch.load('./pt.pth') > > > > print(tensor) > > print(tensor.kkk) > > > > > > 这个用例可以跑过。 但是mindtorch无法恢复出tensor.kkk这个值 > > > > import torch > a = torch.tensor(2.) > > a.kkk = 3 > > pth = torch.save(a, './pt.pth') > > import mindtorch.torch as ms_torch > > tensor = ms_torch.load('./pt.pth') > > print(tensor) > print(tensor.kkk) 需新增_set_obj_state _rebuild_from_type_v2中调用
1 day ago
Erpim commented on pull request OpenI/MSAdapter#918
[WIP]load/save fix用torch_disable来解决该类问题
1 day ago
Erpim commented on pull request OpenI/MSAdapter#918
[WIP]load/save fix需验证,加载torch的pth,保存mindtorch的pth,再加载回来,整个流程的正确性
1 day ago
Erpim commented on pull request OpenI/MSAdapter#918
[WIP]load/save fix这部分data_ptr的作用是否和内存访问相关,频繁调用性能较差
1 day ago
Erpim commented on pull request OpenI/MSAdapter#918
[WIP]load/save fix当前有应该不支持as_subclass?什么场景会进这个函数?
1 day ago
Erpim commented on pull request OpenI/MSAdapter#918
[WIP]load/save fix补充_check_container_source实现
1 day ago
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》