/home/HwHiAiUser/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for <class ‘numpy.float64’> type is zero.
setattr(self, word, getattr(machar, word).flat[0])
/home/HwHiAiUser/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class ‘numpy.float64’> type is zero.
return self._float_to_str(self.smallest_subnormal)
/home/HwHiAiUser/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:549: UserWarning: The value of the smallest subnormal for <class ‘numpy.float32’> type is zero.
setattr(self, word, getattr(machar, word).flat[0])
/home/HwHiAiUser/.local/lib/python3.9/site-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class ‘numpy.float32’> type is zero.
return self._float_to_str(self.smallest_subnormal)
[WARNING] ME(8134:255086668574752,MainProcess):2025-09-04-16:22:35.406.389 [mindspore/context.py:1402] For ‘context.set_context’, the parameter ‘ascend_config’ will be deprecated and removed in a future version. Please use the api mindspore.device_context.ascend.op_precision.precision_mode(),
mindspore.device_context.ascend.op_precision.op_precision_mode(),
mindspore.device_context.ascend.op_precision.matmul_allow_hf32(),
mindspore.device_context.ascend.op_precision.conv_allow_hf32(),
mindspore.device_context.ascend.op_tuning.op_compile() instead.
Traceback (most recent call last):
File “/home/HwHiAiUser/mindnlp/llm/inference/janus_pro/understanding.py”, line 4, in
from mindnlp.transformers import AutoModelForCausalLM
File “/home/HwHiAiUser/mindnlp/mindnlp/init.py”, line 51, in
from mindnlp import transformers
File “/home/HwHiAiUser/mindnlp/mindnlp/transformers/init.py”, line 16, in
from . import models, pipelines
File “/home/HwHiAiUser/mindnlp/mindnlp/transformers/models/init.py”, line 19, in
from . import (
File “/home/HwHiAiUser/mindnlp/mindnlp/transformers/models/albert/init.py”, line 16, in
from . import tokenization_albert, tokenization_albert_fast, configuration_albert, modeling_albert
File “/home/HwHiAiUser/mindnlp/mindnlp/transformers/models/albert/modeling_albert.py”, line 28, in
from mindnlp.core import nn, ops
File “/home/HwHiAiUser/mindnlp/mindnlp/core/init.py”, line 16, in
from . import optim, ops, nn, distributions
File “/home/HwHiAiUser/mindnlp/mindnlp/core/optim/init.py”, line 2, in
from .optimizer import Optimizer
File “/home/HwHiAiUser/mindnlp/mindnlp/core/optim/optimizer.py”, line 28, in
from mindnlp.core.nn import Parameter
File “/home/HwHiAiUser/mindnlp/mindnlp/core/nn/init.py”, line 16, in
from . import utils, functional, init
File “/home/HwHiAiUser/mindnlp/mindnlp/core/nn/utils/init.py”, line 2, in
from . import parametrizations
File “/home/HwHiAiUser/mindnlp/mindnlp/core/nn/utils/parametrizations.py”, line 9, in
from …nn.modules import Module
File “/home/HwHiAiUser/mindnlp/mindnlp/core/nn/modules/init.py”, line 4, in
from .linear import Linear, Identity
File “/home/HwHiAiUser/mindnlp/mindnlp/core/nn/modules/linear.py”, line 11, in
from … import ops
File “/home/HwHiAiUser/mindnlp/mindnlp/core/ops/init.py”, line 7, in
from .creation import *
File “/home/HwHiAiUser/mindnlp/mindnlp/core/ops/creation.py”, line 4, in
from mindspore._c_expression import Tensor as CTensor # pylint: disable=no-name-in-module, import-error
ImportError: cannot import name ‘Tensor’ from ‘mindspore._c_expression’ (/home/HwHiAiUser/.local/lib/python3.9/site-packages/mindspore/_c_expression.cpython-39-aarch64-linux-gnu.so)
用户您好，还请补充一下您使用的环境版本信息，以及脚本信息，方便分析问题。
这个错误通常是mindnlp版本和mindspore版本不匹配造成的，ai pro上mindnlp 0.4.1分支+mindspore 2.5是可以的，不会有这个问题；
还有，janus模型，在ai pro上有个单独的分支，否则会出错，如果是910b或者310p的话，用0.4.1就可以跑janus：
之前根据这个帮忙在ai pro上测试janus跑过，mimdspore 2.4和2.5轻测可以运行；用的是这个镜像，里面自带的cann版本是8.0：
用户您好，MindSpore支撑人已经分析并给出了问题的原因，由于较长时间未看到您采纳回答，这里版主将进行采纳回答的结帖操作，如果还其他疑问请发新帖子提问，谢谢支持~
此话题已在最后回复的 60 分钟后被自动关闭。不再允许新回复。