我使用下面的代码并报告错误，这是一个非常简单的例子，不应该按预期报告错误。
import numpy as np
import torch as th
# Assuming tt is some data (example as list)
tt = [1, 2, 3, 4, 5] # Example data
# Check if tt is a NumPy array, and convert if necessary
if not isinstance(tt, np.ndarray):
tt = np.array(tt)
# Now, convert tt to a PyTorch tensor
tensor_tt = th.from_numpy(tt)
print(tensor_tt)
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
Cell In[26], line 12
9 tt = np.array(tt)
11 # Now, convert tt to a PyTorch tensor
---> 12 tensor_tt = th.from_numpy(tt)
13 print(tensor_tt)
TypeError: expected np.ndarray (got numpy.ndarray)
我正在使用以下 conda 环境：
conda list
# packages in environment at /opt/miniconda3/envs/ethos:
#
# Name Version Build Channel
appnope 0.1.4 pyhd8ed1ab_0 conda-forge
asttokens 2.4.1 pyhd8ed1ab_0 conda-forge
blas 1.0 openblas
bottleneck 1.3.7 py312ha86b861_0
brotli 1.0.9 h80987f9_8
brotli-bin 1.0.9 h80987f9_8
bzip2 1.0.8 h80987f9_6
ca-certificates 2024.9.24 hca03da5_0
click 8.1.7 pypi_0 pypi
colorlog 6.8.2 pypi_0 pypi
comm 0.2.2 pyhd8ed1ab_0 conda-forge
contourpy 1.2.0 py312h48ca7d4_0
cycler 0.11.0 pyhd3eb1b0_0
debugpy 1.6.7 py312h313beb8_0
decorator 5.1.1 pyhd8ed1ab_0 conda-forge
ethos 0.1.0 pypi_0 pypi
exceptiongroup 1.2.2 pyhd8ed1ab_0 conda-forge
executing 2.1.0 pyhd8ed1ab_0 conda-forge
expat 2.6.3 h313beb8_0
filelock 3.16.1 pypi_0 pypi
fonttools 4.51.0 py312h80987f9_0
freetype 2.12.1 h1192e45_0
fsspec 2024.9.0 pypi_0 pypi
h5py 3.12.1 pypi_0 pypi
importlib-metadata 8.5.0 pyha770c72_0 conda-forge
ipykernel 6.29.5 pyh57ce528_0 conda-forge
ipython 8.28.0 pyh707e725_0 conda-forge
jedi 0.19.1 pyhd8ed1ab_0 conda-forge
jinja2 3.1.4 pypi_0 pypi
joblib 1.4.2 pypi_0 pypi
jpeg 9e h80987f9_3
jupyter_client 8.6.3 pyhd8ed1ab_0 conda-forge
jupyter_core 5.7.2 pyh31011fe_1 conda-forge
kiwisolver 1.4.4 py312h313beb8_0
lcms2 2.12 hba8e193_0
lerc 3.0 hc377ac9_0
libbrotlicommon 1.0.9 h80987f9_8
libbrotlidec 1.0.9 h80987f9_8
libbrotlienc 1.0.9 h80987f9_8
libcxx 14.0.6 h848a8c0_0
libdeflate 1.17 h80987f9_1
libffi 3.4.4 hca03da5_1
libgfortran 5.0.0 11_3_0_hca03da5_28
libgfortran5 11.3.0 h009349e_28
libopenblas 0.3.21 h269037a_0
libpng 1.6.39 h80987f9_0
libsodium 1.0.18 h27ca646_1 conda-forge
libtiff 4.5.1 h313beb8_0
libwebp-base 1.3.2 h80987f9_0
llvm-openmp 14.0.6 hc6e5704_0
lz4-c 1.9.4 h313beb8_1
markupsafe 2.1.5 pypi_0 pypi
matplotlib-base 3.9.2 py312h2df2da3_0
matplotlib-inline 0.1.7 pyhd8ed1ab_0 conda-forge
mpmath 1.3.0 pypi_0 pypi
ncurses 6.4 h313beb8_0
nest-asyncio 1.6.0 pyhd8ed1ab_0 conda-forge
networkx 3.3 pypi_0 pypi
numexpr 2.8.7 py312h0f3ea24_0
numpy 2.1.2 pypi_0 pypi
numpy-base 1.26.4 py312he047099_0
openjpeg 2.5.2 h54b8e55_0
openssl 3.3.2 h8359307_0 conda-forge
packaging 24.1 pyhd8ed1ab_0 conda-forge
pandas 2.2.3 pypi_0 pypi
parso 0.8.4 pyhd8ed1ab_0 conda-forge
pexpect 4.9.0 pyhd8ed1ab_0 conda-forge
pickleshare 0.7.5 py_1003 conda-forge
pillow 10.4.0 py312h80987f9_0
pip 24.2 py312hca03da5_0
platformdirs 4.3.6 pyhd8ed1ab_0 conda-forge
prompt-toolkit 3.0.48 pyha770c72_0 conda-forge
psutil 5.9.0 py312h80987f9_0
ptyprocess 0.7.0 pyhd3deb0d_0 conda-forge
pure_eval 0.2.3 pyhd8ed1ab_0 conda-forge
pyarrow 17.0.0 pypi_0 pypi
pygments 2.18.0 pyhd8ed1ab_0 conda-forge
pyparsing 3.1.2 py312hca03da5_0
python 3.12.7 h99e199e_0
python-dateutil 2.9.0.post0 pypi_0 pypi
python-tzdata 2023.3 pyhd3eb1b0_0
pytz 2024.2 pypi_0 pypi
pyzmq 25.1.2 py312h313beb8_0
readline 8.2 h1a28f6b_0
seaborn 0.13.2 pypi_0 pypi
setuptools 75.1.0 py312hca03da5_0
six 1.16.0 pyh6c4a22f_0 conda-forge
sqlite 3.45.3 h80987f9_0
stack_data 0.6.2 pyhd8ed1ab_0 conda-forge
sympy 1.13.3 pypi_0 pypi
tk 8.6.14 h6ba3021_0
torch 2.4.1 pypi_0 pypi
tornado 6.4.1 py312h80987f9_0
tqdm 4.66.5 pypi_0 pypi
traitlets 5.14.3 pyhd8ed1ab_0 conda-forge
typing_extensions 4.12.2 pyha770c72_0 conda-forge
tzdata 2024.2 pypi_0 pypi
unicodedata2 15.1.0 py312h80987f9_0
wcwidth 0.2.13 pyhd8ed1ab_0 conda-forge
wheel 0.44.0 py312hca03da5_0
xz 5.4.6 h80987f9_1
zeromq 4.3.5 h313beb8_0
zipp 3.20.2 pyhd8ed1ab_0 conda-forge
zlib 1.2.13 h18a0788_1
zstd 1.5.5 hd90d995_2
我尝试在 torch 中将 np.array 数据转换为张量；报告的错误令人困惑；我不知道包版本是否存在冲突。
当我降级 numpy 时它仍然报告相同的错误：
import numpy as np
import torch
print(torch.__version__)
print(np.__version__)
x = np.array([[1, 2, 3, 4], [5, 6, 7, 8]], dtype=np.float32)
y = torch.from_numpy(x)
2.4.1
1.26.0
{
"name": "TypeError",
"message": "expected np.ndarray (got numpy.ndarray)",
"stack": "---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
Cell In[1], line 6
4 print(np.__version__)
5 x = np.array([[1, 2, 3, 4], [5, 6, 7, 8]], dtype=np.float32)
----> 6 y = torch.from_numpy(x)
TypeError: expected np.ndarray (got numpy.ndarray)"
}
最佳答案
2
numpy这是一个已知错误，您现在需要降级：
pip install "numpy<1.26.4"
参考：
3
-
这对我来说似乎不可行，当使用 1.26.0 版本的 numpy 时，它会报告同样的错误
– -
你的 Torch 版本是什么？
– -
我的 Torch 版本是 2.4.1
–
|
经过多次尝试，我终于找到了问题的原因。注意，安装中有两个 numpy 版本numpy (1.26.4)和。如果你卸载 numpy (1.26.4) 然后通过 重新安装，你的错误仍然会存在；你需要再次确保版本 2.1.2 也被删除了。numpy-base (2.1.2)conda install numpy==1.26.0pip uninstall numpy
|
|