fail to import detectron2.data - pytorch

I installed detectron2 via anaconda: conda install -c conda-forge detectron2
The installation succeeded but i fail to run: detectron2.data import build_detection_train_loader
the error message is:
i wonder know what i can do to successfully run my code.
details of my conda environmnt:
I tried to re-install my pytorch, to make sure that its version is corresponding to torchvision

Related

No module found torch

I am using jetson NX xavier kit having cuda 10.2.89, open Cv 4.1.1 and tensorRT 7.1.3 . Trying to install pytorch. Tried installing with this line
conda install pytorch torchvision cpuonly -c pytorch
but when i write this line
import torch
It throws an error by saying that module not installed.
How I can verify if pytorch has been installed correctly.
Try this one
conda install -c pytorch pytorch
After executing this command, you need to enter yes(if prompted in the command line) for installing all the related packages. If there is no conflict while installing the libraries, the PyTorch library will be installed.
To check if it is properly installed or not, type the command python in your command line and type import torch to check if it is properly installed or not.

Pytorch Import Error: <pathname>: object file has no loadable segments

I've been trying to install the Pytorch module for my Ubuntu 16.04 LTS through conda. I used conda install pytorch torchvision cpuonly -c pytorch to install it (non CUDA version). However when I type import torch on the Python shell, this is what I see -
ImportError: /home/student/anaconda2/lib/python2.7/site-packages/torch/_C.so: object file has no loadable segments
I have verified that Pytorch was installed using conda list
I had the same issue on Ubuntu 18.04 for conda env with python 3.8. The problem I think is for the incomplete torch installation. So I did pip install from wheel instead of conda install. You may follow as below (assuming you have cuda11 installed):
create conda env
conda create --name=myenv python=3.8
conda activate myenv
Install torch from wheel
pip install torch==1.7.0+cu110 torchvision==0.8.1+cu110 torchaudio===0.7.0 -f https://download.pytorch.org/whl/torch_stable.html
Please note I had to install torchvision==0.8.1+cu110 as reported here

trouble importing Pytorch in Jupyter notebook

Iam new to deep learning and Iam trying to import Pytorch on Jupyter Notebook.
I installed Pytorch with the following lines of code in Anaconda Prompt.
conda create -n pytorch_p37 python=3.7
conda activate pytorch_p37
conda install pytorch torchvision -c pytorch
conda install jupyter
conda list
it all executed well.
but on importing pytorch it shows errors.
import torch
this error below:-
OSError: [WinError 126] The specified module could not be found
error showing image
The problem lied where I was installing a CUDA version.
I tried installing the CPU version and it worked fine (CUDA None).
conda install pytorch torchvision cpuonly -c pytorch
!pip install torch
It worked for me in a Anaconda's Jupyter notebook.
I lost one hour and found that launching the conda environment as stacked version does not lead to error:
in your example:
conda activate pytorch_p37
and from jupyter
import torch # error
from terminal:
conda activate --stack pytorch_p37
and from jupyter:
import torch # success
I could not figure out why :/
https://conda.io/projects/conda/en/latest/user-guide/tasks/manage-environments.html#activating-an-environment

ImportError: cannot import name 'XGBClassifier' from 'xgboost' (unknown location)

xgboost imported successfully, but I'm not able to import XGBClassifier.
Check whether xgboost is properly installed or not.
To install xgboost in anaconda distribution, you can run the following command in anaconda command-line console.
conda install -c conda-forge xgboost=0.6a2or 'conda install -c anaconda py-xgboost'
It will work fine after the installation.

Pytorch installation problems with Anaconda

After having upgraded my environment's python to Python 3.61, I attempted to install pytorch using this command:
conda install -c peterjc123 pytorch
However I got this error:
Fetching package metadata .............
Solving package specifications: .
UnsatisfiableError: The following specifications were found to be in conflict:
-pytorch
-pyqt
I also used the commands
conda install -c peterjc123 pytorch cuda90
conda install -c peterjc123 pytorch cuda80
But the result is still the same. Anyone got a clue how to solve this?
The problem was solved after downgrading from Python 3.6.2 to Python 3.5.1 after running:
conda install -c anaconda python=3.5.1
After running this command, run:
conda install -c peterjc123 pytorch
Pytorch should install as per normal. A similar issue occurs for openCV as well
An alternative way to install pytorch using anaconda is
conda create -n py_env python=3.5
source activate py_env
conda install pytorch-cpu torchvision -c pytorch
Go to python shell and import using the command
import torch
Hope this helps.

Resources