BallonsTranslator
BallonsTranslator copied to clipboard
dmMaze
Bug Report:5070无法运行
Version Info
D:\Tools\BallonsTranslator_dev_src_with_gitpython\launch.py:8: DeprecationWarning: pkg_resources is deprecated as an API. See https://setuptools.pypa.io/en/latest/pkg_resources.html import pkg_resources Python version: 3.10.11 (tags/v3.10.11:7d4cc5a, Apr 5 2023, 00:38:17) [MSC v.1929 64 bit (AMD64)] Python executable: D:\Tools\BallonsTranslator_dev_src_with_gitpython\ballontrans_pylibs_win\python.exe Version: 1.4.0 Branch: dev Commit hash: bdb323c61e97eea69efcba6ec0cc5acc1f48bf36
Description of the Problem
NVIDIA GeForce RTX 5070 with CUDA capability sm_120 is not compatible with the current PyTorch installation. The current PyTorch install supports CUDA capabilities sm_37 sm_50 sm_60 sm_61 sm_70 sm_75 sm_80 sm_86 sm_90 compute_37. If you want to use the NVIDIA GeForce RTX 5070 GPU with PyTorch, please check the instructions at https://pytorch.org/get-started/locally/
Text Detector
None
OCR
None
Inpaint
None
Translator
None
Screenshot
No response
Logs
Additional Information
No response
DUP #825
我已安装cuda12.8，且已安装最新的pytorch，但是显示如下信息
Device name: NVIDIA GeForce RTX 5070 Cuda is available: True Cuda version: 11.8
DUP #825
我已安装cuda12.8，且已安装最新的pytorch，但是显示如下信息
Device name: NVIDIA GeForce RTX 5070 Cuda is available: True Cuda version: 11.8
Did you install the updated torch through python from the package with which you downloaded (Google Drive) or did you install a new torch separately? The log shows that you did not install the update correctly, or you installed it in the wrong env
DUP #825
我已安装cuda12.8，且已安装最新的pytorch，但是显示如下信息
设备名称：NVIDIA GeForce RTX 5070 Cuda 可用：True Cuda 版本：11.8
您是通过 python 从下载的软件包 （Google Drive） 安装更新的 Torch，还是单独安装新的 Torch？日志显示您没有正确安装更新，或者您将其安装在错误的环境中 我是单独安装的
>>> import torch
>>> print(f"PyTorch版本: {torch.__version__}")
PyTorch版本: 2.7.0+cu128
>>> print(f"CUDA版本: {torch.version.cuda}")
CUDA版本: 12.8
>>> print(f"设备名: {torch.cuda.get_device_name(0)}")
设备名: NVIDIA GeForce RTX 5070
>>> print(f"支持架构: {torch.cuda.get_arch_list()}")
支持架构: ['sm_50', 'sm_60', 'sm_61', 'sm_70', 'sm_75', 'sm_80', 'sm_86', 'sm_90', 'sm_100', 'sm_120']
Run this
import sys
import os
python_executable_path = sys.executable
print(f"Path to the Python executable file: {python_executable_path}")
Run this
import sys import os
python_executable_path = sys.executable
print(f"Path to the Python executable file: {python_executable_path}")
Path to the Python executable file: C:\Users\13919\AppData\Local\Programs\Python\Python311\python.exe
当我重新下载最新版的整合版本并点击运行 scripts/local_gitpull.bat 获取更新，从头开始安装时，它自动下载的pytorch版本貌似是基于cuda11.8的
Path to the Python executable file: C:\Users\13919\AppData\Local\Programs\Python\Python311\python.exe
How did you install BT? Send me a screenshot of the BT folder
Python 可执行文件的路径：C：\Users\13919\AppData\Local\Programs\Python\Python311\python.exe
您是如何安装 BT 的？将 BT 文件夹的屏幕截图发送给我
Открываешь консоль в папке с BT
пишешь
ballontrans_pylibs_win\python.exe -m pip install torch torchvision torchaudio --index-url download.pytorch.org/whl/cu128
Оно установит тебе в программу новый драйвер для 5070
Открываешь консоль в папке с BT
пишешь
ballontrans_pylibs_win\python.exe -m pip install torch torchvision torchaudio --index-url download.pytorch.org/whl/cu128 Оно установит тебе в программу новый драйвер для 5070
done，thank you товарищ