detectron2 demo cuda10.0 py3.6

Okay. No machine is immortal. Here’s my 180394930208th installation … ==||

Installation

conda create -n detectron python=3.6 pip
conda activate detectron
conda install pytorch torchvision cudatoolkit=10.0 -c pytorch
pip install cython pyyaml==5.1
pip install -U 'git+https://github.com/cocodataset/cocoapi.git#subdirectory=PythonAPI'
pip install detectron2 -f https://dl.fbaipublicfiles.com/detectron2/wheels/cu100/index.html
# `cudaxxx` depends on your cuda version.
conda install -c menpo opencv

`vi mydemo.py`

# usage: python mydemo.py
import detectron2
from detectron2.utils.logger import setup_logger
setup_logger()
# import some common libraries
import numpy as np
import cv2
import random
# import some common detectron2 utilities
from detectron2 import model_zoo
from detectron2.engine import DefaultPredictor
from detectron2.config import get_cfg
from detectron2.utils.visualizer import Visualizer
from detectron2.data import MetadataCatalog
im = cv2.imread("./input.jpg")
cfg = get_cfg()
# add project-specific config (e.g., TensorMask) here if you're not running a model in detectron2's core library
cfg.merge_from_file(model_zoo.get_config_file("COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml"))
cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.5  # set threshold for this model
# Find a model from detectron2's model zoo. You can use the https://dl.fbaipublicfiles... url as well
cfg.MODEL.WEIGHTS = model_zoo.get_checkpoint_url("COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml")
predictor = DefaultPredictor(cfg)
outputs = predictor(im)
v = Visualizer(im[:, :, ::-1], MetadataCatalog.get(cfg.DATASETS.TRAIN[0]), scale=1.2)
v = v.draw_instance_predictions(outputs["instances"].to("cpu"))
v.save("./output.jpg")

Finally,

wget http://images.cocodataset.org/val2017/000000439715.jpg -O input.jpg
python mydemo.py

output.jpg for example:

References

https://colab.research.google.com/drive/16jcaJoc6bCFAQ96jDe2HwtXj7BMD_-m5
https://github.com/facebookresearch/detectron2/blob/master/INSTALL.md
https://github.com/facebookresearch/detectron2

detectron2 demo cuda10.0 py3.6相关推荐

Win10+CUDA10.0.130＋cudnn7.4.1+tensorflow1.13.1+anaconda3 5.2.0+GTX1060
Win10+CUDA10.0.130+cudnn7.4.1+tensorflow1.13.1+anaconda3 5.2.0+GTX1060 安装环境: 系统:win10 1803 显卡:GTX106 ...
ubuntu18.04安装CUDA10.0
1.更新驱动对于CUDA10.0,要求驱动版本大于410.48,因此首先要更新驱动 ubuntu-drivers devices(查看所有驱动) sudo ubuntu-drivers autoin ...
pytorch 安装 pip+windows10+python3.6+CUDA10.0
pytorch 安装 pip+windows10+python3.6+CUDA10.0 进入官网点击: install previous versions of PyTorch 进入安装网址,选择合 ...
YOLO3实现目标检测（VS2015+OPENCV3.4.2+YOLO3+CUDA10.0+CUDNN7.5）
本人YOLOv3刚刚入坑,走了不少弯路,自己摸索了一下,首先给大家配置一下yolo3,后续会有具体的算法分析讲解.安装VS2015+opencv3.4.2,这个就不用多说了吧,网上很多教程可以参考,主 ...
Ubuntu16.04安装NVIDIA显卡（RTX20系列）驱动+CUDA10.0+cudnn+Pytorch1.1.0
Ubuntu16.04,GPU: Geforce RTX 2070-Super(本博客是显卡驱动和cuda分开安装的教程). 1. Ubuntu16.04安装NVIDIA显卡驱动 1. 下载显卡驱动. ...
实验室服务器18.04LTS+RTX2080Ti+CUDA10.0+CUDAnn7.4+tf1.13装机之路
文章目录一.存在且并未解决的问题 1.安装ubuntu16.04LTS系统失败 2.ubuntu18.04LTS系统,DrCom上网掉线二.安装ubuntu18.04LTS系统三.安装NVIDI ...
linux卸载cuda10.0,Ubuntu下安装CUDA10.0以及问题
tensorflow版本与cuda和cudnn的对应关系: 安装一定要查看CUDA要求的linux下的Driver Version,链接网址如下: 在附加驱动中有本机所用的显卡驱动. 提示Incomp ...
linux终端安装cuda,Ubuntu下安装CUDA10.0以及问题
tensorflow版本与cuda和cudnn的对应关系:https://tensorflow.google.cn/install/source 安装一定要查看CUDA要求的linux下的Driver ...
Windows10配置CUDA10.0+cudnn7.5.1
1.安装CUDA10.0 (1)下载:https://developer.nvidia.com/cuda-10.0-download-archive?target_os=Windows&tar ...

detectron2 demo cuda10.0 py3.6

Installation

`vi mydemo.py`

Finally,

References

detectron2 demo cuda10.0 py3.6相关推荐

最新文章

热门文章