聚合国内IT技术精华文章,分享IT技术精华,帮助IT从业人士成长

  • 505500 views阅读

    Fix an Out Of Memory case

    Here is my code and it caused OOM (“Out Of Memory”) when running: for img_batch in data_load: input = torch.from_numpy(np.asarray(img_batch)).cuda() result = self._net(input.permu...

    分类:技术文章 时间:2022-10-14 11:18 我要评论(0个)

  • 782430 views阅读

    A strange problem in RegNetY-32G

    I have been using RegNetY in DongNiao for almost two years. Previously I was just using small models such as RegNetY-8G. But after having a computer with RTX-3080-TI, I started to use the biggest ...

    分类:技术文章 时间:2022-09-02 09:55 我要评论(0个)

  • 979726 views阅读

    Strange error from Nvidia’s apex library

    apex is a mixed training library from Nvidia. I have been using it since I got an RTX3080TI GPU. A few days ago, I started to use RegNetY-32GF (I just used RegNetY models smaller than16GF previous...

    分类:技术文章 时间:2022-08-26 11:40 我要评论(0个)

  • 832138 views阅读

    Accelerate inference speed of DNN on Intel CPU

    To save the cost on the inference server, I did some experiments on how to accelerate the speed of prediction for our model. import torch.nn as nn import pycls.core.builders as model_builder f...

    分类:技术文章 时间:2022-07-21 12:35 我要评论(0个)

  • 709123 views阅读

    Average weights of two Pytorch models

    After reading this paper, I begin to do an experiment about it. Referencing this snippet, I wrote my code: net1 = model_builder.build_model() net2 = model_builder.build_model() output = mode...

    分类:技术文章 时间:2022-07-15 08:43 我要评论(0个)

  • 591550 views阅读

    Insert multiple lines in a specific position of a file

    I used awk for quite a long time, but not his brother sed. A couple of days ago I want to insert two lines for a CMake file in a specific position and find a perfect answer: here. Now I could a...

    分类:技术文章 时间:2022-07-01 12:35 我要评论(0个)

  • 592738 views阅读

    Model saving error when using Apex

    Apex is a tool to enable mixed-precision training that comes from Nvidia. import apex.amp as amp net, optimizer = amp.initialize(net, optimizer, opt_level="O2") # forward outputs =...

    分类:技术文章 时间:2022-06-03 09:56 我要评论(0个)

  • 1329429 views阅读

    机器学习环境配置:WLS2+Ubuntu+CUDA+cuDNN

    p>最近买了一台新的笔记本,拿到电脑后的首先要做的是配置机器学习环境。中间犯了一些错误,于是整理出来供可能需要的同学参考。 安装Windows子系统 WLS2 从 WSL 1 更新到 WSL 2的主要原因包括: 提高文件系统性能, 支持完全的系统调用兼容性。 WSL 2 使用最新、最强大的虚拟化技术在轻量级实用工具虚拟机 (VM) 中运行 Linux 内核。 但是,WSL 2 不是传...

    分类:技术文章 时间:2022-04-04 20:46 我要评论(0个)

  • 1098831 views阅读

    Debug CUDA error for PyTorch

    After I changed my dataset for my code, the training failed: /tmp/pip-req-build-_tx3iysr/aten/src/ATen/native/cuda/ScatterGatherKernel.cu:310: operator(): block: [0,0,0], thread: [59,0,0] Asser...

    分类:技术文章 时间:2021-04-23 10:40 我要评论(0个)

  • 394536 views阅读

    A struggle to keep the accuracy

    In this August, we have got 0.83 evaluation accuracy for DIB-10K dataset. But since last month, we have updated the dataset and the accuracy could only get to 0.82. The first doubtful point is ...

    分类:技术文章 时间:2020-12-18 08:43 我要评论(0个)

  • 507232 views阅读

    TabNet: a new neural-network architecture for tabular data

    The neural network seems mostly to be used on Computer Vision and Natural Language Processing scenarios, while tree-models like GBDT are mainly used for tabular data. But why? Although this ...

    分类:技术文章 时间:2020-10-08 20:14 我要评论(0个)

  • 292011 views阅读

    The nn.Sigmoid() of PyTorch on Android device

    I have trained an EfficientNet model to classify more than ten thousand different categories of birds, by using PyTorch. To run this model on the mobile device, I built a program by learning the P...

    分类:技术文章 时间:2020-05-19 00:22 我要评论(0个)