YDOOK

YDOOK：onnxruntime session 定义 provider

YDOOK：onnxruntime 加载和运行模型 InferenceSession

YDOOK：onnxruntime 是什么？API Overview

YDOOK：onnx onnxruntime Python APIs 接口文档网址

YDOOK：ONNX onnxruntime APIs 接口文档网址

YDOOK：onnx onnxruntime 官网 origin website

如何解决 VSC 配置 Anaconda 后，anaconda 包引入不了的问题

YDOOK：Librosa 读取 .wav 音频文件并展示显示波形图© YDOOK Jinwei Lin, shiye.workhttp://www.ydook.comhttps://orcid.org/0000-0003-0558-6699文章目录YDOOK：Librosa 读取 .wav 音频文件并展示显示波形图© YDOOK Jinwei Lin, shiye.workhttp://www.ydook.comhttps://orcid.org/0000-0003-0558-66991. Co

YDOOK：通过 ffmpeg 转化格式的音频 前后的采样率一样吗© YDOOK Jinwei Lin, shiye.workhttp://www.ydook.comhttps://orcid.org/0000-0003-0558-6699文章目录YDOOK：通过 ffmpeg 转化格式的音频 前后的采样率一样吗© YDOOK Jinwei Lin, shiye.workhttp://www.ydook.comhttps://orcid.org/0000-0003-0558-6699一样的，数据数

YDOOK：MyPLayer：Jinwei Lin 最新开源 Python 音频视频基本播放器© YDOOK Jinwei Lin, shiye.workhttp://www.ydook.comhttps://orcid.org/0000-0003-0558-6699文章目录YDOOK：MyPLayer：Jinwei Lin 最新开源 Python 音频视频基本播放器© YDOOK Jinwei Lin, shiye.workhttp://www.ydook.comhttps://orcid.or

YDOOK：Python 通过系统直接播放音频音乐的各种方式 mp3 wav© YDOOK Jinwei Lin, shiye.workhttp://www.ydook.comhttps://orcid.org/0000-0003-0558-6699文章目录YDOOK：Python 通过系统直接播放音频音乐的各种方式 mp3 wav© YDOOK Jinwei Lin, shiye.workhttp://www.ydook.comhttps://orcid.org/0000-0003-0558-6

YDOOK：Pydub 使用教程 打开文件 各种格式© YDOOK Jinwei Lin, shiye.workhttp://www.ydook.comhttps://orcid.org/0000-0003-0558-6699文章目录YDOOK：Pydub 使用教程 打开文件 各种格式© YDOOK Jinwei Lin, shiye.workhttp://www.ydook.comhttps://orcid.org/0000-0003-0558-66991. 官网：[http://pydub.c

YDOOK：Pytorch教程：Numpy 手动设置反向梯度递归 神经网络© YDOOK Jinwei Lin, shiye.work文章目录YDOOK：Pytorch教程：Numpy 手动设置反向梯度递归 神经网络© YDOOK Jinwei Lin, shiye.workCode:Outcome:1. 循环 5000 次：设置学习率：learning_rate = 1e-52. 循环 10000 次：设置学习率：learning_rate = 1e-5Code:# import torch

YDOOK：Pytorch教程：反向传播 out.backward() 用法© YDOOK Jinwei Lin, shiye.work文章目录YDOOK：Pytorch教程：反向传播 out.backward() 用法© YDOOK Jinwei Lin, shiye.workCode:Outcome:Code:import numpyimport torchimport numpy as np# a = torch.tensor([1], requires_grad=True, dt

YDOOK：Pytorch教程：tensor 张量变量 requires_grad=True 设置 运算后因变量保留自变量该属性© YDOOK Jinwei Lin, shiye.work文章目录YDOOK：Pytorch教程：tensor 张量变量 requires_grad=True 设置 运算后因变量保留自变量该属性© YDOOK Jinwei Lin, shiye.workCode:Result:Code:import numpyimport torchimport numpy as

YDOOK：Pytorch教程：tensor mean() 函数求平均值 按行按列求平均值© YDOOK Jinwei Lin, shiye.work文章目录YDOOK：Pytorch教程：tensor mean() 函数求平均值 按行按列求平均值© YDOOK Jinwei Lin, shiye.work使用方法：mean(matrix, axis= a) 其中 matrix为一个矩阵， axis= a 为分类参数Code:Result:使用方法：mean(matrix, axis= a)

YDOOK：Pytorch教程：tensor 导数与自动求导 Autograd 设置 requires_grad=True© YDOOK Jinwei Lin, shiye.work文章目录YDOOK：Pytorch教程：tensor 导数与自动求导 Autograd 设置 requires_grad=True© YDOOK Jinwei Lin, shiye.workCode:Result:Code:import numpyimport torchimport numpy as np#

YDOOK：Pytorch教程：直接赋值 tensor 的特性© YDOOK Jinwei Lin, shiye.work文章目录YDOOK：Pytorch教程：直接赋值 tensor 的特性© YDOOK Jinwei Lin, shiye.workKey: 直接赋值 tensor ， 与直接赋值 Python 一般变量相同。result:Key: 直接赋值 tensor ， 与直接赋值 Python 一般变量相同。import numpyimport torchimport numpy

YDOOK：Pytorch教程：使用 dtype= 定义张量矩阵元素类型© YDOOK Jinwei Lin, shiye.work文章目录YDOOK：Pytorch教程：使用 dtype= 定义张量矩阵元素类型© YDOOK Jinwei Lin, shiye.workresult:import numpyimport torchimport numpy as npa = torch.ones([3, 3])print('a = ', a)print('type(a) = ', t

YDOOK：Pytorch教程：numpy 变量与 tensor变量相互转换 转化© YDOOK Jinwei Lin, shiye.workimport numpyimport torchimport numpy as npa = [[1, 2, 3], [4, 5, 6]]print('a = ', a)print('type(a = ', type(a))print()# 普通列表数据转换为 numpy数组n_a = numpy.array(a)print('n_a = '

YDOOK：Pytorch教程：tensor 张量内各个值同时相加一个数© YDOOK Jinwei Lin, shiye.workimport torchimport numpy as npt1 = torch.tensor([[1, 2, 3], [4, 5, 6]])print(t1)a = 6print('a = ', a)print()# 将张量 t1 的各个值都加上 at1_a = t1.add(a)print('t1_a = ', t1_a)outc

YDOOK：Pytorch教程：tensor 张量内各个元素之和 相加值输出© YDOOK Jinwei Lin, shiye.workimport torchimport numpy as npt1 = torch.tensor([[1, 2, 3], [4, 5, 6]])print(t1)# 将张量 t1 的各个值相加print(t1.sum())# 输出print(type(t1.sum()))# 使用 .item()# 传化为Python变量输出和print(t

YDOOK：Pytorch教程：tensor 张量矩阵的点乘与叉乘 数量积与向量积© YDOOK Jinwei Lin, shiye.workimport torchimport numpy as npt1 = torch.tensor([[1, 2, 3], [4, 5, 6]])t2 = torch.tensor([[7, 8], [9, 10], [11, 12]])# 打印 t1 t2print('t1 = ', t1)print('t2 = ', t2)print()#

YDOOK：Pytorch教程：tensor 张量相乘 矩阵相乘© YDOOK Jinwei Lin, shiye.workimport torchimport numpy as npt = torch.tensor([[1, 2, 3], [4, 5, 6]])print(t)# 输出转置张量print(t.T)# 矩阵相乘t_mul = t.matmul(t.T)print('t = ', t)print('t.T = ', t.T)print('\n2X3张量与3X

YDOOK：Pytorch教程：转置矩阵 转置张量 T© YDOOK Jinwei Lin, shiye.workimport torchimport numpy as npt = torch.tensor([[1, 2, 3], [4, 5, 6]])print(t)# 输出转置张量print(t.T)outcome:tensor([[1, 2, 3], [4, 5, 6]])tensor([[1, 4], [2, 5],

YDOOK：Pytorch教程：tensor 多个张量合并© YDOOK Jinwei Lin, shiye.workimport torchimport numpy as npa = torch.tensor([[1, 2, 3], [11, 22, 33]])b = torch.tensor([[4, 5, 6], [44, 55, 66]])c = torch.tensor([[7, 8, 9], [77, 88, 99]])print(a)print(type(a))pri

YDOOK：Pytorch教程：tensor 张量切片 局部获取操作© YDOOK Jinwei Lin, shiye.workimport torchimport numpy as nptensor = torch.rand((3, 3))print(type(tensor))print('tensor = \n', tensor)print()print('tensor 的第0行 = ', tensor[0])print('tensor 的第1行 = ', tensor[1])

YDOOK：Pytorch教程：判断是否可以使用 cuda 加速© YDOOK Jinwei Lin, shiye.workimport torchimport numpy as npprint(torch.cuda.is_available())if torch.cuda.is_available(): print('当前可以使用 cuda GPU加速')else: print('当前不可以使用 cuda GPU加速')outcome:True当前可以使用 c

YDOOK：Pytorch教程：tensor 的属性 调用设置© YDOOK Jinwei Lin, shiye.workimport torchimport numpy as nptensor_variable = torch.rand(3,5)print('输出变量形状：shape = ', tensor_variable.shape)print('输出变量数据类型：dtype = ', tensor_variable.dtype)print('输出变量存储设备：device =

YDOOK：Pytorch教程：通过指定形状shape, 快速生成 tensor© YDOOK Jinwei Lin, shiye.workimport torchimport numpy as npshape = (3, 5)ones_tensor = torch.ones(shape)zeros_tensor = torch.zeros(shape)rand_tensor = torch.rand(shape)print(torch.ones(shape))print()p

YDOOK：Pytorch教程：复制已有的 tensor结构，生成随机张量 tensor rand_like tensor© YDOOK Jinwei Lin, shiye.workimport torchimport numpy as npdata = [[1, 2],[3, 4], [5, 6]]n_data = np.array(data)print(n_data)print(type(n_data))t_data = torch.from_numpy(n_data)t_s

YDOOK：Pytorch教程：复制已有的 tensor结构，生成全一张量 tensor© YDOOK Jinwei Lin, shiye.workimport torchimport numpy as npdata = [[1, 2],[3, 4], [5, 6]]n_data = np.array(data)print(n_data)print(type(n_data))t_data = torch.from_numpy(n_data)t_shape_tensor = tor

YDOOK：Pytorch教程：从 numpy array 数组中构建 tensor© YDOOK Jinwei Lin, shiye.workimport torchimport numpy as npdata = [[1, 2],[3, 4]]n_data = np.array(data)print(n_data)print(type(n_data))t_np_tensor = torch.from_numpy(n_data)print()print(t_np_tensor

YDOOK：Pytorch教程：新建 tensor变量© YDOOK Jinwei Lin, shiye.workimport torchimport numpy as npdata = [[1, 2],[3, 4]]n_data = np.array(data)t_data = torch.tensor(data)print(n_data)print(type(n_data))print()print(t_data)print(type(t_data))输出[[1 2

YDOOK：Pytorch教程：安装 Pythorch© YDOOK Jinwei Lin, shiye.workhttps://pytorch.org/get-started/locally/conda install pytorch torchvision torchaudio cudatoolkit=11.3 -c pytorchpip3 install torch1.10.0+cu113 torchvision0.11.1+cu113 torchaudio===0.10.0+cu11

YDOOK：Pytorch 有哪些 Python API© YDOOK Jinwei Lin, shiye.work**torchtorch.nntorch.nn.functionaltorch.TensorTensor AttributesTensor Viewstorch.autogradtorch.cudatorch.cuda.amptorch.backendstorch.distributedtorch.distributed.algorithms.jointorch

YDOOK：Pytorch 官方文档出处© YDOOK Jinwe Lin, shiye.workhttps://pytorch.org/docs/stable/index.html

YDOOK AI : cuda ：最新版 cuda11 安装详细教程 亲测完美1. 首先当然是是要下载 cuda11 :下载教程：https://ydook.blog.csdn.net/article/details/1124711522. 双击下载后的安装包：3. 开始安装：选择安装路径：4. 安装中：5. 同意并继续：6. 选择需要的安装模式：7. 下一步：8. 下一步：9. 安装中：10. 安装完成！...

YDOOK AI : Pytorch ：使用 tensor.ones() tensor.ones_like() 生成定义 全一数组 全1数组1_1 tensor.ones() 函数：torch._C._VariableFunctions @overload def ones(*size: int, out: Optional[Tensor] = None, dtype: Optional[dtype] = None, layout: layout =

YDOOK AI : Pytorch ：使用 tensor.zeros() tensor.zeros_like() 生成定义 全零数组 全0数组1. tensor.zeros() 函数：torch._C._VariableFunctions @overload def zeros(*size: int, out: Optional[Tensor] = None, dtype: Optional[dtype] = None, layout: l