diff --git a/tutorials/source_zh_cn/beginner/dataset.ipynb b/tutorials/source_zh_cn/beginner/dataset.ipynb index e8a3cce3456a23fdda466e126a3e6e2bee5dfa07..530a01a9793709596f265aaae57014deadd082fe 100644 --- a/tutorials/source_zh_cn/beginner/dataset.ipynb +++ b/tutorials/source_zh_cn/beginner/dataset.ipynb @@ -133,9 +133,7 @@ "\n", "+ `batch_size`：每组包含的数据个数，`batch_size=2`设置每组包含2个数据， `batch_size`值默认大小为32。\n", "\n", - "+ `repeat_num`：重复数据集的个数，`repeat_num=1`即一份数据集，`repeat_num`值默认为1 。\n", - "\n", - "下面的样例实现将数据集随机打乱顺序并将样本两两组成一个批次的功能。" + "+ `repeat_num`：重复数据集的个数，`repeat_num=1`即一份数据集，`repeat_num`值默认为1 。\n" ] }, { diff --git a/tutorials/source_zh_cn/beginner/infer.ipynb b/tutorials/source_zh_cn/beginner/infer.ipynb index 829bdbcb600b5eb6ea7862901906d232dcb6b09e..4e0c174cbc591739b8049be5fc93647f6d38fa42 100644 --- a/tutorials/source_zh_cn/beginner/infer.ipynb +++ b/tutorials/source_zh_cn/beginner/infer.ipynb @@ -548,11 +548,11 @@ "\n", "打开APP后，在首页点击`分类`模块后，即可点击中间按钮进行拍照获取图片，或者点击上侧栏的图像按钮选择进行图片相册用于图像分类功能。\n", "\n", - "![main](https://gitee.com/mindspore/docs/raw/tutorials-develop/tutorials/source_zh_cn/beginner/images/app1.png)\n", + "
![main](https://gitee.com/mindspore/docs/raw/master/tutorials/source_zh_cn/beginner/images/app1.png)
\n", "\n", - "> 在默认情况下，MindSpore Vision`分类`模块内置了一个通用的AI网络模型对图像进行识别分类。\n", + "在默认情况下，MindSpore Vision`分类`模块内置了一个通用的AI网络模型对图像进行识别分类。\n", "\n", - "![result](https://gitee.com/mindspore/docs/raw/tutorials-develop/tutorials/source_zh_cn/beginner/images/app2.png)" + "
![result](https://gitee.com/mindspore/docs/raw/master/tutorials/source_zh_cn/beginner/images/app2.png)
" ] }, { @@ -572,7 +572,7 @@ "{\n", " \"title\": '狗和牛角包',\n", " \"file\": 'mobilenet_v2_1.0_224.ms',\n", - " \"label\": ['牛角包', '狗'],\n", + " \"label\": ['牛角包', '狗']\n", "}\n", "```\n", "\n", @@ -588,11 +588,11 @@ "\n", "为实现手机端狗与牛角包的识别功能，需将标签文件`custom.json`文件和模型文件`mobilenet_v2_1.0_224.ms`一起放置到手机上指定目录下。这里以`Android/data/Download/` 文件夹为例，首先把标签文件和模型文件同时放在上述手机地址，如图所示，点击自定义按钮，然后会弹出系统文件功能，点击左上角的打开文件，然后找到Json标签文件和模型文件存放的目录地址，并选择对应的Json文件。\n", "\n", - "![step](https://gitee.com/mindspore/docs/raw/tutorials-develop/tutorials/source_zh_cn/beginner/images/app3.png)\n", + "
![step](https://gitee.com/mindspore/docs/raw/master/tutorials/source_zh_cn/beginner/images/app3.png)
\n", "\n", "标签与模型文件部署到手机后，即可点击中间按钮进行拍照获取图片，或者点击上侧栏的图像按钮选择图片相册用于图像，就可以进行狗与牛角包的分类识别。\n", "\n", - "![result1](https://gitee.com/mindspore/docs/raw/tutorials-develop/tutorials/source_zh_cn/beginner/images/app4.png)\n", + "
![result1](https://gitee.com/mindspore/docs/raw/master/tutorials/source_zh_cn/beginner/images/app4.png)
\n", "\n", "> 本章仅包含手机侧简单的部署过程，想要了解推理更多内容请参考[MindSpore Lite](https://www.mindspore.cn/lite/docs/zh-CN/master/index.html)。" ] diff --git a/tutorials/source_zh_cn/beginner/quick_start.ipynb b/tutorials/source_zh_cn/beginner/quick_start.ipynb index 0d5c084df9439821881060c0ce1688809ebdcb59..81baa0d828867c5b961872178ada66f216b3ef87 100644 --- a/tutorials/source_zh_cn/beginner/quick_start.ipynb +++ b/tutorials/source_zh_cn/beginner/quick_start.ipynb @@ -86,7 +86,7 @@ }, { "cell_type": "code", - "execution_count": 14, + "execution_count": 3, "metadata": {}, "outputs": [], "source": [ @@ -109,7 +109,7 @@ }, { "cell_type": "code", - "execution_count": 15, + "execution_count": 4, "metadata": {}, "outputs": [], "source": [ @@ -156,7 +156,7 @@ }, { "cell_type": "code", - "execution_count": 16, + "execution_count": 5, "metadata": {}, "outputs": [], "source": [ @@ -216,7 +216,7 @@ }, { "cell_type": "code", - "execution_count": 18, + "execution_count": 7, "metadata": {}, "outputs": [ { @@ -244,9 +244,20 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 8, "metadata": {}, - "outputs": [], + "outputs": [ + { + "data": { + "text/plain": [ + "[]" + ] + }, + "execution_count": 8, + "metadata": {}, + "output_type": "execute_result" + } + ], "source": [ "from mindspore import load_checkpoint, load_param_into_net\n", "\n", @@ -271,14 +282,26 @@ }, { "cell_type": "code", - "execution_count": 20, + "execution_count": 10, "metadata": {}, "outputs": [ + { + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "needs_background": "light" + }, + "output_type": "display_data" + }, { "name": "stdout", "output_type": "stream", "text": [ - "Predicted: \"1\", Actual: \"1\"\n" + "Predicted: \"[4 8 9 1 6 6]\", Actual: \"[4 8 9 1 6 6]\"\n" ] } ], diff --git a/tutorials/source_zh_cn/beginner/tensor.ipynb b/tutorials/source_zh_cn/beginner/tensor.ipynb index cac2c1c1aab5a3d85d9313a7f7ee01a78c2c5729..005803079a7018340c4329c8ebed1b36d81912c5 100644 --- a/tutorials/source_zh_cn/beginner/tensor.ipynb +++ b/tutorials/source_zh_cn/beginner/tensor.ipynb @@ -464,7 +464,11 @@ "source": [ "## 稀疏张量\n", "\n", - "稀疏张量是一种特殊张量，其中绝大部分元素的值为零。在某些应用场景中（比如推荐系统、分子动力学、图神经网络等），数据的特征是稀疏的，若使用普通张量表征这些数据会引入大量不必要的计算、存储和通讯开销。在这种时候就可以使用稀疏张量来表征这些数据。常用的稀疏格式有`COO`、`CSR`、`CSC`、`DIA`等，不同的稀疏格式有其最适合的应用场景。其中，MindSpore现在已经支持最常用的`CSR`和`COO`两种稀疏数据格式。\n", + "稀疏张量是一种特殊张量，其中绝大部分元素的值为零。\n", + "\n", + "在某些应用场景中（比如推荐系统、分子动力学、图神经网络等），数据的特征是稀疏的，若使用普通张量表征这些数据会引入大量不必要的计算、存储和通讯开销。在这种时候就可以使用稀疏张量来表征这些数据。\n", + "\n", + "MindSpore现在已经支持最常用的`CSR`和`COO`两种稀疏数据格式。\n", "\n", "常用稀疏张量的表达形式是``。其中，`indices`表示非零下标元素， `values`表示非零元素的值，shape表示的是被压缩的稀疏张量的形状。 在这个结构下，我们定义了`CSRTensor`、`COOTensor`和`RowTensor`三种稀疏张量结构。\n", "\n", @@ -487,7 +491,7 @@ "\n", "- `shape`: 表示的是被压缩的稀疏张量的形状，数据类型为`Tuple`,目前仅支持2维`CSRTensor`。\n", "\n", - "`CSRTensor`的详细文档，请参考[mindspore.CSRTensor](https://www.mindspore.cn/docs/api/zh-CN/master/api_python/mindspore/mindspore.CSRTensor.html)。\n", + "> `CSRTensor`的详细文档，请参考[mindspore.CSRTensor](https://www.mindspore.cn/docs/api/zh-CN/master/api_python/mindspore/mindspore.CSRTensor.html)。\n", "\n", "下面给出一些CSRTensor的使用示例：" ] @@ -506,7 +510,6 @@ } ], "source": [ - "# CSRTensor的构建\n", "import mindspore as ms\n", "from mindspore import Tensor, CSRTensor\n", "\n", @@ -514,6 +517,8 @@ "indices = Tensor([0, 1])\n", "values = Tensor([1, 2], dtype=ms.float32)\n", "shape = (2, 4)\n", + "\n", + "# CSRTensor的构建\n", "csr_tensor = CSRTensor(indptr, indices, values, shape)\n", "\n", "print(csr_tensor.astype(ms.float64).dtype)" @@ -533,7 +538,7 @@ "\n", "- `shape`: 表示的是被压缩的稀疏张量的形状，目前仅支持2维`COOTensor`.\n", "\n", - "`COOTensor`的详细文档，请参考[mindspore.COOTensor](https://www.mindspore.cn/docs/api/zh-CN/master/api_python/mindspore/mindspore.COOTensor.html)。\n", + "> `COOTensor`的详细文档，请参考[mindspore.COOTensor](https://www.mindspore.cn/docs/api/zh-CN/master/api_python/mindspore/mindspore.COOTensor.html)。\n", "\n", "下面给出一些COOTensor的使用示例：" ] @@ -556,7 +561,6 @@ } ], "source": [ - "# COOTensor的构建\n", "import mindspore as ms\n", "import mindspore.nn as nn\n", "from mindspore import Tensor, COOTensor\n", @@ -564,22 +568,31 @@ "indices = Tensor([[0, 1], [1, 2]], dtype=ms.int32)\n", "values = Tensor([1, 2], dtype=ms.float32)\n", "shape = (3, 4)\n", + "\n", + "# COOTensor的构建\n", "coo_tensor = COOTensor(indices, values, shape)\n", "\n", "print(coo_tensor.values)\n", "print(coo_tensor.indices)\n", "print(coo_tensor.shape)\n", - "# COOTensor转换数据类型\n", - "print(coo_tensor.astype(ms.float64).dtype)" + "print(coo_tensor.astype(ms.float64).dtype) # COOTensor转换数据类型" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ - "上述代码会生成如下`COOTensor`:\n", + "上述代码会生成如下式所示的`COOTensor`:\n", "\n", - "![COOTensor](https://gitee.com/mindspore/docs/raw/tutorials-develop/tutorials/source_en/beginner/images/tensor_1.PNG)" + "$$\n", + " \\left[\n", + " \\begin{matrix}\n", + " 0 & 1 & 0 & 0 \\\\\n", + " 0 & 0 & 2 & 0 \\\\\n", + " 0 & 0 & 0 & 0\n", + " \\end{matrix}\n", + " \\right] \\tag{1}\n", + "$$" ] }, { @@ -596,7 +609,7 @@ "\n", "- `dense_shape`: 表示的是被压缩的稀疏张量的形状。\n", "\n", - "`RowTensor`只能在`Cell`的构造方法中使用。详细内容，请参考[mindspore.RowTensor](https://www.mindspore.cn/docs/api/zh-CN/master/api_python/mindspore/mindspore.RowTensor.html)。代码样例如下：" + "> `RowTensor`只能在`Cell`的构造方法中使用。详细内容，请参考[mindspore.RowTensor](https://www.mindspore.cn/docs/api/zh-CN/master/api_python/mindspore/mindspore.RowTensor.html)。代码样例如下：" ] }, {