ones (m, 1) , x ] ; % Add a column of ones to x 2 From this point on, you will need to remember that the age values from your training data are actually in the second column of x. This will be important : , 2 ) , x∗ theta , ’− ’ ) % remember that x i s now a matrix % with 2 columnsand the second % column contains the time info legend ( ’ Training data ’ , ’ Linear r e g r e s s i o n ’ ) Note that

the petting zoo. If we revisit the plot in Figure 4-1 with the newly assigned labels in the third column of Table 4-2, we can see a pattern. It is possible to linearly separate3 the data points belonging mechanism. On the left is the list of tokens. The tokens are hashed using the hash function in the center column. The output of the hash function % vocab_size is an index in [0, vocab_size - 1] and is used to refer visual representation of the attention the word corporation pays to each sequence element. The left column is a tokenized representation of our example text. The thickness of the connecting lines between

student’s scores on two exams. In your training data, the first column of your x array represents all Test 1 scores, and the second column represents all Test 2 scores, and the y vector uses “1” to label

多维数组可以进行连接，分段等多种操作。我们先来看 vstack(),hstack(),column_stack()函数。 > a = np.arange(3) > b = np.arange(10, 13) > v = np.vstack((a, b)) # 按第1轴连接数组 > h = np.hstack((a, b)) # 按第0轴连接数组 > c = np.column_stack((a, b)) # 按列连接多个一维数组

Figure 3-11: The image on the left is a cut-mix of turtle (3%) and tortoise (97%) images in the center column and the top-right image is their average mix. Bottom-right is a mixup of turtle 8 When the samples For example, the top-right image is an average mix of turtle and tortoise images in the center column. The average mixing is a label mixing technique that averages the sample images to produce the mixed

数据分析库：Pandas Pandas 是一个 BSD 开源协议许可的，面向 Python 用户的高性能和易于上手的数 据结构化和数据分析工具。 数据框（Data Frame）是一个二维带标记的数据结构，每列（column）数据类型 可以不同。我们可以将其当作电子表格或数据库表。 数据读入 pandas.read_csv 方法实现了快速读取 CSV（comma-separated) 文件到数据框的功能。 数据可视化库：matplotlib

a convolutional layer which receives a 3-channel input. Each individual 3x3 matrix is a kernel. A column of 3 kernels represents a channel. As you might notice, with such structured sparsity we can obtain

com/ Misc • Dynamic VS Static Computation Graph a b x_train_tensor Epoch 1 Misc • Dynamic VS Static Computation Graph a b yhat x_train_tensor Misc • Dynamic VS Static Computation Graph a b x_train_tensor y_train_tensor Misc • Dynamic VS Static Computation Graph a b x_train_tensor Epoch 2 Misc • Dynamic VS Static Computation Graph a b yhat x_train_tensor Misc • Dynamic VS Static Computation Graph a b x_train_tensor x_train_tensor yhat loss loss y_train_tensor Misc • Dynamic VS Static Computation Graph Building the graph and computing the graph happen at the same time. Seems inefficient, especially if we

空间。 10.6. 自注意力和位置编码 411 P = P[0, :, :].unsqueeze(0).unsqueeze(0) d2l.show_heatmaps(P, xlabel='Column (encoding dimension)', ylabel='Row (position)', figsize=(3.5, 4), cmap='Blues') 相对位置信息 除了捕获绝对 print(f'performance in Gigaflops: element {gigaflops[0]:.3f}, ' f'column {gigaflops[1]:.3f}, full {gigaflops[2]:.3f}') performance in Gigaflops: element 1.204, column 88.542, full 195.625 11.5.2 小批量 之前我们会理所当然地读取数据的 ai/t/2794 147 https://people.eecs.berkeley.edu/~rcs/research/interactive_latency.html 148 https://static.googleusercontent.com/media/research.google.com/en//people/jeff/Stanford‐DL‐Nov‐2010.pdf 149 http://inst

片的分布尽可能地接近， 从而保证了模型的泛化能力。 图 3.2 MNIST 数据集样例图片 现在来讨论图片的表示方法。一张图片包含了ℎ行(Height/Row)，?列(Width/Column)， 每个位置保存了像素(Pixel)值，像素值一般使用 0~255 的整形数值来表达颜色强度信息， 例如 0 表示强度最低，255 表示强度最高。如果是彩色图片，则每个像素点包含了 R、G、 加速度，型号年份，产地 column_names = ['MPG','Cylinders','Displacement','Horsepower','Weight', 'Acceleration', 'Model Year', 'Origin'] raw_dataset = pd.read_csv(dataset_path, names=column_names,