Regression and AI in practice · 回归与实践中的 AI
| English | 中文 | Pinyin · 拼音 |
|---|---|---|
| regression/rɪˈɡreʃn/ | 回归 | huí guī |
| classification/ˌklæsɪfɪˈkeɪʃn/ | 分类 | fēn lèi |
| linear regression/ˈlɪnɪə rɪˈɡreʃn/ | 线性回归 | xiàn xìng huí guī |
| OCR/ˌəʊ siː ˈɑː/ | 光学字符识别 | guāng xué zì fú shí bié |
Predicting a number, not a name
- Ask a model "is this email spam?" and it must choose from a fixed set of answers. Ask it "what will this house sell for?" and there is no set to choose from: the answer is a number, and being out by a thousand pounds is nearly right.
- Those are two different tasks, and mixing them up is the commonest mistake in this topic. Both are supervised learning; only the shape of the answer differs.
- This lesson is regression 回归 against classification 分类, how linear regression 线性回归 chooses its line, and how real products chain trained models into a pipeline.
预测一个数,而不是一个名字
- 问模型"这封邮件是垃圾邮件吗?",它必须从一组固定的答案里选。问它"这套房子会卖多少钱?",却没有可选的集合:答案是一个数,而差一千英镑几乎就算对了。
- 这是两种不同的任务,把它们混为一谈是本单元最常见的错误。两者都是监督学习;不同的只是答案的形状。
- 这一课讲回归(regression)与分类(classification)的对比、线性回归(linear regression)怎样选出它那条直线,以及真实产品怎样把训练好的模型串成流水线。
Regression and classification
- Classification predicts a category from a fixed set: cat or dog, spam or not spam, which of ten digits.
- Regression predicts a number on a continuous scale: a house price, tomorrow's temperature, how long a journey will take.
- Both are supervised: both learn from labelled examples. The choice between them is decided by one question: is the answer a label or a number?
- The distinction changes how you measure success. A classifier is right or wrong; a regressor is a certain distance out, which is why their loss functions differ.
回归与分类
- 分类从一个固定集合中预测一个类别:猫还是狗、垃圾邮件还是不是、十个数字中的哪一个。
- 回归在连续的尺度上预测一个数:房价、明天的气温、一次行程要多久。
- 两者都是监督的:都从有标签的样例中学习。在它们之间的选择由一个问题决定:答案是一个标签还是一个数?
- 这个区分改变了衡量成败的方式。分类器要么对要么错;回归器则是差了多少,这就是它们的损失函数不同的原因。
The difference between regression and classification is that regression predicts: · 回归和分类之间的区别是回归预测:
Regression outputs a continuous number (price, temperature); classification outputs a category. · 回归输出一个连续的数(价格、温度);分类输出一个类别。
Regression predicts a continuous number (a price, a temperature), while classification predicts a category (cat vs dog). · 回归预测一个连续的数(一个价格、一个温度),而分类预测一个类别(猫对狗)。
Number out = regression; category out = classification — the two main kinds of supervised prediction. · 输出数 = 回归;输出类别 = 分类——监督预测的两个主要种类。
Worked example: which task is which
- Predicting which of five customer types a shopper belongs to. Classification: the answer is one of a fixed set of categories.
- Predicting how much that shopper will spend next month. Regression: the answer is a number, and being close counts as nearly right.
- Predicting whether a machine will fail in the next week. Classification: yes or no. Predicting how many days until it fails. Regression.
- The same underlying data can feed either. Read the question being asked, not the subject matter.
例题:哪个任务是哪一种
- 预测一位购物者属于五种顾客类型中的哪一种。 分类:答案是固定类别集合中的一个。
- 预测那位购物者下个月会花多少钱。 回归:答案是一个数,接近就算基本正确。
- 预测一台机器下周是否会故障。 分类:是或否。预测还有几天会故障。 回归。
- 同样的底层数据可以喂给任何一种。要读被问的问题,而不是看题材。
Match each prediction task to its kind. · 把每个预测任务与它的种类配对。
The same data can feed either. Read the question: a label from a fixed set, or a number on a scale. · 同样的数据两种都能喂。要读问题:是固定集合中的一个标签,还是尺度上的一个数。
Linear regression
- Linear regression fits a straight line, or a flat surface when there are several inputs:
- The coefficients are chosen to minimise the sum of the squared errors between the line's predictions and the actual training values.
- Squared, for two reasons the exam accepts: errors above and below the line then cannot cancel out, and a large error is penalised much more than several small ones.
The line that leaves the least total squared distance
线性回归
- 线性回归拟合一条直线,当输入有多个时则是一个平面:
- 这些系数被选来最小化平方误差之和——即直线的预测值与训练数据实际值之差的平方和。
- 用平方,考试接受两个理由:这样线上方和下方的误差不会互相抵消,而且一个大误差受到的惩罚远大于几个小误差。
留下最小总平方距离的那条线
Fitting a regression line · 拟合一条回归线
Drag the controls. Linear regression draws the straight line that makes the squared distances to the data points as small as possible — then it predicts a number for any new input. · 拖动控件。线性回归画出使到数据点的平方距离尽可能小的那条直线——然后它为任何新输入预测一个数。
Linear regression chooses its coefficients to: · 线性回归选择它的系数来:
It fits the line/hyperplane that makes the squared differences from the data points as small as possible. · 它拟合使与数据点的平方差尽可能小的那条线/超平面。
Why does linear regression minimise the sum of the SQUARED errors rather than the errors themselves? · 线性回归为什么最小化误差的平方和,而不是误差本身?
Both reasons are accepted. Without squaring, a line could have a huge error above and below and still score zero. · 两个理由都被接受。不平方的话,一条线上下各有巨大误差却仍可能得零分。
When a straight line is the right model
- Use linear regression when the relationship looks roughly linear and you want a model a human can read: each coefficient says how much that input matters, and in which direction.
- When the data curves, the same idea still applies with a different model: polynomial regression, a decision tree, or a neural network. In every case the recipe is the same, choose a model, define a loss, minimise it.
- That recipe is what connects this lesson to the last one: gradient descent is simply how the minimising is done when there is no formula for the answer.
什么时候直线是对的模型
- 当关系看起来大致线性、而且你想要一个人能读懂的模型时,就用线性回归:每个系数说出那个输入有多重要、朝哪个方向。
- 当数据是弯的,同样的想法换个模型仍然适用:多项式回归、决策树,或神经网络。每一种情况下配方都一样:选一个模型、定义一个损失、把它最小化。
- 这个配方正是本课与上一课的连接:当答案没有公式时,梯度下降不过是完成"最小化"的方式。
AI in a real product
- Real systems rarely use one model. They chain several trained models into a pipeline, each doing one step.
- Face identification at a door: a camera captures a face, image recognition converts it into a numerical representation, and that is matched against the representations of registered people.
- Reading a sign aloud for a traveller: image recognition locates the text, OCR 光学字符识别 turns the pixels into characters, machine translation converts the words, and text-to-speech produces the audio.
- The exam gives a scenario and asks which models are used, in order. Name the stage and what it produces for the next one.
真实产品里的 AI
- 真实系统很少只用一个模型。它们把几个训练好的模型串成一条流水线,每个做一步。
- 门口的人脸识别:摄像头捕捉人脸,图像识别把它转换成数值表示,再拿它与已登记者的表示做匹配。
- 为旅行者朗读路牌:图像识别定位文字,光学字符识别(OCR)把像素变成字符,机器翻译转换词语,文本转语音产生音频。
- 考试给一个情境,问用到哪些模型、按什么顺序。要说出每个阶段,以及它交给下一个阶段什么。
Match each AI idea to what it means. · 把每个 AI 想法与它的含义配对。
Training fits the model once; inference is the quick prediction each time a user interacts. · 训练拟合模型一次;推理是每次一个用户交互时的快速预测。
A "read a sign aloud" feature typically chains which models? · 一个“大声读一个标志”的功能通常链接哪些模型?
Several trained models form a pipeline: find the text, extract characters, translate, then speak. · 几个训练好的模型形成一个流水线:找到文字,提取字符,翻译,然后说出来。
Put the stages of an app that reads a foreign sign aloud in order. · 把一个朗读外文路牌的应用的各阶段按顺序排列。
Several trained models chained together, each handing its output to the next. Name the stage and what it produces. · 几个训练好的模型串在一起,每个把输出交给下一个。要说出阶段和它产出什么。
Worked example: training against inference
- A phone app recognises a plant from a photograph instantly, yet training its model took weeks on many machines. Explain why.
- Training repeatedly runs forward and backward passes over a very large labelled dataset, adjusting millions of weights over many epochs. That is what took weeks.
- Inference, using the trained model, is a single forward pass through a network whose weights are now fixed: multiply, add, activate, layer by layer, once.
- The intelligence lives in the learned weights, which is why the finished model is small enough and fast enough to ship on a phone.
例题:训练与推理的对比
- 一个手机应用能瞬间从照片认出一种植物,而训练它的模型却在许多台机器上花了几周。解释为什么。
- 训练在一个非常大的有标签数据集上反复进行前向和反向传播,历经很多轮次调整数百万个权重。那才是花掉几周的事。
- 推理,即使用训练好的模型,是在权重已固定的网络上做一次前向传播:逐层相乘、相加、激活,一次而已。
- 智能存在于学到的权重里,这就是完成后的模型足够小、足够快、能装进手机的原因。
Using a trained model to make a prediction requires only a single ____ pass. · 用训练好的模型做一次预测,只需要一次____传播。
Training repeats forward and backward passes over a huge dataset; inference runs the fixed weights once, which is why it is instant. · 训练在庞大数据集上反复做前向和反向传播;推理只用固定的权重跑一次,这就是它瞬间完成的原因。
Marks that slip away
- Classification gives a category, regression gives a number. Both are supervised; that is not the difference.
- Linear regression minimises the sum of squared errors, not "the total distance". Say squared, and say why.
- A pipeline answer must give the stages in order and say what each hands to the next.
- Training is repeated forward and backward passes; inference is one forward pass. Do not describe training when the question asks about use.
容易丢掉的分
- **分类给出类别,回归给出数字。**两者都是监督的;那不是区别所在。
- 线性回归最小化的是平方误差之和,不是"总距离"。要说平方,并说出为什么。
- 流水线的答案必须按顺序给出各阶段,并说出每个交给下一个什么。
- 训练是反复的前向和反向传播;推理是一次前向传播。题目问使用时不要去描述训练。
You've got it
- classification predicts a category, regression predicts a number; both are supervised, and the question's answer shape decides which
- linear regression fits $y = m_1x_1 + \ldots + c$ by minimising the sum of squared errors, so errors cannot cancel and large ones are penalised heavily
- curved data needs another model, but the recipe is unchanged: choose a model, define a loss, minimise it
- real products chain models: image recognition, OCR, translation and text-to-speech; training is many forward and backward passes, inference is one forward pass
你掌握了
- 分类预测类别,回归预测数字;两者都是监督的,由问题答案的形状决定用哪个
- 线性回归通过最小化平方误差之和来拟合 $y = m_1x_1 + \ldots + c$,这样误差不会抵消,大误差受重罚
- 弯曲的数据需要别的模型,但配方不变:选模型、定损失、最小化
- 真实产品把模型串起来:图像识别、OCR、翻译和文本转语音;训练是许多次前向和反向传播,推理是一次前向传播