azure

云計(jì)算 ， 機(jī)器學(xué)習(xí) (Cloud Computing, Machine Learning)

A common question often asked in Data Science is:-

數(shù)據(jù)科學(xué)中經(jīng)常問(wèn)到的一個(gè)常見(jiàn)問(wèn)題是：

Which machine learning algorithm should I use?

我應(yīng)該使用哪種機(jī)器學(xué)習(xí)算法？

While there is no Magic-Algorithm that solves all business problems with zero errors, the algorithm you select should depend on two distinct parts of your Data Science scenario…

雖然沒(méi)有魔術(shù)算法能夠以零錯(cuò)誤解決所有業(yè)務(wù)問(wèn)題，但您選擇的算法應(yīng)取決于數(shù)據(jù)科學(xué)場(chǎng)景的兩個(gè)不同部分……

What do you want to do with your data?: Specifically, what is the business question you want to answer by learning from your past data?

您想對(duì)數(shù)據(jù)做什么？：具體來(lái)說(shuō)，您想從過(guò)去的數(shù)據(jù)中學(xué)到什么業(yè)務(wù)問(wèn)題？

What are the requirements of your Data Science scenario?: Specifically, what is the accuracy, training time, linearity, number of parameters, and number of features your solution supports?

您的數(shù)據(jù)科學(xué)方案的要求是什么 ？：具體來(lái)說(shuō)，您的解決方案支持的準(zhǔn)確性，訓(xùn)練時(shí)間，線性，參數(shù)數(shù)量和功能數(shù)量是多少？

The ML Algorithm cheat sheet helps you choose the best machine learning algorithm for your predictive analytics solution. Your decision is driven by both the nature of your data and the goal you want to achieve with your data.

ML算法備忘單可幫助您為預(yù)測(cè)分析解決方案選擇最佳的機(jī)器學(xué)習(xí)算法。 您的決定取決于數(shù)據(jù)的性質(zhì)和要通過(guò)數(shù)據(jù)實(shí)現(xiàn)的目標(biāo)。

The Machine Learning Algorithm Cheat-sheet was designed by Microsoft Azure Machine Learning (AML), to specifically answer this question:-

機(jī)器學(xué)習(xí)算法備忘單是由Microsoft Azure機(jī)器學(xué)習(xí)(AML)設(shè)計(jì)的 ，專門(mén)用于回答以下問(wèn)題：

What do you want to do with your data?

您想如何處理您的數(shù)據(jù)？

數(shù)據(jù)科學(xué)方法論： (The Data Science Methodology:)

I must state here that we need to have a solid understanding of the iterative system of methods that guide Data Scientists on the ideal approach to solving problems using the Data Science Methodology. Otherwise, we may never fully understand the essence of the ML Algorithm Cheat-sheet.

在此我必須指出，我們需要對(duì)方法的迭代系統(tǒng)有深入的了解，這些方法可以指導(dǎo)數(shù)據(jù)科學(xué)家使用數(shù)據(jù)科學(xué)方法論解決問(wèn)題的理想方法 。 否則，我們可能永遠(yuǎn)無(wú)法完全理解ML算法備忘單的本質(zhì)。

Azure機(jī)器學(xué)習(xí)算法備忘單： (The Azure Machine Learning Algorithm Cheat-sheet:)

The AML cheat-sheet is designed to serve as a starting point, as we try to choose the right model for predictive or descriptive analysis. It is based on the fact that there is simply no substitute for understanding the principles of each algorithm and the system that generated your data.

AML速查表旨在作為起點(diǎn)，因?yàn)槲覀儑L試選擇正確的模型進(jìn)行預(yù)測(cè)或描述性分析。 它基于這樣一個(gè)事實(shí)，那就是無(wú)可替代地理解每種算法和生成數(shù)據(jù)的系統(tǒng)的原理。

The AML Algorithm cheat-sheet can be downloaded here.

AML算法備忘單可在 此處 下載 。

img_creditimg_credit

AML算法備忘單概述： (Overview of the AML Algorithm Cheat-sheet:)

The Cheat-sheet covers a broad library of algorithms from classification, recommender systems, clustering, anomaly detection, regression, and text analytics families.

備忘單涵蓋了廣泛的算法庫(kù)，這些算法來(lái)自分類 ， 推薦系統(tǒng) ， 聚類 ， 異常檢測(cè) ， 回歸和文本分析系列。

Every machine learning algorithm has its own style or inductive bias. So, for a specific problem, several algorithms may be appropriate, and one algorithm may be a better fit than others.

每種機(jī)器學(xué)習(xí)算法都有自己的風(fēng)格或歸納偏差。 因此，對(duì)于一個(gè)特定的問(wèn)題，幾種算法可能是合適的，并且一種算法可能比其他算法更合適。

But it’s not always possible to know beforehand which is the best fit. Therefore, In cases like these, several algorithms are listed together in the Cheat-sheet. An appropriate strategy would be to compare the performance of related algorithms and choose the best-befitting to the requirements of the business problem and data science scenario.

但是，并非總是可能事先知道哪種方法最合適。 因此，在這種情況下，速查表中同時(shí)列出了幾種算法。 合適的策略是比較相關(guān)算法的性能，并選擇最適合業(yè)務(wù)問(wèn)題和數(shù)據(jù)科學(xué)場(chǎng)景的需求。

Bear in mind that the machine learning process is a highly iterative process.

請(qǐng)記住，機(jī)器學(xué)習(xí)過(guò)程是一個(gè)高度迭代的過(guò)程。

AML算法備忘單應(yīng)用程序： (AML Algorithm Cheat-sheet Applications:)

1.文字分析： (1. Text Analytics:)

If the solution requires extracting information from text, then text analytics can help derive high-quality information, to answer questions like:-

如果解決方案需要從文本中提取信息，則文本分析可以幫助獲取高質(zhì)量的信息，以回答諸如以下的問(wèn)題：

What information is in this text?

本文中有什么信息？

Text-based algorithms listed in the AML Cheat-sheet include the following:

AML備忘單中列出的基于文本的算法包括以下內(nèi)容：

• Extract N-Gram Features from Text: This helps to featurize unstructured text data, creating a dictionary of n-grams from a column of free text.

  提取的N-gram從文本功能 ：這有助于特征化非結(jié)構(gòu)化的文本數(shù)據(jù)，從自由文本的一列創(chuàng)建的正克的字典。

• Feature Hashing: Used to transform a stream of English text into a set of integer features that can be passed to a learning algorithm to train a text analytics model.

  功能散列： 用過(guò)的 將英語(yǔ)文本流轉(zhuǎn)換為一組整數(shù)特征，可以將這些整數(shù)特征傳遞給學(xué)習(xí)算法以訓(xùn)練文本分析模型。

• Preprocess Text: Used to clean and simplify texts. It supports common text processing operations such as stop-words-removal, lemmatization, case-normalization, identification, and removal of emails and URLs.

  預(yù)處理文本： 用過(guò)的 清理和簡(jiǎn)化文本。 它支持常見(jiàn)的文本處理操作，例如停用詞刪除，詞形還原，大小寫(xiě)規(guī)范化，標(biāo)識(shí)以及電子郵件和URL的刪除。

• Word2Vector: Converts words to values for use in NLP tasks, like recommender, named entity recognition, machine translation.

  Word2Vector： 將單詞轉(zhuǎn)換為用于NLP任務(wù)(例如推薦器，命名實(shí)體識(shí)別，機(jī)器翻譯)的值。

2.回歸： (2. Regression:)

We may need to make predictions on future continuous values such as the rate-of-infections and so on… These can help us answer questions like:-

我們可能需要對(duì)未來(lái)的連續(xù)值進(jìn)行預(yù)測(cè)，例如感染率等。這些可以幫助我們回答以下問(wèn)題：

How much or how many?

多少個(gè)？

Regression algorithms listed in the AML Cheat-sheet include the following:

AML備忘單中列出的回歸算法包括以下內(nèi)容：

• Fast Forest Quantile Regression: -> Predicts a distribution.

  快速森林分位數(shù)回歸： ->預(yù)測(cè)分布。

• Poisson Regression: -> Predicts event counts.

  泊松回歸： ->預(yù)測(cè)事件計(jì)數(shù)。

• Linear Regression: -> Fast training linear model.

  線性回歸： ->快速訓(xùn)練線性模型。

• Bayesian Linear Regression: -> Linear model, small data sets

  貝葉斯線性回歸： ->線性模型，小型數(shù)據(jù)集

• Decision Forest Regression: -> Accurate, fast training times

  決策森林回歸： ->準(zhǔn)確，快速的培訓(xùn)時(shí)間

• Neural Network Regression: -> Accurate, long training times

  神經(jīng)網(wǎng)絡(luò)回歸： ->準(zhǔn)確，長(zhǎng)訓(xùn)練時(shí)間

• Boosted Decision Tree Regression: -> Accurate, fast training times, large memory footprint

  增強(qiáng)的決策樹(shù)回歸： ->準(zhǔn)確，快速的培訓(xùn)時(shí)間，大內(nèi)存占用

3.推薦人： (3. Recommenders:)

Well, just like Netflix and Medium, we can generate recommendations for our users or clients, by using algorithms that perform remarkably well on content and collaborative filtering tasks. These algorithms can help answer questions like:-

好吧，就像Netflix和Medium，我們可以使用在內(nèi)容和協(xié)作過(guò)濾任務(wù)上表現(xiàn)出色的算法為用戶或客戶生成推薦。 這些算法可以幫助回答以下問(wèn)題：

What will they be interested in?

他們會(huì)對(duì)什么感興趣？

The Recommender algorithm listed in the AML Cheat-sheet includes:-

AML備忘單中列出的推薦算法包括：-

• SVD Recommender: -> The SVD Recommender is based on the Single Value Decomposition (SVD) algorithm. It can generate two different kinds of predictions:

  SVD推薦器： -> SVD推薦器基于單值分解 ( SVD )算法。 它可以生成兩種不同的預(yù)測(cè)：

Predict ratings for a given user and item.

預(yù)測(cè)給定用戶和項(xiàng)目的等級(jí) 。

Recommend items to a user

向用戶推薦商品

The SVD Recommender also has the following features:- Collaborative filtering, better performance with lower cost by reducing the dimensionality.

SVD Recommender還具有以下功能：-協(xié)作過(guò)濾，通過(guò)降低尺寸，以更低的成本獲得了更好的性能。

4.聚類： (4. Clustering:)

If we want to seek out the hidden structures in our data and to separate similar data points into intuitive groups, then we can use clustering algorithms to answer questions like:-

如果我們想找出數(shù)據(jù)中的隱藏結(jié)構(gòu)并將相似的數(shù)據(jù)點(diǎn)分為直觀的組，則可以使用聚類算法來(lái)回答以下問(wèn)題：

How is this organized?

這是如何組織的？

The Clustering algorithm listed in the AML Cheat-sheet include:-

AML備忘單中列出的聚類算法包括：-

• K-Means: -> K-means is one of the simplest and the best known unsupervised learning algorithms. You can use the algorithm for a variety of machine learning tasks, such as:

  K均值： -> K均值是最簡(jiǎn)單，最著名的無(wú)監(jiān)督學(xué)習(xí)算法之一。 您可以將算法用于各種機(jī)器學(xué)習(xí)任務(wù)，例如：

Detecting abnormal data

檢測(cè)異常數(shù)據(jù)

Clustering text documents

聚類文本文檔

Analyzing datasets before we use other classification or regression methods.

在使用其他分類或回歸方法之前，先分析數(shù)據(jù)集。

5.異常檢測(cè)： (5. Anomaly Detection:)

This technique is useful as we try to identify and predict rare or unusual data points. For example in IoT data, we could use anomaly-detection to detect and raise an alarm as we analyze the logs-data of a machine. This could be used to identify strange IP addresses or unusually high attempts to access the system or any other anomaly that could pose a serious threat.

當(dāng)我們嘗試識(shí)別和預(yù)測(cè)稀有或異常數(shù)據(jù)點(diǎn)時(shí)，此技術(shù)很有用。 例如，在物聯(lián)網(wǎng)數(shù)據(jù)中，當(dāng)我們分析機(jī)器的日志數(shù)據(jù)時(shí)，我們可以使用異常檢測(cè)來(lái)檢測(cè)并發(fā)出警報(bào)。 這可用于識(shí)別奇怪的IP地址或異常高的訪問(wèn)系統(tǒng)嘗試或任何其他可能構(gòu)成嚴(yán)重威脅的異常情況。

Anomaly detection can be used to answer questions like:-

異常檢測(cè)可用于回答以下問(wèn)題：

Is this weird, is this abnormal?

這很奇怪，這異常嗎？

The Anomaly detection algorithm listed in the AML Cheat-sheet include:-

AML速查表中列出的異常檢測(cè)算法包括：-

• PCA-Based Anomaly Detection: -> For example, to detect fraudulent transactions, you often don’t have enough examples of fraud to train on. But you might have many examples of good transactions.

  基于PCA的異常檢測(cè)： ->例如，要檢測(cè)欺詐性交易，您通常沒(méi)有足夠的欺詐性實(shí)例來(lái)進(jìn)行培訓(xùn)。 但是您可能有許多交易良好的例子。

The PCA-Based Anomaly Detection module solves the problem by analyzing available features to determine what constitutes a “normal” class. The module then applies distance metrics to identify cases that represent anomalies.

基于PCA的異常檢測(cè)模塊通過(guò)分析可用功能以確定什么構(gòu)成“正常”類來(lái)解決該問(wèn)題。 然后，該模塊應(yīng)用距離度量來(lái)識(shí)別代表異常的案例。

This approach lets you train a model by using existing imbalanced data. PCA records fast training times.

通過(guò)這種方法，您可以使用現(xiàn)有的不平衡數(shù)據(jù)來(lái)訓(xùn)練模型。 PCA記錄了快速的培訓(xùn)時(shí)間。

• Train Anomaly Detection model: -> This takes as input, a set of parameters for an anomaly detection model, and an unlabeled dataset. It returns a trained anomaly detection model, together with a set of labels for the training data.

  訓(xùn)練異常檢測(cè)模型 ： ->將異常檢測(cè)模型的一組參數(shù)和未標(biāo)記的數(shù)據(jù)集作為輸入。 它返回一個(gè)經(jīng)過(guò)訓(xùn)練的異常檢測(cè)模型，以及一組用于訓(xùn)練數(shù)據(jù)的標(biāo)簽。

6.多類分類： (6. Multi-Class Classification:)

Often, we may need to pick the right answers from complex questions with multiple possible answers. For tasks like these, we need a Multi-class classification algorithm. This can help us answer questions like:-

通常，我們可能需要從具有多個(gè)可能答案的復(fù)雜問(wèn)題中選擇正確的答案。 對(duì)于此類任務(wù)，我們需要一種多類分類算法。 這可以幫助我們回答以下問(wèn)題：

Is this A or B or C or D?

這是A或B還是C或D？

Multi-class algorithms listed in the AML Cheat-sheet include the following:

AML備忘單中列出的多類算法包括以下內(nèi)容：

• Multi-Class Logistic Regression: -> Logistic regression is a well-known method in statistics that is used to predict the probability of an outcome, and is popular for classification tasks. The algorithm predicts the probability of occurrence of an event by fitting data to a logistic function.

  多類Logistic回歸： -> Logistic回歸是統(tǒng)計(jì)中眾所周知的方法，用于預(yù)測(cè)結(jié)果的概率，并且在分類任務(wù)中很流行。 該算法通過(guò)將數(shù)據(jù)擬合到邏輯函數(shù)來(lái)預(yù)測(cè)事件發(fā)生的概率。

Usually a Binary-Classifier, but in Multi-class logistic regression, the algorithm is used to predict multiple outcomes.

通常是二進(jìn)制分類器，但在多類邏輯回歸中，該算法用于預(yù)測(cè)多個(gè)結(jié)果。

Features: Fast training times, linear model.

特點(diǎn) ：快速的訓(xùn)練時(shí)間，線性模型。

• Multi-class Neural Network: -> A neural network is a set of interconnected layers. The inputs are the first layer and are connected to an output layer by an acyclic graph comprised of weighted edges and nodes.

  多類神經(jīng)網(wǎng)絡(luò)： ->神經(jīng)網(wǎng)絡(luò)是一組相互連接的層。 輸入是第一層，并通過(guò)包含加權(quán)邊和節(jié)點(diǎn)的非循環(huán)圖連接到輸出層。

Between the input and output layers, you can insert multiple hidden layers. Most predictive tasks can be accomplished easily with only one or a few hidden layers. However, recent research has shown that deep neural networks (DNN) with many layers can be effective in complex tasks such as image or speech recognition, with successive layers used to model increasing levels of semantic depth.

在輸入和輸出層之間，可以插入多個(gè)隱藏層。 僅需一層或幾層隱藏層即可輕松完成大多數(shù)預(yù)測(cè)性任務(wù)。 但是，最近的研究表明，具有多層結(jié)構(gòu)的深度神經(jīng)網(wǎng)絡(luò)(DNN)在諸如圖像或語(yǔ)音識(shí)別之類的復(fù)雜任務(wù)中可能是有效的，連續(xù)的層用于對(duì)語(yǔ)義深度的遞增級(jí)別進(jìn)行建模。

Features: Accuracy, long training times.

特點(diǎn) ：準(zhǔn)確性高，訓(xùn)練時(shí)間長(zhǎng)。

• Multiclass Decision Forest: -> The decision forest algorithm is an ensemble learning method for classification.

  多類決策森林：-> 決策森林算法是用于分類的整體學(xué)習(xí)方法。

Decision trees, in general, are non-parametric models, meaning they support data with varied distributions. In each tree, a sequence of simple tests is run for each class, increasing the levels of a tree structure until a leaf node (decision) is reached.

通常，決策樹(shù)是非參數(shù)模型，這意味著它們支持具有不同分布的數(shù)據(jù)。 在每棵樹(shù)中，為每個(gè)類運(yùn)行一系列簡(jiǎn)單測(cè)試，從而增加樹(shù)結(jié)構(gòu)的級(jí)別，直到達(dá)到葉節(jié)點(diǎn)(決策)為止。

Features: Accuracy, fast training times.

特點(diǎn)：準(zhǔn)確性，快速的訓(xùn)練時(shí)間。

• One-vs-All Multiclass: -> This algorithm implements the one-versus-all method, in which a binary model is created for each of the multiple output classes. In essence, it creates an ensemble of individual models and then merges the results, to create a single model that predicts all classes.

  一對(duì)多所有類： ->此算法實(shí)現(xiàn)了一對(duì)多方法，其中為多個(gè)輸出類中的每一個(gè)創(chuàng)建二進(jìn)制模型。 本質(zhì)上，它創(chuàng)建了單個(gè)模型的集合，然后合并結(jié)果，以創(chuàng)建一個(gè)預(yù)測(cè)所有類的模型。

Any binary classifier can be used as the basis for a one-versus-all model

任何二進(jìn)制分類器都可以用作“一對(duì)多”模型的基礎(chǔ)

Features: Depends on the two-class classifier.

特點(diǎn)：取決于兩類分類器。

• Multiclass Boosted Decision Tree: -> This algorithm creates a machine learning model that is based on the boosted decision trees algorithm.

  多類增強(qiáng)決策樹(shù)： ->此算法創(chuàng)建基于增強(qiáng)決策樹(shù)算法的機(jī)器學(xué)習(xí)模型。

A boosted decision tree is an ensemble learning method in which the second tree corrects for the errors of the first tree, the third tree corrects for the errors of the first and second trees, and so forth. Predictions are based on the ensemble of trees together.

增強(qiáng)決策樹(shù)是一種集成學(xué)習(xí)方法，其中第二棵樹(shù)糾正第一棵樹(shù)的錯(cuò)誤，第三棵樹(shù)糾正第一棵樹(shù)和第二棵樹(shù)的錯(cuò)誤，依此類推。 預(yù)測(cè)是基于樹(shù)木的整體。

Features: Non-parametric, fast training times, and scalable.

特點(diǎn)：非參數(shù)，快速的培訓(xùn)時(shí)間和可擴(kuò)展性。

7.二進(jìn)制分類： (7. Binary Classification:)

Binary classification tasks are the most common classification tasks. These often involve a yes or no, true or false, type of response. Binary classification algorithms help us to answer questions like:-

二進(jìn)制分類任務(wù)是最常見(jiàn)的分類任務(wù)。 這些通常涉及是或否，正確或錯(cuò)誤的響應(yīng)類型。 二進(jìn)制分類算法可幫助我們回答以下問(wèn)題：

Is this A or B?

這是A還是B？

Binary-classifier algorithms listed in the AML Cheat-sheet include the following:

AML備忘單中列出的二進(jìn)制分類器算法包括以下內(nèi)容：

• Two-Class Support Vector Machine: -> Support vector machines (SVMs) are a well-researched class of supervised learning methods. This particular implementation is suited to the prediction of two possible outcomes, based on either continuous or categorical variables.

  兩類支持向量機(jī)： ->支持向量機(jī)(SVM)是一種經(jīng)過(guò)嚴(yán)格研究的監(jiān)督學(xué)習(xí)方法。 這種特定的實(shí)現(xiàn)方式適合基于連續(xù)或分類變量來(lái)預(yù)測(cè)兩種可能的結(jié)果。

Features: Under 100 features, linear model.

特征：根據(jù)100個(gè)特征，線性模型。

• Two-Class Averaged Perceptron: -> The averaged perceptron method is an early and simple version of a neural network. In this approach, inputs are classified into several possible outputs based on a linear function, and then combined with a set of weights that are derived from the feature vector — hence the name “perceptron.”

  兩類平均感知器： -> 平均感知器方法是神經(jīng)網(wǎng)絡(luò)的早期和簡(jiǎn)單版本。 在這種方法中，基于線性函數(shù)將輸入分類為幾個(gè)可能的輸出，然后與從特征向量派生的一組權(quán)重結(jié)合在一起，因此得名“ perceptron”。

Features: Fast training, linear model.

特點(diǎn)：快速訓(xùn)練，線性模型。

• Two-Class Decision Forest: -> Decision forests are fast, supervised ensemble models. This algorithm is a good choice if you want to predict a target with a maximum of two outcomes. Generally, ensemble models provide better coverage and accuracy than single decision trees.

  兩級(jí)決策森林： ->決策森林是快速，受監(jiān)督的集成模型。 如果要預(yù)測(cè)最多兩個(gè)結(jié)果的目標(biāo)，則此算法是一個(gè)不錯(cuò)的選擇。 通常，集成模型比單個(gè)決策樹(shù)提供更好的覆蓋范圍和準(zhǔn)確性。

Features: Accurate, fast training.

特點(diǎn)：準(zhǔn)確，快速的訓(xùn)練。

• Two-Class Logistic Regression: -> Logistic regression is a well-known statistical technique that is used for modeling many kinds of problems. This algorithm is a supervised learning method; therefore, you must provide a dataset that already contains the outcomes to train the model. In this method, the classification algorithm is optimized for dichotomous or binary variables only.

  兩類Logistic回歸： -> Logistic回歸是一種眾所周知的統(tǒng)計(jì)技術(shù)，用于對(duì)多種問(wèn)題進(jìn)行建模。 該算法是一種監(jiān)督學(xué)習(xí)方法。 因此，您必須提供一個(gè)已經(jīng)包含結(jié)果的數(shù)據(jù)集以訓(xùn)練模型。 在這種方法中，分類算法僅針對(duì)二分變量或二進(jìn)制變量進(jìn)行了優(yōu)化。

Features: Fast training, linear model.

特點(diǎn)：快速訓(xùn)練，線性模型。

• Two-Class Boosted Decision Tree: -> This method creates a machine learning model that is based on the boosted decision trees algorithm.

  兩類增強(qiáng)決策樹(shù)： ->此方法創(chuàng)建基于增強(qiáng)決策樹(shù)算法的機(jī)器學(xué)習(xí)模型。

Features: Accurate, fast training, large memory footprint.

特點(diǎn)：準(zhǔn)確，快速的培訓(xùn)，大內(nèi)存占用。

• Two-Class Neural Network: -> This algorithm is used to create a neural network model that can be used to predict a target that has only two values.

  兩類神經(jīng)網(wǎng)絡(luò)： ->他的算法用于創(chuàng)建神經(jīng)網(wǎng)絡(luò)模型，該模型可用于預(yù)測(cè)只有兩個(gè)值的目標(biāo)。

Classification using neural networks is a supervised learning method, and therefore requires a tagged dataset, which includes a label column. For example, you could use this neural network model to predict binary outcomes such as whether or not a patient has a certain disease, or whether a machine is likely to fail within a specified window of time.

使用神經(jīng)網(wǎng)絡(luò)進(jìn)行分類是一種有監(jiān)督的學(xué)習(xí)方法，因此需要帶有標(biāo)簽的數(shù)據(jù)集 ，其中包括標(biāo)簽列。 例如，您可以使用該神經(jīng)網(wǎng)絡(luò)模型來(lái)預(yù)測(cè)二進(jìn)制結(jié)果，例如患者是否患有某種疾病，或者機(jī)器是否有可能在指定的時(shí)間范圍內(nèi)發(fā)生故障。

Features: Accurate, long training times.

特點(diǎn)：準(zhǔn)確，訓(xùn)練時(shí)間長(zhǎng)。

8.圖像分類： (8. Image Classification:)

If the analysis requires extracting information from images, then computer vision algorithms can help us to derive high-quality information, to answer questions like:-

如果分析需要從圖像中提取信息，則計(jì)算機(jī)視覺(jué)算法可以幫助我們獲得高質(zhì)量的信息，以回答諸如以下的問(wèn)題：

What does this image represent?

該圖像代表什么？

The computer vision algorithms listed in the AML Cheat-sheet include:-

AML備忘單中列出的計(jì)算機(jī)視覺(jué)算法包括：-

• DenseNet and ResNet: -> These classification algorithms are supervised learning methods that require a labeled dataset. You can train the model by providing a labeled image directory as inputs. The trained model can then be used to predict values for new unseen input examples.

  DenseNet和ResNet： ->這些分類算法是需要標(biāo)記數(shù)據(jù)集的監(jiān)督學(xué)習(xí)方法。 您可以通過(guò)提供標(biāo)記的圖像目錄作為輸入來(lái)訓(xùn)練模型。 然后，可以將訓(xùn)練后的模型用于預(yù)測(cè)新的看不見(jiàn)的輸入示例的值。

Features: High accuracy, better efficiency.

特點(diǎn)：精度高，效率更高。

摘要： (Summary:)

The Azure Machine Learning Algorithm Cheat Sheet helps you with the first consideration: What do you want to do with your data? On the Machine Learning Algorithm Cheat Sheet, look for a task you want to do, and then find an Azure Machine Learning designer algorithm for the predictive analytics solution.

Azure機(jī)器學(xué)習(xí)算法備忘單可幫助您首先考慮以下事項(xiàng)： 您想對(duì)數(shù)據(jù)做什么？ 在機(jī)器學(xué)習(xí)算法備忘單上，查找要執(zhí)行的任務(wù)，然后找到用于預(yù)測(cè)分析解決方案的Azure機(jī)器學(xué)習(xí)設(shè)計(jì)器算法。

The Azure Machine Learning experience is quite intuitive and easy to grasp. The Azure Machine Learning designer is a drag-and-drop visual interface that makes it engaging and fun to build ML pipelines, assemble algorithms and run iterative ML jobs, build, train and deploy models all within the Azure portal. Once deployed, your models can be consumed by authorized, external, third-party applications in real-time.

Azure機(jī)器學(xué)習(xí)體驗(yàn)非常直觀并且易于掌握。 Azure機(jī)器學(xué)習(xí)設(shè)計(jì)器是一個(gè)拖放式可視化界面，它使構(gòu)建ML管道，組裝算法和運(yùn)行迭代ML作業(yè)，構(gòu)建，訓(xùn)練和部署模型都變得引人入勝且充滿樂(lè)趣。 部署后，您的模型可以被授權(quán)的外部第三方應(yīng)用程序?qū)崟r(shí)使用。

下一步： (Next Steps:)

After deciding on the right model to choose for the business problem, using the Azure Machine Learning Cheat-sheet, the next step is to answer the second question:-

在確定了適合業(yè)務(wù)問(wèn)題的正確模型之后，使用Azure機(jī)器學(xué)習(xí)備忘單 ，下一步是回答第二個(gè)問(wèn)題：

• What are the requirements of your Data Science scenario?: Specifically, what is the accuracy, training time, linearity, number of parameters, and number of features your solution supports?

  您的數(shù)據(jù)科學(xué)方案有哪些要求 ？：具體來(lái)說(shuō)，您的解決方案支持的準(zhǔn)確性，訓(xùn)練時(shí)間，線性，參數(shù)數(shù)量和功能數(shù)量是多少？

To get the best possible outcome for these metrics, kindly go through the details at the Azure Machine Learning site.

為了獲得這些指標(biāo)的最佳結(jié)果，請(qǐng)仔細(xì)查看Azure機(jī)器學(xué)習(xí)站點(diǎn)上的詳細(xì)信息。

Cheers!!

干杯！！

關(guān)于我： (About Me:)

Lawrence is a Data Specialist at Tech Layer, passionate about fair and explainable AI and Data Science. I hold both the Data Science Professional and Advanced Data Science Professional certifications from IBM. After earning the IBM Data Science Explainability badge, my mission is to promote Fairness and Explainability in AI… I love to code up my functions from scratch as much as possible. I love to learn and experiment…And I have a bunch of Data and AI certifications and I’ve written several highly recommended articles.

Lawrence是Tech Layer的數(shù)據(jù)專家，對(duì)公平和可解釋的AI和數(shù)據(jù)科學(xué)充滿熱情。 我同時(shí)擁有 IBM 的 Data Science Professional 和 Advanced Data Science Professional 認(rèn)證。 獲得 IBM數(shù)據(jù)科學(xué)可解釋性徽章后 ，我的任務(wù)是促進(jìn)AI的公平性和可解釋性。我喜歡盡可能地從頭開(kāi)始編寫(xiě)功能。 我喜歡學(xué)習(xí)和實(shí)驗(yàn)……而且我獲得了大量數(shù)據(jù)和AI認(rèn)證，并且撰寫(xiě)了幾篇強(qiáng)烈推薦的文章。

Feel free to connect with me on:-

隨時(shí)與我聯(lián)系：-

Github

Github

Linkedin

領(lǐng)英

Twitter

推特

翻譯自: https://medium.com/towards-artificial-intelligence/the-azure-ml-algorithm-cheat-sheet-451547832cad

azure

總結(jié)

以上是生活随笔為你收集整理的azure_Azure ML算法备忘单的全部?jī)?nèi)容，希望文章能夠幫你解決所遇到的問(wèn)題。

如果覺(jué)得生活随笔網(wǎng)站內(nèi)容還不錯(cuò)，歡迎將生活随笔推薦給好友。