100% ML: Diamond price prediction using machine learning, python, SVM, KNN, Neural networks

Detailed steps & Python code included.

Amongst a variety of items not quantitatively or statistically valued by buyers, Diamonds are possibly the most common. The purchase is far less from rational with a heavy bent on emotional ties. The jewelers would entice every man (and woman) by marketing it as a necessity for the occasion and as a status symbol, and by calling this pricey and unaffordable item as priceless. But what if you get the power to accurately estimate Diamond price before negotiating? Won’t that be a really cool edge!

debeers diamond industry

However, the actual value of a diamond is determined by a gemologist after inspecting its various “features” (using proper machine learning basics terminology now since this article is on Diamond price prediction using machine learning in python) and applying a relative valuation principle of “compare and price”.

CONSUMER’S GUIDE to Buying Diamond

This article is classified as a short course on diamond price prediction using Machine Learning with Python code

Machine Learning course with diamond price prediction project
diamonds square

Mini course describing detailed steps on Machine Learning basics with Python code explained using different Machine Learning models. The mini project here is "diamond price prediction" with dataset from Kaggle

Course Provider: Organization

Course Provider Name: Five Step Guide

Editor's Rating:

It’s holiday season! You can use machine learning regression for price prediction of the diamond you really desire — a little help in negotiating price with local retailers.

This article is a no-brainer, easy-to-follow article to grasp machine learning basics in 15 mins for a Linear regression problem. The machine learning examples use diamond price prediction dataset with Python to show how to predict a number using minimal dataset at a fairly good accuracy.

In last 2 decades, the valuation and pricing has become more or less quantitative i.e. calculations based on values of many properties not just limiting to 4Cs (carat, cut, colour, clarity).

Wikipedia – Diamond cutting

Properties like culet, pavilion, crown, girdle, girdle thickness polish, symmetry, fluorescence, table, depth and so on are the most easily identifiable and recordable features while the diamond is actually cut.

Let me now start by showing you DIAMOND price prediction using machine learning & Python. In this article, I will show very simple steps of a machine learning process using a diamond price prediction dataset from Kaggle.

Get the data

The first steps in Machine learning process are to Import all basic libraries.

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
%matplotlib inline
from sklearn import preprocessing

Let’s begin training by using the diamond price dataset from Kaggle  —  it has basic data on more than 54000 diamonds. In a follow-up article, we will train the same machine learning model to predict diamond prices based on DIAMOND MARKET LIVE DATA from pricescope.com.

df = pd.read_csv(“diamonds.csv”)
df.drop(‘Unnamed: 0’, axis=1, inplace=True)

The output is something like the following:

diamond price prediction dataset
read csv diamond

Data preprocessing

The next steps for diamond price prediction using Machine learning begin with basic data prepossessing. We will first check if any null values or unexpected data type are present in the Kaggle Diamond price prediction dataset to ensure accurate diamond price prediction using python.


Now we plot a Pair-plot of Price vs. 4 Cs (Carat, Cut, Color, Clarity) — the most popular and marketed properties of a diamond.

Read more about 4Cs at the following link:


# plot price vs. carat
sns.pairplot(df, x_vars=[’carat’], y_vars = [’price’])
# plot carat vs other Cs
sns.pairplot(df, x_vars=[’cut’, 'clarity’, 'color’], y_vars = [’carat’])

The output is something like the following:

pairplot diamond price prediction
pairplot diamond
50off50 9740499

We can see that the properties charts reveal a lot about how and where bulk of diamonds fall under each property value e.g. most bigger diamonds (higher carat) fall in Fair cut, I1 clarity and H-I color. These are poor (commercial grade) diamonds generally sold by retail jewelry shops across the world to attract consumers with advertisements like ‘Diamonds at 50% off

The price vs. carat chart also show that there are some outliers in the dataset i.e. few diamonds that are really over priced!

Now we need to see the distribution of the kaggle dataset for Diamond price prediction using Python. We will create a histogram plot for this. Being able to plot and visually interpret the models used in machine learning to predict diamond prices is a very vital part in completely grasping the motive of this article – learn Machine Learning basics.

Therefore, those steps in machine learning process needed for plotting and visually interpreting the data will be shown and explained everywhere in this article.

FiveStepGuide may earn an Affiliate Commission if you purchase something through links on this page.

I suggest you try some python course from Coursera or the brilliant Python Learning Path titled Linkedin Learning Python Essential Training. Click on the image below to start your 30 day trial.

linkedin learning python essential training

First we define the histplot() function.

def histplot(df, listvar):
 fig, axes = plt.subplots(nrows=1, ncols=len(listvar), figsize=(20, 3))
 for ax in axes:
 df.hist(column=listvar[counter], bins=20, ax=axes[counter])
 counter = counter+1

Now we list the continuous variables and leave out the categorical variable. A continuous variable e.g. carat is one which has numerical values whereas a categorical variable is the one with alphanumeric values as categories e.g. clarity

linear_vars = df.select_dtypes(include=[np.number]).columns

Output is [‘carat’, ‘depth’, ‘table’, ‘price’, ‘l’, ‘w’, ‘d’]

Now we plot the histogram


This revels the distribution of each property. As expected, we see that the data is not normally distributed. After all, how can you expect a 1 carat diamond to be priced just at twice the price of a half-carat given all properties remain the same, while a 1 carat diamond looks much bigger to the eye when in a ring or earrings for that matter than a half carat one.

Diamond database

histplot to predict price of diamonds
histplot diamond

You may like to read the following similar articles

Machine Learning model to predict car mileage
Detailed Python code & steps on several Machine Learning prediction algorithms to…
increase readability score in wordpress
Use SEO readability checker to check readability score in WordPress. How to…
How to QUICKLY setup FREE Cloudflare CDN for WordPress website. Cloudflare features,…

Convert the features to log scale

One of the key steps in Machine learning process is to convert features having bimodal distribution to ones with BINOMIAL distribution.

bimodal distribution
Bimodal distribution vs Binomial distribution

1. Check for any ZERO value

Amongst features namely tabledepthlwd, check if any continuous variable has zero value. This results in a division by zero error when converted to log. Add a tiny number 0.01 to any zero value.

print(‘0 values →’, 0 in df.values)
df[linear_vars] = df[linear_vars] + 0.01
print(‘Filled all 0 values with 0.01. Now any 0 values? →’, 0 in df.values)

The output is:

0 values --> True
Filled all 0 values with 0.01. Now any 0 values? --> False

A fantastic highly rated Udemy course on Machine learning. Click the video below for 30-day money back guarantee

udemy sundog1 sm

2. View and remove outliers using z-score

Since we could briefly sense some outliers in the pairplot charts, lets dwell deeper and see whether there genuinely are any outliers. Let’s first begin by printing top X values of each diamond

def sorteddf(df, listvar):
    for var in listvar:
        display('sorted by ' + var + ' --> ' + str(list(df[listvar].sort_values(by=var,ascending=False)[var].head())))

Output is like:

sorted by carat --> [5.02, 4.51, 4.14, 4.02, 4.02]'
sorted by depth --> [79.01, 79.01, 78.21000000000001, 73.61, 72.91000000000001]'
sorted by table --> [95.01, 79.01, 76.01, 73.01, 73.01]'
sorted by price --> [21646.459999999995, 21640.709999999995, 21626.909999999996, 21624.609999999997, 21623.459999999995]'
sorted by l --> [10.75, 10.24, 10.15, 10.03, 10.02]'
sorted by w --> [58.91, 31.810000000000002, 10.549999999999999, 10.17, 10.11]'
sorted by d --> [31.810000000000002, 8.07, 6.99, 6.7299999999999995, 6.4399999999999995]'

From this list itself, we see that there are some outliers for w,d. Lets visualize those using boxplots. Create a boxplot function

def dfboxplot(df, listvar):
 fig, axes = plt.subplots(nrows=1, ncols=len(listvar), figsize=(20, 3))
 for ax in axes:
 df.boxplot(column=listvar[counter], ax=axes[counter])
 counter = counter+1

Call dfboxplot to view outliers for all properties

dfboxplot(df, linear_vars)
dfboxplot of dataset used in diamond price prediction
dfboxplot diamond

We can now clearly visualize that there are outliers for tablew and d properties. Lets call removeoutliers() function to remove the outliers based on z-score. There are several methods to remov outliers but I am going to follow the z-score process here since its the easiest to implement and delivers optimal results — after all, this is a no-brainer quickie article for newbie users.

For more on all other alogs to remove outliers, check these out:

Stackoverflow: Outliers in Pandas dataframe

def removeoutliers(df, listvars, z):
    from scipy import stats
    for var in listvars:
        df1 = df[np.abs(stats.zscore(df[var])) < z]
    return df1
df = removeoutliers(df, linear_vars,2)

Now, after calling dfboxbplot again to view outliers for all properties, we see the output as:

dfboxplot diamond after removing outliers
dfboxplot diam2

3. Convert to log scale

Since we saw earlier that most features (or properties of a diamond) are not normally distributed, and one of the most favored approach, if not a prerequisite, is to use Gaussian distributed (another name for normally distributed) data.

Quora: Gaussian vs. normal distribution

Medium: Gaussian-distribution-in-data-science

So in order to do perfect diamond price prediction using machine learning, we would need to logarithmize the values in the Kaggle diamond price prediction dataset.

# this log converts dataframe's features inplace
def convertfeatures2log(df, listvars):
   for var in listvars:
      df[var] = np.log(df[var])convertfeatures2log(df, linear_vars)
histplot(df, linear_vars)

The output now is:

hisplot of kaggle diamonds dataset
hisplot diam2

Convert categorical variable column to numerical column using labelencoder

We now have to convert all categorical columns to numerical columns using labelencoder. You may read more about labelencoder in the following webpage — quick and easy to understand description there.

Towards Data Science: encoding categorical features

First we define the convert_catg() function to convert categorical columns to numerical columns

def convert_catg(df1):
 from sklearn.preprocessing import LabelEncoder
 le = LabelEncoder()
# Find the columns of object type along with their column index
  object_cols = list(df1.select_dtypes(exclude=[np.number]).columns)
  object_cols_ind = []
  for col in object_cols:
# Encode the categorical columns with numbers 
  for i in object_cols_ind:
    df1.iloc[:,i] = le.fit_transform(df1.iloc[:,i])

Next, we run the function and see the head of the dataframe.

convert category of diamond price prediction dataset

Learn label encoder, one hot encoding and other such nuances of data preparation using this free course from LinkedIn Learning. Start with your free trial by clicking the AI and Machine Learning Learning Path image below.

linkedin aiml

Let’s now prepare our data further to start building the Machine learning model to predict diamond prices.

Divide the data into independent variable features (X) and dependent variable (y)

The next step in our journey of diamond price prediction in python is to set the independent variable (X) and dependent variable (y).

X is the matrix (or DataFrame) for all the properties (independent features) and y is the vector for output (dependent variables) i.e. diamond price

X_df = df.drop([‘price’, ‘l’, ‘w’, ‘d’], axis=1)
y_df = df[[‘price’]] # two [[ to create a DF

Now, we determine correlations between price and all other attributes.

  • I will be combining both X (already converted categorical to numerical) and y to form a new dataframe for correlation
df_le = X_df.copy()
# add a new column in dataframe — join 2 dataframe columns-wise
df_le[‘price’] = y_df[‘price’].values

The statement df_le = X_df means that df_le will be like a pointer to X_df. Any change made to df_le will actually be a change to X_df. Therefore df_le = X_df.copy() is better.

  • It seems diamond price is highly correlated with carat and fairly with table, color and clarity, but not much with cut
correlation in kaggle diamonds dataset
category correlation

Note on Feature scaling — it seems its not needed here since we have already converted the diamond price prediction dataset properties into log. Nevertheless, if I had feature scaled, then the code would have been as written below:

from sklearn.preprocessing import StandardScaler
sc_X = StandardScaler()
X_df = sc_X.fit_transform(X_df)

Train Test Split

Data scientists generally split the dataset used for machine learning projects into either two or three subsets: 2 subsets for training and testing, while 3 for training, validation and testing. I will talk about it in detail a bit later. This data split prevents an algorithm from overfitting and underfitting. I have explained overfitting and underfitting briefly in my other post here:


The code for train_test_split is as follows:

from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X_df, y_df, test_size=0.3, random_state=42)

Now I will take you through the steps needed to predict diamond price using machine learning and Python on SVM, KNN, Neural networks models.

Once you are really ready with your clean & corrected diamond price prediction dataset, coding a simple Machine Learning algorithm in python is a cakewalk.

Let me now demonstrate a few machine learning algorithms with code, output and charts.

An outline of what we will be doing in the next few pages is:

  • split the diamond price prediction dataset into training set and test set.
  • Train the algorithm on training set data
  • Use the trained algorithm (or trained ML model) to estimate price from diamond properties in test data.
  • Verify / visualize / measure the differences between the predicted prices and actual prices using scatterplots, histograms, accuracy metrics etc.
ETF Advantages
Why etfs are good? Know these ETF Advantages before putting your money…
trade etfs for full time income sm
Know how ETF works | Read detailed information and steps to be…
Options trading for beginners
Introduction to Options Trading for beginners. Learn options trading via options trading…

Machine Learning algorithms

Now is the time to learn regression models for diamond price prediction and start coding machine learning algorithms on this dataset.

machine learning regression for diamond price prediction

After completing all necessary data pre-processing steps, let’s move ahead and see some ML algorithms in action.

Linear regression

Let’s start with the simplest of all, the ubiquitous linear regression model to predict price of a diamond.

  1. Import LinearRegression class from Sci-kit learn
  2. Create an object of LinearRegression model
  3. Fit the model to X_train and y_train
  4. Make predictions
# Import the class from Sci-kit learn
from sklearn.linear_model import LinearRegression
# Create an object of LinearRegression model
reg_all = LinearRegression()
# Fit the model to X_train and y_train
# Make predictions

Now visualize the discrepancy of predictions vs. actual prices using scatterplot and histogram

import matplotlib.pyplot as plt
scatterplot linear regression diamonds dataset
Scatterplot linear regression diamonds
import seaborn as sns
distplot linear regression diamonds dataset
Distplot linear regression Diamonds dataset

Note that this is a comparison between logarithm of actual prices from the diamond price prediction dataset, and the predictions from our linear regression ML model → See the code again. It is written as plt.scatter(y_test,y_pred) after all features were converted to logarithm using convertfeatures2log()

This means to see actual discrepancies, we should un-logarithm it i.e. find the exponent of every price and prediction and then plot. The following code does this:

# convert prices and predictions back to exp
y_pred2 = np.exp(y_pred)
y_test2 = np.exp(y_test)

Now see the scatterplot and histogram again:

distplot linear regression diamonds dataset

How to select the best machine learning algorithm for a specific problem? Take the following course today to see several ML Algorithms used in a variety of problems. The best part is that you get a FREE 30 day trial by clicking on the image below.

linkedin appliedml

K-nearest neighbors (KNN)

Because so many API libraries exist for several machine learning algorithms, the code for all simple machine learning algorithms is straightforward.

On the lines of linear regression code, we use sklearn library for to predict diamond price using KNN or K-nearest neighbors.

from sklearn.neighbors import KNeighborsRegressor
reg_all = KNeighborsRegressor(n_neighbors = 8, metric = ‘minkowski’, p = 2)

The distplot and scatterplot for logarithmic features are as follows. You can confirm from values in x-axis and y-axis that y_test and y_pred are log here.

distplot KNN log diamonds dataset
distplot KNN log diamonds dataset

The distplot and scatterplot for absolute values of features are as follows. You can confirm from values in x-axis and y-axis that y_test and y_pred are NOT log here.

distplot KNN diamonds dataset
distplot KNN diamonds dataset

Support vector machines (SVM)

Let’s now look at diamond price prediction using SVM in Python.

  1. Import LinearRegression class from Sci-kit learn
  2. Create an object of LinearRegression model
  3. Fit the model to X_train and y_train
  4. Make predictions
from sklearn.svm import SVR
regressor = SVR(kernel=’rbf’)
y_pred = regressor.predict(X_test)

Below is a comparison of scatterplots of log values of features (left plot) and absolute values (right plot) of features.

scatterplot SVM diamond kaggle dataset
scatterplot SVM – Log & No log

As of now, we can deduce that diamond price prediction using SVM is a better option because it gives a better metrics score and a better scatterplot than Linear Regression and KNN

In the following course from Udemy, you will learn how to apply SVMs to practical applications: image recognition, spam detection, medical diagnosis, and regression analysis. 30-Day money back guarantee if you do not like.

udemy lazyprog1

Regression Evaluation Metrics

Here are three common evaluation metrics for any regression problem including our diamond price prediction using machine learning problem:

mean absolute error formula machine learning regression
Mean Absolute Error formula
Mean squared Error formula regression machine learning
Mean Squared Error formula
root mean squared error formula regression machine learning
Root Mean Squared Error formula

The code to call functions related to all 3 evaluation metrics is mentioned below.

from sklearn import metricsprint(‘MAE:’, metrics.mean_absolute_error(y_test, y_pred))
print(‘MSE:’, metrics.mean_squared_error(y_test, y_pred))
print(‘RMSE:’, np.sqrt(metrics.mean_squared_error(y_test, y_pred)))
# Output for SVM is :
MAE: 0.08815796872409032
MSE: 0.012811091991743056
RMSE: 0.11318609451581522

Random Forest

Random forest is one of the most popular algorithms in most use cases / projects across industries. Its fast, easier to implement, needs lesser data, doesnt require extensive training and produces almost equally good results.

We will now use the sklearn code and kaggle dataset for Diamond price prediction using python with Random Forest machine learning algorithm.

from sklearn.ensemble import RandomForestRegressor
rf = RandomForestRegressor(n_estimators = 10)
y_pred = rf.predict(X_test)

Now the metrics and their outputs

MAE: 0.08413000449351056
MSE: 0.012880789432555585
RMSE: 0.1134935655997977

It turns out that Random Forest is similar to the far slower SVM for diamond price prediction.

Diamond price prediction using neural networks

neural networks to predict diamond price

Generally, neural networks or ANNs are more suited for classification problems requiring lots of complex logic / decision making and huge computations. They require larger datasets for their optimization to get the benefit of generalization and nonlinear mapping. But, if there’s not enough data, a plain regression model may be better suited despite a few nonlinearities.

You can read more about whether Neural networks are really needed for regression problems or not by reading Neural Networks for Regression. I would any day suggest you also subscribe to a very popular course on AI foundations and Neural networks by clicking the LinkedIn learning image below. You will get a 30-day FREE trial as well.

linkedin neural networks

However, just for sake of completeness, I will show you the steps needed for diamond price prediction using Neural networks.

And for ANN, the best, fastest and easiest to use and code library is Keras. Read more about Keras at their official website:

keras for neural networks

How to create an Artificial neural network in python, Keras & kerasregressor

The step-wise process to create a neural network to predict diamond prices is:

  1. Construct a baseline neural network in keras
  2. Compile and return a Keras model using KerasRegressor,
  3. Use the K-fold function to get perfect results
  4. Fit the estimator to diamond price prediction dataset (training data)
  5. Finally, predict the output (the diamond price predicted using neural networks)
  6. Evaluate the model using metrics between y-predicted vs. y-test

Now we go through each step in detail to create a neural network in keras.

Firstly import the libraries for our project:

from keras.models import Sequential
from keras.layers import Dense
from keras.optimizers import RMSprop
from keras.wrappers.scikit_learn import KerasRegressor
from sklearn.model_selection import cross_val_score
from sklearn.model_selection import KFold

Next, we construct a baseline_model() function to create and return a Keras model

# define base model
def baseline_model():
    # create model
    model = Sequential()
    # add 1st layer
    model.add(Dense(output_dim=18, input_dim=11, activation='relu')) # kernel_initializer='normal',
    # add hidden layer
    model.add(Dense(output_dim=12, kernel_initializer='normal', activation='relu'))
    # add output layer
    model.add(Dense(1, kernel_initializer='normal'))
    # Compile model
    model.compile(loss='mean_squared_error', optimizer='adam')
    return model

After designing the baseline neural network in keras, we run the KerasRegressor() function.

  • It returns the baseline Artificial neural network model built in Keras. It takes as input the model, epochs and batch-size.
  • An epoch defines the number of times the learning algorithm will work through the entire training dataset to update the weights for a neural network. An epoch that has one batch is called the batch gradient descent learning algorithm. For batch training, all the training samples pass through the learning algorithm simultaneously in one epoch before weights are updated.
  • The batch size is a number of samples processed before the model is updated.

More information on KerasRegressor can be found at Tensorflow website:

tensorflow for machine learning using neural networks
estimator = KerasRegressor(build_fn=baseline_model, epochs=10, batch_size=5)
kf = KFold(n_splits=5)
results = cross_val_score(estimator, X_train, y_train, cv=kf)
print(“Results: %.2f (%.2f) MSE” % (results.mean(), results.std()))

A fantastic highly rated Udemy course on Neural networks. Click for 30-day money back guarantee

udemy holczer balazs

Using the neural network built in keras, we run the KFold() function.

K-Fold Cross Validation uses a given data set, splits it into a K number of folds where each fold is used as a testing set while other K-1 are used as training set.

  • For example, for 10-Fold cross validation (K=10), the dataset is split into 10 folds.
  • In the first iteration, the first fold is used for validation while the 9 remaining folds form the training set.
  • In the second iteration, 2nd fold is used as the testing set while the rest serve as the training set.
  • This process is repeated until each fold of the 10 folds have been used as testing sets.

Subsequently, the cross_val_score function takes the model, X and y, and kfold’s result as inputs and outputs multiple results — a list of regression model metrics scores. The cross_val_score function splits the data, using KFold as described above, into K pieces, trains on each combination of K-1 folds and gives back the metrics of the model.

Brief information and steps on KFold and cross_val_score can be found in my other post; link below.


Finally, we fit the estimator to our training data to get predictions from our mini-project on diamond price prediction using neural networks.

estimator.fit(X_train, y_train)
y_pred = estimator.predict(X_test)# Plot a scatter plot like above to see prediction perfection 

The scatterplot output for log(features) is as below. It turns out that it is not as bad as we expected when I said that ANNs are mostly used for classification, not regression. You can yourself decide if you would like to invest resources in ANN deployment simply for diamond price prediction using machine learning.

scatterplot for diamond price prediction using neural networks
ANN scatterplot for log(features)


For this particular post, I have referred to several websites including Beyond4Cs, Machine Learning Mastery and others few posts and websites on internet.

The following course was my first ever course on Udemy and what a fantastic learning I must say. I can any day suggest you to immediately subscribe to this course since I believe it is probably the best course you should take up after Andrew Ng’s course. Notice the number of people who reviewed it.

Click on the following image to directly go to Udemy’s course. It comes with a 30-day money back guarantee – I have myself availed it for 2 other courses 🙂

udemy kirill1

The same post was first written on Medium.com. Please click here for part I and here for part II of my initial two posts on medium.

Thanks for reading this post. I have written several other such Machine Learning related articles on this website. If you liked this post, kindly comment and like using the comment form below.

Frequently Asked Questions (FAQ)

What are different types of Machine Learning?

This is one of the very first questions by readers of any guide on Machine Learning. Generally speaking,
There are 3 types of machine learning:
1. Supervised Learning, where an ML model makes predictions based on known or tagged or labeled data (data with tags or labels, thereby seemingly more meaningful) e.g. our diamond price prediction in python
2. Unsupervised Learning, where there’s no labeled data. An ML model here identifies patterns, relationships and anomalies to predict the outcome.
3. Reinforcement Learning, where the ML model for prediction learns continually based on the output its predicts and the rewards it receives for its previous actions.

What is the process followed to create Machine learning models for prediction?

One of the first steps in Machine Learning process is to understand the business requirements and then identify the ML model. Then, a very simple 3-step machine learning basic process is followed to create ML models for prediction:
1. Train the model: Split the entire data to be used to predict diamond price into train and test data using train-test-split, or any other method. The train data is run on the agreed ML model for prediction. The model is tweaked regularly by editing model parameters unless accuracy, precision and recall and other tests match your expectations.
2. Test the model: Once the output is up to satisfactory levels, you use the test data to check if the model predicts with agreed accuracy. This would help determine if training is effective. On errors, change your ML model for prediction or retrain it with more data.
3. Deploy the model: Finally, upon several rounds of testing and training, once the model reaches your levels of expectations, you deploy the model into production.

What is bias and variance?

Bias is the distance of predicted values (y-hat) from the actual values (y). Bias is High when the average predicted values are far from the actual values.
Variance is high when the model performs good on the training set but not on test set. Variance shows how scattered are the predicted values from actual values.

What can I get from a confusion matrix?

A Confusion matrix is used for evaluating the performance not for ML models for prediction but for classification. It is an 2 x 2 matrix which compares the actual values with those predicted by the machine learning model. It is the first step in machine learning basics on performance evaluation. Furthermore, it helps tell you how well the ML classification model is performing and the errors it is making.
True Positive (TP) : The predicted value matches the actual value. E.g. actual value = positive, predicted value = positive
True Negative (TN): The predicted value matches the actual value. E.g. actual value = negative, predicted value = negative
False Positive (FP), also called Type 1 error: Wrong prediction. E.g. actual value = negative , predicted value = positive
False Negative(FP), also called Type 2 error: Wrong prediction. E.g. actual value = positive, predicted value = negative 

What is Semi-supervised Machine Learning?

Supervised learning uses labeled data while unsupervised learning uses no training data.
The third type, semi-supervised learning uses a small amount of labeled data and a large amount of unlabeled data.

Precision and Recall formula in 10 seconds!

Another important item falling under Machine Learning basics.
Precision = (True Positive) / (True Positive + False Positive)
Recall = (True Positive) / (True Positive + False Negative)

Example of Precision & recall using one-class classification:
Precision means how many times you can “correctly recollect” the date when your girlfriend gave you the watch you are wearing.
Recall means how many times you can “recollect” that the watch was actually gifted by her, not by your Dad.

What’s k-fold cross validation?

K-fold cross validation is a procedure used to estimate the skill of the model on unseen data.
K = number of groups your data is split into.
K-fold cross validation procedure:
1. Shuffle the dataset randomly by splitting the dataset into k groups (10 groups)
2. For each unique group: (for i = 1 to 10)
a) Take the group as test data set (train_test_split = 10% test & 90% train) + remaining groups as a training data set
b) Fit a model on the training set and evaluate it on this test set
c) Retain the evaluation score and discard the model
Summarize the skill of the model using the sample of model evaluation scores

Leave a Comment

Five Ste🅿 Guide
Follow me