https://www.datacamp.com/community/podcast/kaggle-future-data-science

<p>Anthony Goldbloom, CEO of Kaggle, speaks with Hugo about Kaggle, data science communities, reproducible data science, machine learning competitions and the future of data science in the cloud.</p>

Anthony Goldbloom, CEO of Kaggle, speaks with Hugo about Kaggle, data science communities, reproducible data science, machine learning competitions and the future of data science in the cloud.

Kaggle and the Future of Data Science

https://www.datacamp.com/community/news/my-first-kaggle-contest-th9roqvvaz9

My First Kaggle Contest

https://www.datacamp.com/community/news/conducting-an-internal-data-science-contest-at-appnexus-tsbj5q96dwa

The AppNexus Data Science team is always looking for ways to spread data science knowledge across AppNexus. We’re not the first company to recognize that scaling Machine Learning (ML) means improving…

Conducting an Internal Data Science Contest at AppNexus

https://www.datacamp.com/community/news/is-data-science-the-right-fit-for-you-khd6m0xdp5d

EDA of the Kaggle Survey from 2017.  See how you measure up!

Is Data Science the right fit for you?

https://www.datacamp.com/community/news/my-first-kaggle-contest-and-r-package-48kmh6hkl1m

Within an R project, leadr maintains a personal leaderboard for every model built. The package also handles model saving and organization. By the time I unit tested, documented, and built a pkgdown website leadr cost me days in the competition and my score no doubt suffered. Totally worth it.

My First Kaggle Contest and R Package

<p>We&#39;ve had an amazing turn out for the second FB live-coding event! Thanks to all that joined Hugo Bowne-Anderson in December in submitting several submissions to Kaggle&#39;s infamous <a href="https://www.kaggle.com/c/titanic">Titanic Machine Learning Competition!</a>. For those that missed it, here is a recap to find all the useful links and notebooks to take you from zero to one with machine learning in Python.</p>
<p>This live code-along session covers how to build an algorithm that predicts whether any given passenger on the Titanic survived or not, given data on them such as the fare they paid, where they embarked and their age. Hugo shows you how to do so using the Python programming language, Jupyter notebooks and state-of-the-art packages such as <code>pandas</code>, <code>scikit-learn</code> and <code>seaborn</code>. Dive into this rich dataset and build your chops in exploratory data analysis, data munging and cleaning, and machine learning. No previous experience with machine learning necessary, just a will to learn!</p>
<p>This code along session is meant for beginners and intermediates alike. Some programming fundamentals and Python basics will help though (e.g., variables, for loops). Hugo uses Jupyter Notebooks and the terminal. However, if you’re not super familiar with these tools, never fear! We&#39;ll link all the resources you need to complete this project.</p>
<p>You&#39;ll learn how to put all of these tools together to produce informative figures such as this:</p>
<p><a href="https://github.com/datacamp/datacamp_facebook_live_titanic"><img src="http://res.cloudinary.com/dyd911kmh/image/upload/q_auto:best/v1515431047/output_65_0_knd6e9_dsokhg.png" alt=""></a></p>
<p>Feel free to code-along or just watch Hugo do his thing! For those that do wish to code please follow the <a href="https://github.com/datacamp/datacamp_facebook_live_titanic">instructions to set up</a> before starting the video. You’ll need, amongst other things, to clone <a href="https://github.com/datacamp/datacamp_facebook_live_titanic">the repo</a> and download the <a href="https://www.anaconda.com/download/">Anaconda Distribution</a> for Python 3.6 if you haven’t already.</p>
<p>Now you are ready to watch Hugo and follow along. The video lasts about 1 hour and 40 minutes. You can skip to the 10th minute mark where Hugo comes on and the session actually starts.</p>
<p><center><iframe src="https://www.facebook.com/plugins/video.php?href=https%3A%2F%2Fwww.facebook.com%2F726282547396228%2Fvideos%2F1889371314420673%2F&width=560&show_text=false&height=315&appId" width="560" height="520" style="border:none;overflow:hidden" scrolling="no" frameborder="0" allowTransparency="true" allowFullScreen="true"></iframe></center></p>
<p>You can find 3 detailed tutorials showing the various submissions with the solutions here:</p>
<ul>
<li><a href="https://www.datacamp.com/community/tutorials/kaggle-machine-learning-eda">EDA &amp; Machine Learning</a></li>
<li><a href="https://www.datacamp.com/community/tutorials/kaggle-tutorial-machine-learning">Your First Machine Learning Model</a></li>
<li>3rd one coming January 10!</li>
</ul>
<p>Can you improve on Hugo&#39;s submissions? Let us know how high you can rank on Kaggle!</p>
<p>Help us make these sessions better by sharing feedback on Twitter to <a href="http://twitter.com/DataCamp">@DataCamp</a> &amp; <a href="http://twitter.com/hugobowne">@hugobowne</a>.</p>
<p>See you in the next live coding session!</p>


https://www.datacamp.com/community/blog/kaggle-live-coding-recap

Recap post outlining all the resources from the second live-coding session!

Live-Coding Recap: Kaggle Competition with Machine Learning

<h2>Getting Started with Python</h2><p>In this chapter we will go trough the essential steps that you will need to take before beginning to build predictive models.</p><h2>Predicting with Decision Trees</h2><p>After making your first predictions in the previous chapter, it's time to bring you to the next level. In chapter 2 you</p><h2>Improving your predictions through Random Forests</h2><p>What techniques can you use to improve your predictions even more? One possible way is by making use of the machine learning method Random Forest. Namely, a forest is just a collection of trees...</p>

https://www.datacamp.com/community/open-courses/kaggle-python-tutorial-on-machine-learning

Always wanted to compete in a Kaggle competition but not sure you have the right skill set? In this course, you'll learn to predict survival rate for Kaggle's Titanic competition.

Kaggle Python Tutorial on Machine Learning

<h2>Analyzing the 2013 American Community Survey</h2><p>In this first chapter you will use data from the <a href="https://www.kaggle.com/census/2013-american-community-survey" target="_blank" >2013 American Community Survey</a> to figure out whether it makes sense to pursue a PhD or not. The end result of this chapter will be <a href="https://www.kaggle.com/martijn/d/census/2013-american-community-survey/data-exploration-with-kaggle" target="_blank">your own Kaggle script</a> that you can add to your Kaggle account.</p><h2>Pigeons, Chopsticks, Spanish Silver and Marijuana</h2><p>Analyze pigeon racing data, discover the optimal size for your chopsticks, plot Spanish silver production in the 18th century and map the average price of marijuana in the different states of the US.</p>

https://www.datacamp.com/community/open-courses/data-exploration-with-kaggle-scripts

In this course, you'll begin learning the art and science of data exploration. You'll also become familiar with some of the most widely used visualization packages in data science.

Data Exploration With Kaggle Scripts

<p><strong>Table of Contents</strong></p>
<ul>
<li><a href="#importing-the-data">Importing the Data</a></li>
<li><a href="#preprocessing">Preprocessing</a></li>
<li><a href="#training">Training</a></li>
<li><a href="#getting-the-validation-set-predictions">Getting the Validation Set Predictions</a></li>
<li><a href="#getting-the-test-set-predictions">Getting the Test Set Predictions</a></li>
<li><a href="generating-the-output-dataframe">Generating the Output Dataframe</a></li>
<li><a href="#generating-the-output-file">Generating the Output File</a></li>
</ul>
<p>In this tutorial, we will go through the Titanic dataset, analyze it, and submit the results to the actual competition. You have already learned how to fetch the data and import it in your Kaggle Notebook in <a target="_blank" href="https://www.datacamp.com/community/tutorials/tutorial-kaggle-datasets-tutorials-kaggle-notebooks">this tutorial</a>. Now we will be using DataCamp Workspace as the data science notebook. You will be learning how to use both tools on different platforms so that you can choose your favorite.</p>
<p>First, refer to the "Competitions" section in Kaggle and search for Titanic (Figure 5.4). Before going directly into the notebook, you should understand all the information shared in the "Description and Rules" sections. It is important to also check other users' notebooks, as this could give you more ideas about the solution.</p>
<center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/v1646748483/image6_irvam2.png"></center>
<center>Figure 5.4: Titanic - Machine Learning from Disaster</center>

<p>The competition is about using machine learning to create a model that predicts which passengers would have survived the Titanic shipwreck. We will be using a dataset that includes passenger information like name, gender, age, etc. There will be 2 different datasets that we will be using. The first one is titled &#39;train.csv&#39;. In this dataset, besides the passenger information, there are also the ground truth values such as &#39;survived&#39; and &#39;not survived&#39;. In the second dataset which is titled &#39;test.csv&#39;, there will not be any <a target="_blank" href="https://towardsdatascience.com/in-ai-the-objective-is-subjective-4614795d179b">ground truth</a> values, since we are asked to predict those and submit our results.</p>
<p>The submission file will have only 2 columns: PassengerId and Survived. The submission system will produce errors if you try to upload a file with more than 2 values. Please note that you would be able to submit 10 submissions a day.</p>
<p>With the information provided in the "Competition Description" section, the next step would be to fetch the data to our DataCamp Workspace and analyze the data.</p>
<p>Firstly, you need to download the dataset in Data Explorer by going into the Data tab (Figure 5.5). You should be downloading both &#39;train.csv&#39; and &#39;test.csv&#39;.</p>
<p>After downloading the files, you should be navigating to <a target="_blank" href="https://www.datacamp.com/workspace/dashboard">DataCamp Workspace</a>. Please note that you need to create an account first. Creating an account and using the workspace is free.</p>
<p>You can easily drag and drop files on the left side of the notebook (Figure 5.6).</p>
<br>
<center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/v1646748482/image2_v8jxnm.png"></center>
<br>
<center>Figure 5.5: Downloading the Dataset</center>
<br>
<center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/v1646748482/image5_vv8uem.png"></center>
<br>
<center>Figure 5.6: Adding the Dataset to DataCamp Workspace</center>
<br>
<p>The next step is to check the data dictionary in the dataset description section. This will help us to decide which columns to visualize.</p>
<ul>
<li>survival: Survival</li>
<li>pclass: Ticket class</li>
<li>sex: Sex</li>
<li>Age: Age in years</li>
<li>sibsp: # of siblings / spouses aboard the Titanic</li>
<li>parch: # of parents / children aboard the Titanic</li>
<li>ticket: Ticket number</li>
<li>fare: Passenger fare</li>
<li>cabin: Cabin number</li>
<li>embarked: Port of embarkation</li>
</ul><div style="padding-bottom:20px;" id="importing-the-data"></div>
<h2>Importing the Data</h2>
<p>You can use bash commands in jupyter notebooks by placing an exclamation mark in front of the cell. We will be checking the location of the uploaded dataset files using pwd and ls commands.</p>
<pre><code class="lang-python">!pwd; ls
</code></pre><p>/work/files/workspace</p>
<p>notebook.ipynb    test.csv  train.csv</p>
<p>Pandas is a Python library and it is used for data manipulation and data analysis processes. We have now removed <strong>Ticket, Name,</strong> and <strong>PassengerId</strong> columns since they do not have any meaningful numerical or categorical data. Now you will be able to see the rest of the features to be used.</p>
<p>In the next couple of lines, the training and the test dataset files are read using the <strong>read_csv</strong> method of Pandas. Note that <strong>Survived</strong> and <strong>PassengerId</strong> columns are added respectively to train and test dataframes. We will be feeding the model separately using the features in <strong>columns_to_be_added_as_features</strong> and the labels in the <strong>Survived</strong> column. But in order to make the preprocessing simpler, we can combine them for now. The <strong>test_df_matcher</strong> variable includes <strong>PassengerId</strong> as well. We will be using it to match the PassengerId with the test set predictions, to generate the CSV file which will be uploaded to Kaggle as our result.</p>
<pre><code class="lang-python">import pandas as pd
columns_to_be_added_as_features = [&#39;Sex&#39;,&#39;Age&#39;,&#39;SibSp&#39;,&#39;Parch&#39;,&#39;Pclass&#39;,&#39;Fare&#39;,&#39;Embarked&#39;]
train_df = pd.read_csv(&#39;train.csv&#39;, usecols=columns_to_be_added_as_features + [&#39;Survived&#39;])
test_df_matcher = pd.read_csv(&#39;test.csv&#39;, usecols=columns_to_be_added_as_features + [&#39;PassengerId&#39;])
test_df = test_df[columns_to_be_added_as_features]
</code></pre><p>With the <strong>head()</strong> method, we can see the first 5 rows alongside their column names. You can define the amount of the initial rows to be shown as the parameter inside the <strong>head()</strong> method. The number of rows' values in the training set and in the test set are printed respectively in the following cell.</p>
<pre><code class="lang-python">print(train_df.head())
</code></pre><center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/v1646748482/image1_azdsq9.png"></center>
<center>Figure 5.7: Dataframe</center>

<pre><code class="lang-python">print(&quot;Number of rows in training set: {}&quot;.format(len(train_df)))
print(&quot;Number of rows in test set: {}&quot;.format(len(test_df)))
</code></pre><p>Number of rows in training set: 891
Number of rows in test set: 418</p>
<h2 id="preprocessing">Preprocessing</h2>
<p>We should convert the type of the values to <strong>float</strong>, which will be used as features in the model. This conversion will allow us to perform some numerical operations when normalizing the data.</p>
<pre><code class="lang-python">for column_title in columns_to_be_added_as_features:
    if column_title in [&#39;Embarked&#39;, &quot;Sex&quot;]:
        continue
    train_df[column_title] = pd.to_numeric(train_df[column_title], downcast=&quot;float&quot;)
    test_df[column_title] = pd.to_numeric(test_df[column_title], downcast=&quot;float&quot;)

train_df[&quot;Survived&quot;] = pd.to_numeric(train_df[&quot;Survived&quot;], downcast=&quot;float&quot;)
</code></pre><p>We need to feed the model using numerical values. In the code piece below the strings are converted into numerical values</p>
<pre><code class="lang-python">train_df[&#39;Embarked&#39;].replace(&#39;Q&#39;, 0,inplace=True)
train_df[&#39;Embarked&#39;].replace(&#39;S&#39;, 1,inplace=True)
train_df[&#39;Embarked&#39;].replace(&#39;C&#39;, 2,inplace=True)

test_df[&#39;Embarked&#39;].replace(&#39;Q&#39;, 0,inplace=True)
test_df[&#39;Embarked&#39;].replace(&#39;S&#39;, 1,inplace=True)
test_df[&#39;Embarked&#39;].replace(&#39;C&#39;, 2,inplace=True)

train_df[&#39;Sex&#39;].replace(&#39;male&#39;, 0,inplace=True)
train_df[&#39;Sex&#39;].replace(&#39;female&#39;, 1,inplace=True)

test_df[&#39;Sex&#39;].replace(&#39;male&#39;, 0,inplace=True)
test_df[&#39;Sex&#39;].replace(&#39;female&#39;, 1,inplace=True)

print(train_df.head())
</code></pre><center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/v1646748482/image3_glx4ay.png"></center>
<center>Figure 5.8: Dataframe with Numerical Values</center>

<p>Below, all the features are normalized, which means scaled between 0 and 1 in this case.</p>
<pre><code class="lang-python">#Credit: Michael Aquilina,
#https://stackoverflow.com/questions/26414913/normalize-columns-of-pandas-data-frame
def normalize(df):
    result = df.copy()
    for feature_name in df.columns:
        max_value = df[feature_name].max()
        min_value = df[feature_name].min()
        result[feature_name] = (df[feature_name] - min_value) / (max_value - min_value)
    return result

train_df = normalize(train_df)
test_df = normalize(test_df)

print(train_df.head())
</code></pre><center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/v1646748482/image4_iebh1v.png"></center>
<center>Figure 5.9: Normalized Dataframe</center>

<p>Since the dataset will be split into training and validation, we will be shuffling the dataset first to avoid groupings in only one set.</p>
<pre><code class="lang-python">#Credit: Kris, https://stackoverflow.com/questions/29576430/shuffle-dataframe-rows
train_df = train_df.sample(frac=1).reset_index(drop=True)
</code></pre><p>All the empty values will be replaced with 0; this choice has been made for time-saving purposes. However, you can read <a target="_blank" href="https://towardsdatascience.com/7-ways-to-handle-missing-values-in-machine-learning-1a6326adf79e">this tutorial</a> to learn how to handle missing values in machine learning.</p>
<pre><code class="lang-python">train_df = train_df.fillna(0)
test_df = test_df.fillna(0)
</code></pre><p>Below, the validation's ratio was set to 0.2, which means 20% of data will be used to validate it. To split the dataset into training and validation sets, we will be using Sklearn&#39;s <strong>train_test_split</strong> method. Then we will be splitting the features and labels as shown in the last 4 lines of the code piece below.</p>
<pre><code class="lang-python">validation_set_ratio = 0.2
validation_set_size = int(len(train_df)*validation_set_ratio)
training_set_size = len(train_df) - validation_set_size

print(&quot;Total set size: {}&quot;.format(len(train_df)))
print(&quot;Training set size: {}&quot;.format(training_set_size))
print(&quot;Validation set size: {}&quot;.format(validation_set_size))

from sklearn.model_selection import train_test_split

train, val = train_test_split(train_df, test_size=validation_set_ratio)

train_X = train[columns_to_be_added_as_features]
train_Label = train[[&#39;Survived&#39;]]

val_X = val[columns_to_be_added_as_features]
val_Label = val[[&#39;Survived&#39;]]
</code></pre><h2 id="training">Training</h2>
<p>Support vector machines are supervised learning models that are used for classification and regression analysis. This model was chosen with the parameters below. You should be trying other models with different parameter combinations. There is no need to start and stop the code every time you want to modify a parameter or to try another model, as you can automate this process. If you are interested in Support Vector Machines, you can view the <a target="_blank" href="https://www.datacamp.com/community/tutorials/svm-classification-scikit-learn-python">Support Vector Machines with Scikit-learn</a> tutorial on DataCamp.</p>
<pre><code class="lang-python">from sklearn.svm import SVC

SVM_KERNEL = &quot;linear&quot;
SVM_C = 10
SVM_GAMMA = 0.00001

svm_model = SVC(kernel = SVM_KERNEL, C = SVM_C, gamma = SVM_GAMMA)
svm_model.fit(train_X, train_Label)
</code></pre><h2 id="getting-the-validation-set-predictions">Getting the Validation Set Predictions</h2>
<p>In this part, we will be feeding the model validation set features and getting the predictions. Then we will be comparing them with the ground truth values. The Sklearn&#39;s <strong>metrics.accuracy_score</strong> method was used to make the accuracy calculation easier. The validation set accuracy is calculated as 0.799, which is a good start.</p>
<pre><code class="lang-python">from sklearn import metrics

y_pred = svm_model.predict(val_X)
print(&quot;Accuracy:&quot;,metrics.accuracy_score(val_Label, y_pred))
</code></pre><p>Accuracy: 0.7988826815642458</p>
<h2 id="getting-the-test-set-predictions">Getting the Test Set Predictions</h2>
<p>Now we proceed to get the predictions for the actual test set. Because there are no ground truth values for the test set, we can only know the accuracy after the predictions file is uploaded to Kaggle.</p>
<pre><code class="lang-python">test_pred = svm_model.predict(test_df)
</code></pre><h2 id="generating-the-output-dataframe">Generating the Output Dataframe</h2>
<p>The PassengerId column is fetched from the dataframe named <strong>test_df_matcher</strong>. When it is directly returned, the type of the return value will be <strong>Series</strong>. For this reason, we will be  casting it to Pandas DataFrame type. Then we will be inserting the test set predictions as the second column of the dataframe.</p>
<pre><code class="lang-python">result = pd.DataFrame(test_df_matcher[&#39;PassengerId&#39;])
print(result.head(10))
</code></pre><center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/v1646748483/image9_l38s1p.png"></center>
<center>Figure 5.10: PassengerId Dataframe</center>

<pre><code class="lang-python">result.insert(1, &quot;Survived&quot;, test_pred, True)
result[&quot;Survived&quot;] = pd.to_numeric(result[&quot;Survived&quot;], downcast=&quot;integer&quot;)
print(result.head(10))
</code></pre><center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/v1646748483/image8_talxf3.png"></center>
<center>Figure 5.11: Finalized Dataframe</center>

<h2 id="generating-the-output-file">Generating the Output File</h2>
<p>Finally, the output format is ready. As the last step, the index values will be removed by passing <strong>False</strong> to the <strong>index</strong> parameter. Now you can download the generated file and upload it to Kaggle. This way, you will learn the test set accuracy, after which you will be placed on the leaderboard.</p>
<pre><code class="lang-python">result.to_csv(&quot;out.csv&quot;, index=False)
</code></pre><p>We will now return to the  competition page and click on "Submit Predictions" and then upload the predictions file. After a few seconds, you will get your test accuracy (Figure 5.12).</p>
<center><img src="https://res.cloudinary.com/dyd911kmh/image/upload/v1646748483/image7_rdjv5h.png"></center>
<center>Figure 5.12: Accuracy of the Test Set</center>

<h2 id="conclusion">Conclusion</h2>
<p>In this tutorial, you have learned how to analyze a dataset and submit your results to a Kaggle competition.  If you enjoy doing data science competitions you can  learn more about DataCamp Competitions <a target="_blank" href="https://www.datacamp.com/data-science-competitions">here</a>.    </p>


https://www.datacamp.com/community/tutorials/tutorial-kaggle-competition-tutorial-machine-learning-from-titanic

Prepare for your first Kaggle Competition with this step-by-step tutorial.  

Learn how to analyze the Titanic dataset in Kaggle and submit the results to the competition.  

Kaggle Competition Tutorial: Machine Learning from the Titanic

<nav><h2 id="table-of-contents">Table of Contents</h2>
<ul>
<li><a href="#kaggle-notebooks">Kaggle Notebooks</a></li>
<li><a href="#kaggle-datasets">Kaggle Datasets</a>
<ul>
<li><a href="#what-are-kaggle-datasets-">What Are Kaggle Datasets?</a></li>
<li><a href="#dataset-search-filters">Dataset Search Filters</a></li>
<li><a href="#data-explorer">Data Explorer</a></li>
<li><a href="#datasets-for-beginners">Datasets For Beginners</a></li>
<li><a href="#custom-datasets">Custom Datasets</a></li>
</ul>
</li>
</ul>
</nav>
<h2 id="kaggle-notebooks">KAGGLE NOTEBOOKS</h2>
<p>A "Kaggle Notebook" is a free jupyter notebook server that can be GPU integrated. Just like DataCamp Workspace notebooks, it allows you to perform machine learning operations on cloud computers instead of doing it on your own computer. Each time you create a Kaggle Notebook, you can edit and run its content in the browser. There is no need to set up your own jupyter notebook environment, just enter Kaggle, create a notebook and start using it on the browser. You can see the notebooks you have created before on the Kaggle Notebook page and you can also review other people&#39;s notebooks.</p>
<p>To create a Notebook, click on &quot;New Notebook&quot; after you navigate to the Kaggle Code page (Figure 3.1). After this process, cloud resources will be allocated for you and a notebook will be created instantly. You can give your notebook a name by clicking on the notebook text in the upper left corner. As you can see at a first glance, many options and features found in jupyter notebooks are also available here.</p>
<p>The most frequent questions about Kaggle Notebooks are how to share a notebook publicly, how to add another person as a collaborator, how to import a dataset to a notebook, and how to use the GPU. You can see the buttons required to perform each of these operations in Figure 3.2.</p>
<centre><img src="https://res.cloudinary.com/dyd911kmh/image/upload/v1646226116/create_notebook_ycwbc8.png" alt="Figure 3.1: Kaggle Code"></centre>
<center>Figure 3.1: Kaggle Code</center>
<br>
<p>First, you have to commit a version of your Notebook to make it public. You can create a new version of your Notebook by clicking the &quot;Save Version&quot; button at the top right. Then, you can make your notebook accessible to everyone by clicking the &quot;Share&quot; button on the left of this button, or you can add others as collaborators using the same menu. To import a dataset, simply click on the &quot;Add data&quot; button under the &quot;Save Version&quot; button on the right menu, and select the dataset you want to add. To activate the GPU, you need to select the GPU option from the accelerator section in the menu on the right side. The maximum GPU time you can use on Kaggle is set at 30 hours per week.</p>
<centre><img src="https://res.cloudinary.com/dyd911kmh/image/upload/v1646226116/notebook_r620gk.png" alt="Figure 3.2: Kaggle Notebook"></centre>
<center>Figure 3.2: Kaggle Notebook</center>

<p>All other features of notebooks are explained in detail in the <a target="_blank" href="https://www.kaggle.com/docs/notebooks">Kaggle documentation.</a></p>
<h2 id="kaggle-datasets">KAGGLE DATASETS</h2>
<h3 id="what-are-kaggle-datasets-">WHAT ARE  KAGGLE DATASETS?</h3>
<p>Kaggle is a data science platform but it also supports dataset handling. "Kaggle Datasets" allows you to create your own custom datasets, share them with others and easily import them into your notebooks. Additionally, you can add private datasets which would only be visible to you.</p>
<p>What makes this feature one of the most important ones in Kaggle is that it gives you  access to a wide variety of top-quality datasets shared by other users. You can easily find the datasets you want with just a few search and filtering methods.</p>
<h3 id="dataset-search-filters">DATASET SEARCH FILTERS</h3>
<p>To search for a dataset, write your keywords in the search field, as shown in Figure 4.1. Here you can see that we can access several datasets about the pandemic just by typing &quot;Covid&quot; in the search bar.</p>
<p>If you click on &quot;Filters&quot; on the right side of the search bar, more filtering options will appear (Figure 4.2). With these, you can narrow your search by entering dataset tags, file type, and other values like the minimum or maximum size of the dataset (Figure 4.3).</p>
<br>
<centre><img src="https://res.cloudinary.com/dyd911kmh/image/upload/v1646226092/datasets_z4c6lk.png" alt="Figure 4.1: Dataset Search Filters"></centre>
<center>Figure 4.1: Dataset Search Filters</center>
<br>
<p>Kaggle allows you to download any dataset for free, but depending on what you are going to use it for, you may need to pay attention to the license type of the datasets. In some cases, it is possible that you may need to obtain additional permissions from their owners in case you want to use a dataset for an academic paper or in case you intend to use it for commercial purposes, for example.</p>
<centre><img src="https://res.cloudinary.com/dyd911kmh/image/upload/v1646226092/datasets_computer_vision_kcadmg.png" alt="Figure 4.2: Dataset Search Filters by Tags"></centre>
<center>Figure 4.2: Dataset Search Filters by Tags</center>
<br>
<p>There are  three main license types on Kaggle:</p>
<ol>
<li><strong>Creative Commons</strong>: There are several kinds of Creative Commons licenses:</li>
<ol type = "a">
<li>CCO, which stands for public domain and means that the dataset is available to everyone under any circumstances.</li>
<li>CC-BY, which requires the dataset user to credit its owner.</li>
<li>CC-BY-SA, which also requires the owner to be credited and adds the condition that the dataset keeps the same kind of license even after it&#39;s modified.</li>
</ol>
<li>GPL: This license basically provides four main usage options:</li>
<ol type = "a">
<li>Firstly, you get unlimited use of a dataset.</li>
<li>You also have the possibility to examine how the dataset works, and modify it.</li>
<li>Additionally, you are entitled to the unlimited distribution of copies of the dataset.</li>
<li>And lastly, you can distribute the modified version of the dataset as well.</li>
</ol>
<li>Open Database: This allows users to share, modify and use the dataset but it makes it mandatory to establish the same kind of license for the modified dataset.</li>
</ol>
<centre><img src="https://res.cloudinary.com/dyd911kmh/image/upload/v1646226091/datasets_filter_opjfyj.png" alt="Figure 4.3: Dataset Search Filters"></centre>
<center>Figure 4.3: Dataset Search Filters</center>
<br>
<h3 id="data-explorer">DATA EXPLORER</h3>
<p>The Data Explorer section allows you to quickly browse through the content and structure of the datasets. It gives you an overview of the files and the columns in the data, as well as their histogram graphs (Figure 4.4).</p>
<centre><img src="https://res.cloudinary.com/dyd911kmh/image/upload/v1646226091/dataset_explorer_ebgf9u.png" alt="Figure 4.4: Data Explorer"></centre>
<center>Figure 4.4: Data Explorer</center>

<h3 id="datasets-for-beginners">DATASETS FOR BEGINNERS</h3>
<p>The following datasets are fun and easy to play with as a beginner. You can fetch these to your notebooks and start getting your hands dirty by visualizing the data. Once you come up with an idea, you can even build machine learning models with some of these datasets.</p>
<ul>
<li><a target="_blank" href="https://www.kaggle.com/rtatman/digidb">Digimon Database</a>: A database of Digimon and their moves from Digimon Story CyberSleuth.</li>
<li><a target="_blank" href="https://www.kaggle.com/rtatman/animal-bites">Animal Bites</a>: Data on over 9,000 bites, including rabies tests.</li>
<li><a target="_blank" href="https://www.kaggle.com/crawford/80-cereals">80 Cereals</a>: Nutrition data on 80 cereal products.</li>
<li><a target="_blank" href="https://www.kaggle.com/datafiniti/womens-shoes-prices">Women&#39;s Shoe Prices</a>: A list of 10,000 women&#39;s shoes and the prices at which they are sold.</li>
<li><a target="_blank" href="https://www.kaggle.com/groundhogclub/groundhog-day">Groundhog Day Forecasts and Temperatures</a>: How accurate is Punxsutawney Phil&#39;s winter weather forecast?</li>
</ul>
<h3 id="custom-datasets">CUSTOM DATASETS</h3>
<p>You can also upload and use your own datasets in Kaggle. This feature comes in handy when you have your own dataset or when you've modified a dataset and want to use it in your notebook. In order to upload a dataset, first, you need to zip your main dataset file. Then click on &quot;New Dataset&quot; in the Datasets section. Give your dataset a name and upload your zip file (Figure 4.5).</p>
<centre><img src="https://res.cloudinary.com/dyd911kmh/image/upload/v1646226092/dataset_import_qljtq7.png" alt="Figure 4.5: Importing Custom Datasets"></centre>
<center>Figure 4.5: Importing Custom Datasets</center>
<br>
<p>And that&#39;s it. You can now fetch the uploaded dataset to your notebook and start using it, as shown in section 2. If you want to keep the dataset private, make sure that the label in the right bottom corner of the uploading screen reads &quot;Private&quot;. If you don&#39;t want it to be private, you can click on the label and change it to "Public".</p>


https://www.datacamp.com/community/tutorials/tutorial-kaggle-datasets-tutorials-kaggle-notebooks

Learn about Kaggle datasets and notebooks and get a head start on creating your Kaggle profile.

Explore kaggle with this tutorial and get a head start on creating your Kaggle profile. Learn your way around Kaggle Notebooks.

community

datacamp

kaggle

News (5)

My First Kaggle Contest

Is Data Science the right fit for you?

Is Data Science the right fit for you?

Tutorials (5)

Kaggle Competition Tutorial: Machine Learning from the Titanic

Kaggle Datasets Tutorial: Kaggle Notebooks

Cheat Sheets (0)

Open Courses (3)

Kaggle Python Tutorial on Machine Learning

Data Exploration With Kaggle Scripts

Podcast (1)

Kaggle and the Future of Data Science (with Anthony Goldbloom)

Blog Posts (1)

Live-Coding Recap: Kaggle Competition with Machine Learning

Tech Thoughts (0)