<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Manu Ganji on Medium]]></title>
        <description><![CDATA[Stories by Manu Ganji on Medium]]></description>
        <link>https://medium.com/@manuganji?source=rss-543a89825ff3------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*zlqBNHAi0Dw4h3NMeuz__w.jpeg</url>
            <title>Stories by Manu Ganji on Medium</title>
            <link>https://medium.com/@manuganji?source=rss-543a89825ff3------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 22:55:47 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@manuganji/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Freelance Programming When Excellence Is Your Dope]]></title>
            <link>https://medium.com/@manuganji/freelance-programming-when-excellence-is-your-dope-128b2923e28e?source=rss-543a89825ff3------2</link>
            <guid isPermaLink="false">https://medium.com/p/128b2923e28e</guid>
            <category><![CDATA[freelancing]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[learning-to-code]]></category>
            <category><![CDATA[pitching]]></category>
            <category><![CDATA[excellence]]></category>
            <dc:creator><![CDATA[Manu Ganji]]></dc:creator>
            <pubDate>Sun, 14 Jan 2018 19:22:18 GMT</pubDate>
            <atom:updated>2018-01-15T03:25:48.821Z</atom:updated>
            <content:encoded><![CDATA[<p>Is this post for you? To answer that question we need to know what drives you.</p><ul><li>The <strong>I-Type </strong>people are driven by intrinsic rewards like “Doing a good job”, “Learning a new language feature”, and “Having a positive impact on the world”.</li><li>The other kind are the <strong>X-Type</strong> for whom external rewards work well — Fame, Money and Power, etc.</li></ul><p>Nobody is purely one type but most people lean towards one of those types. Make no mistake I-Types do need market pay but money doesn’t take a lot of space in their head. Read the <a href="https://www.goodreads.com/book/show/6452796-drive">book</a> to understand the nuances. If the I-Type sounds like you, then this post is for you. Although I will probably never say no to a million dollars but I’m an I-Type.</p><p>So, Freelancing shouldn’t be that hard for you, you say? Here’s some common advice that I get — “In the initial days accept every project that comes your way”, “Use fake testimonials”, “Price yourself cheap”, “Use a pseudonym”, “Don’t waste time writing tests”, “Code all day”, “Be nice to everyone”, “Don’t turn down any lead”... I have a mental block against doing any of them so none of them were useful.</p><p>Over time I figured out what works for me. Let me break my freelancing cycle into 4 stages.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*b0NOI8SiGBd9Hprjwbj6Rg.png" /><figcaption>My Freelancing Work Cycle — Learn, Pitch, Execute, Reflect and Learn Again</figcaption></figure><p>1. LEARN — Staying Updated With My Craft</p><p>2. PITCH — Landing a project</p><p>3. EXECUTE — Performing the job</p><p>4. REFLECT — Project Postmortem / Ideation</p><h3>Learning</h3><p>Memorizing the syntax isn’t good enough. Time, Budget, and Infrastructure constraints cut down the option pool. I learn to create <em>the best solution under constraints. </em>And I keep getting better at it.</p><p>You have to know when and how to seek help. Learning to Debug and Reading the logs, documentation and source code go a long way. Apart from basic Engineering ideas like Redundancy, Speed and Consistency, learn how to set up your stack for immediate alerts when things go wrong. Take your questions to the most helpful community for that technology.</p><p><strong>Excellence comes from better understanding.</strong></p><p>How do you stay updated about the latest releases/best practices? Newsletters (<a href="http://github.com/explore">Github Explore</a>, <a href="http://pycoders.com/">Pycoders Weekly</a>, etc), following reputed developers on Twitter, and Going to Meetups/Conferences.</p><h3>Landing a Project</h3><p>Trust is hard to come by on the Internet. I don’t enjoy sales calls but I have to sell if I’m freelancing. So what do I do to win their Trust? I think the best way to take an honest consultative approach to <strong>grow their business </strong>because every body wants results from their investment.</p><p>You may get flooded by leads but you can only pick a few projects. Because each project will take a lot of time and attention. I often end up working more than I had originally estimated. The implications?</p><ol><li>I only do one project at a time</li><li>My service won’t be cheap</li></ol><p>The alternative approach doesn’t work for me. Here’s how it goes.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*sqjNRsbPCPAeQL17IPuzyQ.png" /><figcaption>Low Pricing game will leave me frustrated for not doing my best work</figcaption></figure><p>This is the reason marketplaces never work for me. They place unreasonable value on competitive pricing which is a recipe for disaster in my opinion. Some good sources for projects:</p><ol><li>Blogging</li><li>Video</li><li>Answering questions on Stack Overflow</li><li>Friends</li><li>Google Ads</li></ol><h3>Execution</h3><p>Execution is the most favorite part of my work cycle. And I’d really prefer it if I never have to worry about landing my next project and instead be constantly <em>doing</em> great projects.</p><p>I obsess over doing my projects<strong> the best way I can</strong>. Some habits that I have developed so far:</p><ol><li>Understand the Client requirements as well as their Owner. My clients are often surprised when I suggest an improvement.</li><li>Explain changes clearly and resolve differences in-time. Also, introduce any change in small <strong>reversible</strong> chunks. Use Version Control System, Test Driven Development, Modular Design.</li><li>Transparent Design. Some Urban Architects advocate making the drinking water and waste water disposal channels sufficiently visible so that the citizens react well in times of a catastrophe. I adopted this idea to make systems <em>sufficiently</em> transparent so that the users can suggest intelligent improvements.</li><li>Documentation. Slack saved me multiple times when the client forgot his earlier request for a particular application behavior. Use Email and IM (Slack) to archive the reasons for decisions.</li></ol><h3>Leisure and Reflection</h3><p>Archimedes was <strong>not working</strong> when the Eureka! moment happened. Leisure is the most under-appreciated part in the work cycle. We need to set aside some time daily to do absolutely nothing productive or efficient or <em>disruptive</em>.</p><blockquote><em>Don’t do something. Just sit there — </em>Read on <a href="http://headspace.com/">Headspace App</a></blockquote><p>On <em>Average</em>, while on project, I work only 20–30 hours per week (<em>I break this rule often because I just cannot stop myself </em>) to have time to reflect on every design decision. I do do all nighters but only when we’re fire-fighting.</p><p>After the project is done, I go slow for a few days to pause and reflect on how the project went and what could be improved — from client communication to deployment to billing.</p><p>I absorb inspiration from several sources —Women, Books, Art, Kids, Governments... To afford leisure though I must price my service correctly.</p><p>That’s my understanding from programming for the last 6.5 years — 2.5 years of which was freelancing. With this obsession for Excellence I just cannot compete for every project on the market. It’s kind of my <em>positioning</em> in marketing lingo. I only work with great tools, drop many project leads but when I do take up a project it ends up well-made. Hope this post helps in your freelancing journey. Thanks for reading!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=128b2923e28e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Picking Great Technology]]></title>
            <link>https://medium.com/@manuganji/picking-great-technology-e33b93855578?source=rss-543a89825ff3------2</link>
            <guid isPermaLink="false">https://medium.com/p/e33b93855578</guid>
            <category><![CDATA[technology]]></category>
            <category><![CDATA[lindy-effect]]></category>
            <category><![CDATA[nassim-nicholas-taleb]]></category>
            <category><![CDATA[artesanal-team]]></category>
            <category><![CDATA[software]]></category>
            <dc:creator><![CDATA[Manu Ganji]]></dc:creator>
            <pubDate>Sat, 16 Dec 2017 15:49:27 GMT</pubDate>
            <atom:updated>2017-12-16T15:56:02.554Z</atom:updated>
            <content:encoded><![CDATA[<p>IMO, the most important rationale for technology decisions is the <a href="https://en.wikipedia.org/wiki/Lindy_effect">Lindy Effect</a>. I learnt it from <a href="https://medium.com/u/f138bf5466fe">Nassim Nicholas Taleb</a> books:</p><blockquote>If a book has been in print for forty years, I can expect it to be in print for another forty years. But, and that is the main difference, if it survives another decade, then it will be expected to be in print another fifty years. This, simply, as a rule, tells you why things that have been around for a long time are not “aging” like persons, but “aging” in reverse. Every year that passes without extinction doubles the additional life expectancy. This is an indicator of some robustness. The robustness of an item is proportional to its life!</blockquote><p>So Java/PHP are not dying anytime soon folks. :) And that latest software library launched 3 months ago has an expected life of 3 more months. Investing on something with 3 months of expected life is a bad idea. If you take only one thing away from this post make it this one.</p><p>I don’t mean you have to wait 10 years each time you bring in new technology. But considering Lindy Effect will help you cut out a lot of noise.</p><p>If you still want to do it, I’ll give you a few more pointers for your decisions. Picking great technology is similar to a good chef picking ingredients — you have to know the sources and processes.</p><ul><li>A product<strong> from a financially viable company — </strong>For long term results there must be a group financially invested in its success.</li><li>The version number —<strong> at least 1.0</strong> please.</li><li><strong>Bulletproof QA — </strong>preferably automated.</li><li><strong>External interest</strong> (for e.g. Github stars) is a decent signal but it can be gamed.</li><li><strong>Singular focus — </strong>Frequently described as<em> does one thing well</em>. Yeah, I’m highly skeptical of any product that promises world peace.</li><li><strong>Artesanal Team — </strong>I mean professionals who view their job as <strong>craft</strong> and are proud of what they sign their name on. And yeah, a decent number of experienced and married folks to provide adult supervision.</li></ul><p>Also think about the learning curve and integration effort. Integration is not just about the technical requirements. Many tools have second order effects that change your work-flow entirely. For example, adding a live chat widget to your website. Will it actually lead to a higher customer satisfaction if your staff have to answer support requests at mid-night hours?</p><p>Lastly, you must answer why this takes priority over all the other things you can do with that time and resources. Yes even if it takes just 1 day you must still have an answer.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e33b93855578" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[We must question the assumptions in our work place]]></title>
            <link>https://medium.com/@manuganji/we-must-question-the-assumptions-in-our-work-place-d3c82a37627a?source=rss-543a89825ff3------2</link>
            <guid isPermaLink="false">https://medium.com/p/d3c82a37627a</guid>
            <category><![CDATA[organization]]></category>
            <category><![CDATA[work]]></category>
            <category><![CDATA[organizational-culture]]></category>
            <dc:creator><![CDATA[Manu Ganji]]></dc:creator>
            <pubDate>Fri, 17 Jul 2015 16:26:35 GMT</pubDate>
            <atom:updated>2015-07-17T16:26:35.414Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rpE_cxgvOAvDOobJtg3xXA.jpeg" /></figure><p>The underlying assumption for too many procedures in our organizations seems to be that people are inherently bad. That given an option they would plunder the resources on useless ideas, use them for their personal benefit or just simply stay unproductive. Why do we need people to log their hours, get approval for a tool replacement, get their stuff checked by security, get approval to use the company car and so on? Do we assume that they are incapable of stepping beyond their self and thinking about the whole?</p><p>Our organizations are making thousands of guesses about people behavior every day. Every starting guess is that the outcome is negative for the organization without strong oversight. As an employee, how often is your first tendency to cheat? How often is it true that you were going to break ethics and the oversight discouraged you? 1 out of 10 times? 1 out of 100 times? 1 out of 1000?</p><p>Imagine that you begin typing “full” into Google Search and 9 times out of 10, the f-word gets suggested. How frustrating would that be? Scaled to millions of employees making at least 5 decisions per day, organizations are making tens of millions of wrong guesses about people behavior. How many lost hours here? This is consistent low productivity. How did this come to be seen as acceptable use of resources?</p><p>Is it time that we treat people as adults in our work place?</p><p>— — — — —</p><p>This post is inspired by <a href="http://www.reinventingorganizations.com/">Reinventing Organizations</a> by Frederic Laloux and I’m in no way affiliated to the author.</p><p>Image credits: <em>Fingerprint scanner identification” by Rachmaninoff — Own work. Licensed under CC BY-SA 3.0 via Wikimedia Commons — </em><a href="https://commons.wikimedia.org/wiki/File:Fingerprint_scanner_identification.jpg#/media/File:Fingerprint_scanner_identification.jpg"><em>https://commons.wikimedia.org/wiki/File:Fingerprint_scanner_identification.jpg#/media/File:Fingerprint_scanner_identification.jpg</em></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d3c82a37627a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Tools and libraries for A/B testing]]></title>
            <link>https://medium.com/@manuganji/tools-and-libraries-for-a-b-testing-8bc140ce4433?source=rss-543a89825ff3------2</link>
            <guid isPermaLink="false">https://medium.com/p/8bc140ce4433</guid>
            <dc:creator><![CDATA[Manu Ganji]]></dc:creator>
            <pubDate>Thu, 03 Apr 2014 03:07:05 GMT</pubDate>
            <atom:updated>2014-04-03T03:07:05.343Z</atom:updated>
            <content:encoded><![CDATA[<p>We were looking for some good A/B testing tools and libraries for our internal products. These are some good projects we came across useful for production level deployments with commercially friendly licenses.</p><ul><li><a href="https://github.com/seatgeek/sixpack">Six pack</a> by <a href="http://seatgeek.com/">SeatGeek</a>. I’m especially awed by the elegance of their approach. I was cracking my head about how to setup the infrastructure to run so many experiments for my startup. Here they are with such a dead simple approach to all of that big problems. It’s cross platform. Requires very minimal setup and a lot of clients are already available in a language of your choice.</li><li><a href="http://www.bingocardcreator.com/abingo">A/Bingo</a> by <a href="http://www.bingocardcreator.com/">Bingo Card Creator</a>. They also have a nice <a href="http://www.bingocardcreator.com/abingo/compare">comparison</a> with other tools that take a more front end approach to A/B testing. You should definitely check their site even if you don’t use it as the author explains some points about the importance of statistical significance in A/B testing.</li><li><a href="http://vanity.labnotes.org/">Vanity</a> for Rails people</li><li><a href="https://www.optimizely.com/">Optimizely</a> is a commercial alternative. This one also takes a more front end approach leading to ease of configuration but limited in the functionality.</li></ul><p>I’ll follow this up with a short tutorial on how to implement one of these for your project. In a few days. ;)</p><p>There will also be another post on why A/B testing is so important for organizations of all sizes.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8bc140ce4433" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Configuring mysql server on Ubuntu in Azure]]></title>
            <link>https://medium.com/@manuganji/configuring-mysql-server-on-ubuntu-in-azure-7c8caaf0cd3d?source=rss-543a89825ff3------2</link>
            <guid isPermaLink="false">https://medium.com/p/7c8caaf0cd3d</guid>
            <dc:creator><![CDATA[Manu Ganji]]></dc:creator>
            <pubDate>Tue, 17 Dec 2013 18:55:57 GMT</pubDate>
            <atom:updated>2014-06-22T17:45:39.261Z</atom:updated>
            <content:encoded><![CDATA[<h4>With remote access to a specified table</h4><p>I cross posted the same post <a href="http://manuganji.github.io/deployment/config-mysqlon-azure/">here</a>. Because Medium is screwing up the code snippets. For example the single quote character is screwed up. Please contact me there for any questions.</p><p>We recently had to create a mysql server on an Azure VM and give remote access to one database.</p><p>I followed the very good documentation here — <a href="http://www.windowsazure.com/en-us/manage/linux/common-tasks/mysql-on-a-linux-vm/">http://www.windowsazure.com/en-us/manage/linux/common-tasks/mysql-on-a-linux-vm/</a>. The steps I had to follow were connecting via ssh</p><pre>ssh -i myprivatekey.key instanceusername@instancename.cloudapp.net</pre><p>Then, to install mysql server</p><pre>sudo apt-get install mysql-server</pre><p>This automatically creates mysql service which makes sure that mysql is restarted when VM is rebooted</p><p>And then we have to secure this mysql installation. Running this command helps you set the values for common settings by asking you questions and recommending good options</p><pre>mysql_secure_installation</pre><p>After this log into mysql as root</p><pre>mysql -u root -p</pre><p>To create a new MySQL user, run the following at the mysql&gt; prompt</p><pre>CREATE USER ‘mysqluser’@’localhost’ IDENTIFIED BY ‘password’;</pre><p>Create a database and give all privileges to mysqluser</p><pre>CREATE DATABASE testdatabase;<br><br>GRANT ALL ON testdatabase.* TO ‘mysqluser’@’localhost’ IDENTIFIED BY ‘password’;</pre><p>Other options in this command are</p><pre># give permission on <strong>all</strong> databases to <strong>mysqluser</strong> from localhost<br>GRANT ALL ON *.* TO ‘mysqluser’@’localhost’ IDENTIFIED BY ‘password’;<br><br># give permission on testdatabase to <strong>mysqluser</strong> connecting <br># from the particular ip 123.234.132.412 <br>GRANT ALL ON testdatabase.* TO ‘mysqluser’@’123.234.132.412’ IDENTIFIED BY ‘password’;</pre><p>For a complete list of options available you should consult <a href="http://dev.mysql.com/doc/refman/5.5/en/grant.html">MySQL Docs</a></p><p>To permit the user <strong>mysqluser</strong> to connect from any ip,</p><pre># permitting <strong>mysqluser</strong> from <strong>any</strong> ip to access <strong>students</strong> table <br># on testdatabase<br>GRANT ALL ON testdatabase.students TO ‘mysqluser’@’%’ IDENTIFIED BY ‘password’;</pre><p>Notice that we used ‘%’ as a wild card for ip while ‘*’ for database/table names wildcard. For a remote user to connect with the correct privileges you need to have that user created in both the localhost and ‘%’ as in.</p><pre># creating <strong>mysqluser</strong> at locahost to access <strong>students </strong>table <br># on testdatabase<br>GRANT ALL ON testdatabase.students TO ‘mysqluser’@’locahost’ IDENTIFIED BY ‘password’;</pre><p>Exit the mysql shell</p><pre>quit</pre><p>This is not all. We have to add the port where mysql is running (default is 3306) as an endpoint in <a href="http://manage.windowsazure.com/">azure management portal</a>. We have to change the bind-address in mysql config that is located at <strong>/etc/mysql/my.cnf</strong>. As per <a href="http://stackoverflow.com/questions/15663001/remote-connections-mysql-ubuntu/15684341#15684341">this SO answer</a>, you should edit that file and put this line for bind-address</p><pre>bind-address = 0.0.0.0</pre><p>Then mysql needs to be restarted. As per <a href="http://stackoverflow.com/questions/12196873/not-able-to-stop-mysql/12196963#12196963">this answer</a></p><pre>sudo service mysql stop<br>sudo service mysql start</pre><p>You should now be able to connect and access <em>students</em> table in <em>testdatabase</em> as mysqluser from anywhere in the world by providing the following params.</p><pre>host url : instancename.cloudapp.net<br>port no. : 3306 [optional if default]<br>username : mysqluser<br>password : password</pre><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=7c8caaf0cd3d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Installation of Opencv, numpy, scipy inside a virtualenv]]></title>
            <link>https://medium.com/@manuganji/installation-of-opencv-numpy-scipy-inside-a-virtualenv-bf4d82220313?source=rss-543a89825ff3------2</link>
            <guid isPermaLink="false">https://medium.com/p/bf4d82220313</guid>
            <dc:creator><![CDATA[Manu Ganji]]></dc:creator>
            <pubDate>Mon, 16 Dec 2013 13:48:53 GMT</pubDate>
            <atom:updated>2014-06-22T17:46:43.098Z</atom:updated>
            <content:encoded><![CDATA[<h4>Useful for python server side deployments that need computer vision</h4><p>I cross posted the same post <a href="http://manuganji.github.io/deployment/install-opencv-numpy-scipy-virtualenv-ubuntu-server/">here</a>. Because Medium is screwing up the code snippets. For example the single quote character is screwed up. Please contact me there for any questions.</p><p>I did this for a ubuntu 13.04 server. This should work for 12.10 without much of a problem</p><p>Activate the virtualenv</p><pre>source bin/activate <br># or<br>workon &lt;environment_name&gt;</pre><p>For numpy and scipy it’s fairly simple</p><pre>pip install numpy scipy</pre><p>You can go by two approaches. One is to install opencv globally in your system and then moving those libraries to your virtualenv. This approach has some shortcomings. The globally installed opencv might have been compiled for a different version of python. It gets difficult during deployment to point to the right environment.</p><h4>Copy globally installed opencv into your virtualenv</h4><p>Use this incase you want to copy your system installed opencv over to your virtualenv. In the virtualenv you make for your project, do the following</p><p>Copy cv.* (from OpenCV-2.2.0/lib directory) to the virtualenv site-packages</p><pre>cp /usr/lib/pymodules/python2.7/cv* $VIRTUAL_ENV/lib/python2.7/site-packages/</pre><p>Again, I have seen this approach used only by an intern. I can’t vouch for it’s reliability.</p><p>The second approach mitigates some of those problems. Thanks to some helpful people on the web, I could figure out how to install OpenCV inside a virtualenv without touching the global python</p><h4>Dependencies</h4><p>For opencv, there are system dependencies which can be installed like so,</p><pre>sudo apt-get install cmake libgtk2.0-dev pkg-config libavcodec-dev libavformat-devlibswscale-dev libamd2.2.0 libblas3gf libc6 libgcc1 libgfortran3 liblapack3gf libumfpack5.4.0 libstdc++6 build-essential gfortran libatlas-dev libatlas3-base libblas-dev liblapack-dev libjpeg-dev libpng-dev libtiff-devlibjasper-dev</pre><h4>Download opencv, build it and install it</h4><ul><li>Downloading</li></ul><p>Navigate to a folder to download opencv and download it</p><pre># change to target directory<br>cd ~/download_directory<br>wget -Oopencv-2.4.5.tar.bz2 http://downloads.sourceforge.net/project/opencvlibrary/opencv-unix/2.4.5/opencv-2.4.5.tar.gz?ts=1371369803&amp;use_mirror=jaist</pre><pre># unzip<br>tar xvf opencv-2.4.5.tar.bz2</pre><ul><li>Building it</li></ul><p>Now make sure that your virtual environment is activated and then execute this command.</p><pre>cd opencv-2.4.5/<br>mkdir release<br>cd release</pre><pre># notice that in the command below, <br># -D INSTALL_PYTHON_EXAMPLES=ON <br># is optional.</pre><pre>cmake -D MAKE_BUILD_TYPE=RELEASE -D CMAKE_INSTALL_PREFIX=$VIRTUAL_ENV/local/ -D PYTHON_EXECUTABLE=$VIRTUAL_ENV/bin/python -D PYTHON_PACKAGES_PATH=$VIRTUAL_ENV/lib/python2.7/site-packages -D INSTALL_PYTHON_EXAMPLES=ON ..</pre><ul><li>Installing it</li></ul><pre>make -j8<br>make install</pre><h4>Testing if everything went right</h4><p>Get into python shell and try the following commands</p><pre># enter python shell<br># make sure your environment is activated<br>python</pre><p>now inside the shell, try the following imports</p><pre>import cv<br>import cv2<br>import numpy<br>import scipy</pre><p>these imports shouldn&#39;t throw any error</p><p>Later, you might run into a decoder error. As a precaution, reinstall Pillow</p><pre>pip freeze | grep Pil # to find the Pillow version we&#39;re using</pre><pre>pip -I install Pillow==2.0.0 # or whichever version you find from the step above</pre><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=bf4d82220313" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>