<?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 Yodi D on Medium]]></title>
        <description><![CDATA[Stories by Yodi D on Medium]]></description>
        <link>https://medium.com/@yo__di?source=rss-1e75a2ef988f------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*t6iG5nligO05o6DC.jpeg</url>
            <title>Stories by Yodi D on Medium</title>
            <link>https://medium.com/@yo__di?source=rss-1e75a2ef988f------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 21:10:08 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@yo__di/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[Loncat-Loncat atau Tidak?]]></title>
            <link>https://medium.com/@yo__di/loncat-loncat-atau-tidak-462e0c7ac5e2?source=rss-1e75a2ef988f------2</link>
            <guid isPermaLink="false">https://medium.com/p/462e0c7ac5e2</guid>
            <category><![CDATA[employment]]></category>
            <category><![CDATA[karir]]></category>
            <category><![CDATA[work]]></category>
            <category><![CDATA[gaji]]></category>
            <dc:creator><![CDATA[Yodi D]]></dc:creator>
            <pubDate>Wed, 23 Sep 2020 04:03:47 GMT</pubDate>
            <atom:updated>2020-09-23T04:03:47.478Z</atom:updated>
            <content:encoded><![CDATA[<p>Tren karyawan jaman dulu adalah bekerja lama disebuah perusahaan, 5 tahun kerja minimal, baru pindah, bahkan bisa seumur hidup di 1 kantor. Berbeda dengan jaman sekarang, setidaknya di lingkungan saya, pindah tiap tahun mulai menjadi hal yang lumrah. Kedua hal itu membuat saya penasaran, mana sebenarnya yang lebih baik? Ada banyak hal untuk mendefinisikan ‘lebih baik’, bisa suasana kantor, bentuk apresiasi, karakter supervisor, fasilitas, jarak, pendapatan dan lain-lain. Jadi ‘lebih baik’ itu kompleks. Tapi, bukan berarti tidak bisa diukur, sehingga saya coba mengukur hal-hal yang bisa diukur terlebih dahulu seperti pendapatan dan tentu dengan banyak penyederhanaan. Jadi, saya membuat simulasi orang yang bekerja loncat-loncat vs orang yang hanya 1 kantor seumur hidup.</p><p>Saya membuat karakter fiktif, ada Anton dan Bambang. Anton dan Bambang adalah karyawan swasta, tapi mereka berbeda. Anton suka pindah-pindah, tiap 3 tahun sekali pasti pindah. Bukan tanpa alasan Anton pindah-pindah tepat 3 tahun sekali. Tiap pindah kantor dia bisa naik gaji minimal 10%, kalau di 1 kantor saja naik sekitar 5%. Betapa menggiurkannya untuk Anton. Selain itu, dia belajar di Undang-Undang Ketenagakerjaan bahwa jika karyawan kerja minimal 3 tahun, maka dia akan memperoleh Uang Penghargaan Masa Kerja (UPMK), besarannya adalah 2x gaji. Bagi dia semakin cepat dapat UPMK, semakin baik. UPMK bertambah 1x gaji setiap 3 tahun. Menurut dia lebih lama daripada itu rugi, 3 tahun dapat 2x gaji, kok tambah 3 tahun cuman nambah 1x gaji.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/457/1*BjYOi76UfFi5MP0kPnZfpw.jpeg" /><figcaption>Sumber : <a href="https://www.gadjian.com/blog/2018/02/01/perhitungan-uang-penghargaan-masa-kerja-bagi-karyawan-di-phk/">https://www.gadjian.com/blog/2018/02/01/perhitungan-uang-penghargaan-masa-kerja-bagi-karyawan-di-phk/</a></figcaption></figure><p>Berbeda dengan Bambang. Bambang seumur hidupnya bekerja hanya di 1 kantor, 25 tahun! Bayangkan, sebetah apa Bambang disana. Bambang juga tidak keberatan selama 25 tahun gajinya naik minimal 5% tahun. Entah apa yang membuat betah Bambang. Bambang juga sadar kalau berhenti kerja nanti dapat UPMK 10x gaji.</p><p>Walaupun berbeda filosofi, ternyata Anton dan Bambang memiliki kesamaan. Selain sama-sama naik gaji 5%/tahun jika tetap di kantor, ternyata gaji mereka di tahun pertama juga sama, katakanlah 1X atau 1x gaji. Dan dalam 1 tahun mereka selalu memproleh 13 x gaji di perusahaan manapun.</p><p>Dari informasi-informasi ini dapat kita buat simulasi, berapa sih total pendapatan Anton vs Bambang setelah selama 25 tahun? Berikut simulasinya :</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GvcDg3bC0JGIo9s-23ftug.png" /><figcaption>Simulasi 6 tahun pertama Anton vs Bambang</figcaption></figure><p>Gambar di atas adalah simulasi 6 tahun pertama Anton dan Bambang bekerja. Tahun ke 1 mereka sama-sama memproleh 13x gaji. Tahun ke 2 mereka juga sama-sama memperoleh 14.33x gaji karena mereka naik gaji dengan besaran yang sama yaitu 5%. Gaji Anton dan Bambang di tahun ke-2 dan seterusnya adalah besar gaji relatif terhadap tahun pertama. Berbeda ketika di penghujung tahun ketika, Anton mendapat 2.21x gaji UPMK. Bahkan di awal tahun keempat, Anton naik gaji 10%, 5% lebih banyak daripada Bambang, sehingga di tahun ke-4 mulai berbeda. Anton memperoleh 15.77x gaji, sedangkan Bambang memproleh 15.05x gaji. Kejadian ini berlangsung selama 25 tahun. Inilah yang terjadi setelah 25 tahun :</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*62sjZpjk4RgHdMBpTHtctw.png" /><figcaption>Simulasi 5 tahun terakhir Anton vs Bambang</figcaption></figure><p>Ternyata Anton memproleh 155x gaji lebih banyak daripada Bambang. Ingat, 155x gaji tahun pertama kerja. Jika relatif terhadap gaji Bambang di tahun terakhir, 42x gaji, selisihnya adalah sekitar 48 bulan gaji terakhir dia [155x gaji/(42x gaji /13 periode gajian)]. Rasanya seperti Anton lebih lama kerja sekitar 4 tahun daripada Bambang, relatif terhadap tahun terakhir Bambang.</p><p>Apakah Bambang sakit hati dengan perbedaan 4 bulan gaji? Siapa yang tahu, tapi sepertinya tidak. Kenapa? Karena Bambang ternyata punya istri bernama Cinta. Cinta juga adalah seorang karyawan. Tapi, berbeda dengan Bambang, Cinta ternyata lebih mirip Anton dalam urusan kerja, hanya 1 perbedaan. Jika Anton tiap pindah minimal naik gaji 10%, Cinta justru naik gaji minimal 30%. Setelah 25 tahun, ternyata Cinta mampu menghasilkan pendapatan sebesar berikut:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*HIMUnI3lMSWiv_tcXzDK0g.png" /><figcaption>Simulasi 6 tahun terakhir Cinta vs Bambang</figcaption></figure><p>1,254x gaji lebih besar daripada Anton, atau setara dengan</p><p>1254x gaji / (42x gaji/13 bulan gajian) = 388 bulan gajian Bambang tahun terakhir.</p><p>388 bulan / 13 periode gaji = 30 tahun kerja Bambang relatif terhadap tahun terakhir.</p><p>2 Bambang saja tidak cukup untuk menyaingi Cinta.</p><p>Jadi, pindah atau tidak?</p><p>Jadi pindah atau tidak?</p><p>— — — — — — — — — — — — — — — — — — — — — — —</p><p>Note :</p><ul><li>Anton dan Bambang adalah simulasi dengan pendekatan konservatif.</li><li>Cinta adalah simulasi dengan pendekatan optimis terancam naif.</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=462e0c7ac5e2" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Pandas Slicing : Readable Way]]></title>
            <link>https://medium.com/mytake/pandas-slicing-readable-way-91484799d80c?source=rss-1e75a2ef988f------2</link>
            <guid isPermaLink="false">https://medium.com/p/91484799d80c</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[python]]></category>
            <category><![CDATA[towards-data-science]]></category>
            <category><![CDATA[programming]]></category>
            <category><![CDATA[pandas]]></category>
            <dc:creator><![CDATA[Yodi D]]></dc:creator>
            <pubDate>Sun, 29 Sep 2019 15:54:11 GMT</pubDate>
            <atom:updated>2019-12-02T13:46:27.893Z</atom:updated>
            <content:encoded><![CDATA[<h3>Pandas Slicing : Readable Way</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7rDK_LL3Be_X7JAiI-fSIA.jpeg" /><figcaption>Image source : <a href="https://www.sunkist.com/tips/how-to-slice-oranges/">https://www.sunkist.com/tips/how-to-slice-oranges/</a></figcaption></figure><p>Let say we have df1.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/409/1*k9wtGzYCZLTkJOJnNTrTBw.png" /><figcaption>df1</figcaption></figure><p>Most of us, doing things like this when slicing dataframe.</p><pre>value9 = df1.loc[df1[‘ID’]==’1&#39;, &#39;value9&#39;]</pre><p>It’s ok actually, but it gets worse when we have multiple criteria</p><pre>#THIS ALL IN ONE LINE ACTUALLY<br>value9 = df1.loc[(df1[‘ID’]==’1&#39;) &amp; (df1[‘value1’]==’10&#39;) &amp; (df1[‘value2’]==’21&#39;) &amp; (df1[‘value3’]==’22&#39;) &amp; (df1[‘value4’]==’24&#39;), &#39;value9&#39;]</pre><p>It becomes not readable. Why don’t we add a new line?</p><pre>value9 = df1.loc[(df1[‘ID’]==’1&#39;) &amp; \<br>(df1[‘value1’]==’10&#39;) &amp; \<br>(df1[‘value2’]==’21&#39;) &amp; \<br>(df1[‘value3’]==’22&#39;) &amp; \<br>(df1[‘value4’]==’24&#39;), \<br>&#39;value9&#39;]</pre><p>Well ummh, it’s readable, but the codes become ugly, any solution ?</p><p><strong>What if we separate the selection part and put it in an object ?</strong></p><pre>select = (df1[‘ID’]==’1&#39;) &amp; \<br>(df1[‘value1’]==’10&#39;) &amp; \<br>(df1[‘value2’]==’21&#39;) &amp; \<br>(df1[‘value3’]==’22&#39;) &amp; \<br>(df1[‘value4’]==’24&#39;)</pre><pre>value9 = df1.loc[select, &#39;value9&#39;]</pre><p>Yeap, it’s better now. Even, we can re-use the selection if we need other value/data.</p><pre>select = (df1[‘ID’]==’1&#39;) &amp; \<br>(df1[‘value1’]==’10&#39;) &amp; \<br>(df1[‘value2’]==’21&#39;) &amp; \<br>(df1[‘value3’]==’22&#39;) &amp; \<br>(df1[‘value4’]==’24&#39;)</pre><pre>#JUST EXAMPLE, I know u always copy-paste the slicing part, even me<br>value9 = df1.loc[select, &#39;value9&#39;]<br>value8 = df1.loc[select, &#39;value8&#39;]<br>shape = df1.loc[select, &#39;value9&#39;].shape<br>index = df1[select].index</pre><p>But there is something that bugs me. I have new line marks (‘\’) which make me :</p><ul><li>I don’t know what is first slicing criteria, is it slice by ‘ID’ on first-line or ‘value1’ at second-line or else ? Besides, ‘ID’ looks hidden.</li><li>I need to read lines one by one where is the end of ‘select’ operation. Maybe in the middle of the line is a different process of code.</li></ul><p>So, instead of using new lines, what if we <strong>use parenthesis?</strong></p><pre>select = (<br>     (df1[‘ID’]==’1&#39;) &amp;<br>     (df1[‘value1’]==’10&#39;) &amp;<br>     (df1[‘value2’]==’21&#39;) &amp;<br>     (df1[‘value3’]==’22&#39;) &amp;<br>     (df1[‘value4’]==’24&#39;)<br>)</pre><pre>value9 = df1.loc[select, &#39;value9&#39;]</pre><p>Now, I know where is ‘select’ code start and end at one sight. More positive things using parenthesis is, it’s easier to read if we have nested criteria, for example :</p><pre>select = (<br>     (df1[‘ID’]==’1&#39;) &amp;<br>     (<br>          (df1[‘value1’]==’10&#39;) | <br>          (df1[‘value1’]==’10&#39;)<br>     ) &amp;<br>     (<br>          (df1[‘value2’]==’21&#39;) | <br>          (df1[‘value2’]==’21&#39;)<br>     ) &amp;<br>     (df1[‘value3’]==’22&#39;) &amp;<br>     (df1[‘value4’]==’24&#39;)<br>)</pre><pre>value9 = df1.loc[select, &#39;value9&#39;]</pre><p>Wow, it’s ugly again, but it’s readable. Can we still improve? Add comments!</p><pre>select = (<br>     #CRITERIA1 =============<br>     (df1[‘ID’]==’1&#39;) &amp;</pre><pre>     #CRITERIA2 =============<br>     (<br>          (df1[‘value1’]==’10&#39;) | <br>          (df1[‘value1’]==’10&#39;)<br>     ) &amp;</pre><pre>     #CRITERIA3 =============<br>     (<br>          (df1[‘value2’]==’21&#39;) | <br>          (df1[‘value2’]==’21&#39;)<br>     ) &amp;</pre><pre>     #CRITERIA4 =============<br>     (df1[‘value3’]==’22&#39;) &amp;</pre><pre>     #CRITERIA5 =============<br>     (df1[‘value4’]==’24&#39;)<br>)</pre><pre>#DO SELECTION<br>value9 = df1.loc[select, &#39;value9&#39;]</pre><p>Well, this one is still ugly but very readable (at least for me). Can you help me to improve it? beside the chaining method?</p><p>Feedback really appreciated. Thanks!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=91484799d80c" width="1" height="1" alt=""><hr><p><a href="https://medium.com/mytake/pandas-slicing-readable-way-91484799d80c">Pandas Slicing : Readable Way</a> was originally published in <a href="https://medium.com/mytake">MyTake</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Python Hack : DF Merging]]></title>
            <link>https://medium.com/@yo__di/python-hack-df-merging-ffb576d41a88?source=rss-1e75a2ef988f------2</link>
            <guid isPermaLink="false">https://medium.com/p/ffb576d41a88</guid>
            <category><![CDATA[python]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[pandas]]></category>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[towards-data-science]]></category>
            <dc:creator><![CDATA[Yodi D]]></dc:creator>
            <pubDate>Fri, 27 Sep 2019 15:27:40 GMT</pubDate>
            <atom:updated>2019-09-27T15:37:06.600Z</atom:updated>
            <content:encoded><![CDATA[<h3>Python Hack : DF Merging</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Y5zUJ5xYk5-wVix_DwRZEA.jpeg" /><figcaption>Image : <a href="https://www.space.com/rare-earth-meteorites-formed-in-asteroid-vesta-crash.html">https://www.space.com/rare-earth-meteorites-formed-in-asteroid-vesta-crash.html</a> (©Makiko Haba)</figcaption></figure><p>We always facing this disasters, especially people who play with data.</p><p>I have two dataframe, df1 and df2.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/123/1*yxpGX9KiQvfa_4TBh2oFug.png" /><figcaption>df1</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/129/1*TDqJYfmXWangEDpyN7vjWA.png" /><figcaption>df2</figcaption></figure><p>I merge it. Done! A new df1! Nothing wrong here right?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/207/1*1uWL1Vp4ZeH5xK718-6zow.png" /><figcaption>New df1</figcaption></figure><p>Then, my stubborn hand is running at it again. Boom! Then I did not expect the df1 would become like this. (Sometime I run it more than twice! disasters-gross-dataframe!)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/330/1*gDlEOlMUbLqxf6_YGp0vIQ.png" /><figcaption>New new df1</figcaption></figure><p>But no problem, just run it from the beginning! Hold on, I just realize the df1 is 10GB and df2 is 5GB. Poor me :(</p><p>Imagine, I need to load this data again. I need 1 to 5 minutes just to load this csv data, using old grandma computer it will take forever. If I were smart enough, I already make it as pickle files. But still, there are gigantic line of preprocessing code, I need to run it again just because this stubborn finger. This problem is not once or twice happened, but uncountable times! How can I avoid this thing?</p><p>So, I design a safer way to merge the data. The root concept is simple,</p><blockquote>NEVER OVERRIDE THE MAIN DATA DIRECTLY,</blockquote><blockquote>N-E-V-E-R!</blockquote><p>What we need to do is, instead of clashing all of df as one, I will ADD the data into df1, it can be its column or row. How?</p><p>First, define the key. In my case it is ID. Call the df1 with its KEY ONLY, something like this :df1[[‘ID’]]</p><p>Second, select the column(s) you want to copy from df2, for example ‘value3’ and ‘value4’.</p><p>Then, the final one will be like this :</p><pre>df1[‘value3’] = df1[[‘ID’]].merge(df2, on=’ID’)[‘value3’]</pre><pre>df1[‘value4’] = df1[[‘ID’]].merge(df2, on=’ID’)[‘value4’]</pre><p>No matter how many times you run this cell, it won’t destroy the dataframe as before. But, you know what, we can do much better by storing all column as a list :</p><pre>cols_selected = [&#39;value3&#39;, &#39;value4&#39;]<br>df1[cols_selected] = df1[[‘ID’]].merge(df2, on=’ID’)[cols_selected]</pre><p>We still can do much better. In df2 we have exactly 3 columns. It won’t bite anything. But what if we have 200 columns? I guess it will eat your RAM when merge is processed. So, how can we solve this? Simple, pass the cols_selected into df2. DON’T FORGET to add the KEY or the ID! because if you pas cols_selected only, the ID on df2 is missing.</p><pre>cols_selected = [&#39;value3&#39;, &#39;value4&#39;]</pre><pre>#since cols_selected is a list, wrap &#39;ID&#39; as list too<br>df1[cols_selected] = df1[[‘ID’]].merge(<br>      <strong>df2[[&#39;ID&#39;] + cols_selected]</strong>, <br>      on=’ID’<br>)[cols_selected]</pre><p>What if I have a lot of key column, let say ID1 or ID2? Do the same things as value columns.</p><pre>keys_selected = [&#39;ID1&#39;, &#39;ID2&#39;]<br>cols_selected = [&#39;value3&#39;, &#39;value4&#39;]</pre><pre>df1[cols_selected] = df1[keys_selected].merge(<br>      df2[keys_selected + cols_selected], <br>      on=keys_selected<br>)[cols_selected]</pre><p>But, but, this whole solution only works for left join, how about, let say, cross join or right join? No problem! Since it generate different dimensions and stories, not just adding additional information into df1, it’s not relevant anymore to keep it in df1. It’s better to generate a new dataframe. Am I right?</p><p>Oh, yeah, be careful if you have unsorted or sliced index. It will screwed you up. Always reset the index after do the selections.</p><p>So, that’s all folks! See you in the next episode. Feedback are welcome. Thanks!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ffb576d41a88" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Jupyter Notebook Merger]]></title>
            <link>https://medium.com/@yo__di/jupyter-notebook-merger-d687f80f2e49?source=rss-1e75a2ef988f------2</link>
            <guid isPermaLink="false">https://medium.com/p/d687f80f2e49</guid>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[towards-data-science]]></category>
            <category><![CDATA[communication]]></category>
            <category><![CDATA[jupyter-notebook]]></category>
            <category><![CDATA[python]]></category>
            <dc:creator><![CDATA[Yodi D]]></dc:creator>
            <pubDate>Thu, 21 Feb 2019 16:02:26 GMT</pubDate>
            <atom:updated>2019-02-21T16:09:22.338Z</atom:updated>
            <content:encoded><![CDATA[<p>Communication is a key for data scientist. Without communication, your hard work will be nothing. There are a lot of way to communicate. Some people prefer to use MS Word or Power Point, some people use data aggregating tools such as Tableau or Kibana, some people use markdown or Jupyter Notebook, even the very simple form such as notepad.</p><p>For me, I use all of them, but most frequently used are MS Power Point and Jupyter Notebook. If I use MS Power Point it means I have a lot of time and discard most of the details. But if the boss want very comprehensive detail in sort time and frequently updated, Notebook came as solutions.</p><h3>Here comes the problem…</h3><p>Unfortunately, generate report with Jupyter Notebook always bug me with these problem :</p><ul><li><strong>Tons of code</strong>, my boss don’t want to see it. Solved by plugins, we can hide the codes.</li><li><strong>Tons of output</strong>, my boss only want to see several output. So, i need to save some output, manually handpick, and load output in other notebook.</li><li><strong>Working with a lot people</strong>. Boss want to see X from Alice, Y from Bob, Z from Charlie. To make a report, we need to dig into tons of code and output scattered from (possibly) tons of notebook.</li><li>Notebook is not reliable anymore as mentioned above : detail, fast, and updated. Looks like MS Word or Power Point become the only solution. But no, thanks.</li></ul><p>I become tired to read tons of notebook, code, and output from them, but i need to make report for the boss. So, I decide to thinking first,</p><blockquote>“Can we combine several notebook together?”</blockquote><h3>Concept</h3><p>And here comes the solution. I decide to make notebook merger. After thinking and research for a while, I think I found the solution. The concept is very simple :</p><ul><li>let say I have notebook A and B</li><li>I want to show X from A, and Y from B</li><li>A has 10 output, X is 9th output. B has 10 output Y is 6th output.</li><li>I will give special keyword to X and Y, let say #keyword_x and #keyword_y.</li><li>After giving keywords, i convert A and B into html. So we have A.html and B.html.</li><li>from html file, we will do scrapping, because it easier to do at html instead of ipynb. We will scrap the output with #keyword_x from A.html and #keyword_y from B.html. The form of both will be in HTML format.</li><li>Last, we can show scrapped element as HTML element.</li><li>We can do converting, scrapping, and displaying in notebook C.</li><li>So, if A and B have updated, we don’t need to open the notebooks one by one, seek X and Y, save the output, and load in other notebook.</li><li>Now, just simply rerun the notebook C.</li></ul><h3><strong>Example</strong></h3><p>For more brief explanation, let’s make an example :</p><ul><li>Install the library</li></ul><pre>pip install git+git://github.com/aiodia/junomerger.git</pre><ul><li>load the library</li></ul><pre>import junomerger</pre><ul><li>Create <a href="https://github.com/aiodia/junomerger/blob/master/examples/notebook_a.ipynb"><strong>notebook_a</strong></a> and <a href="https://github.com/aiodia/junomerger/blob/master/examples/notebook_b.ipynb"><strong>notebook_b</strong></a></li><li>We will merge notebook_a and notebook_b at <a href="https://github.com/aiodia/junomerger/blob/master/examples/notebook_c.ipynb"><strong>notebook_c</strong></a></li><li>Give unique keyword to particular output in notebook_a and b. run, <strong>save :</strong></li></ul><pre>junomerger.print_hidden_tag(&#39;my_unique_keyword&#39;)</pre><ul><li>in notebook_c, we convert notebook_a and b into html :</li></ul><pre>notebook_a = junomerger.convert_ipynb2html(&#39;notebook_a.ipynb&#39;)<br>notebook_b = junomerger.convert_ipynb2html(&#39;notebook_b.ipynb&#39;)</pre><ul><li>Then we show particular output based on unique keyword has been created :</li></ul><pre>notebook_a.show_jupyter_output(keyword=&#39;my_unique_keyword&#39;)</pre><h3>Conclusions</h3><p>Now, my task become simpler. I just need to create skeleton to merge all of the notebook, so i can see the progress daily without opening notebook one by one.</p><p><a href="https://github.com/aiodia/junomerger">aiodia/junomerger</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d687f80f2e49" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AnaLieSys : Bar Chart]]></title>
            <link>https://medium.com/@yo__di/analiesys-bar-chart-5ccc74707bbd?source=rss-1e75a2ef988f------2</link>
            <guid isPermaLink="false">https://medium.com/p/5ccc74707bbd</guid>
            <category><![CDATA[analysis]]></category>
            <category><![CDATA[analytics]]></category>
            <category><![CDATA[visualization]]></category>
            <category><![CDATA[fraud]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Yodi D]]></dc:creator>
            <pubDate>Thu, 07 Jun 2018 16:19:07 GMT</pubDate>
            <atom:updated>2018-07-13T14:17:02.217Z</atom:updated>
            <content:encoded><![CDATA[<p>Pak Blangkon adalah lurah di desa Wakanda Wetan. Suatu hari Pak Blangkon didatangi sekretarisnya dengan tergopoh-gopoh. Dengan nafas tak teratur dia bilang, “gawat Pak Lurah, gawat”. Kemudian ia menyodorkan selembar kertas, isinya adalah perbandingan gaji Pak Blangkon dengan lurah sebelah, Pak Dengklek. Pak Blangkon naik pitam, ternyata gaji dia menurut mereka jauh betul dibandingin Pak Dengklek. Berikut isi kertasnya :</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/700/0*38bN7dANTiD2Gu90" /></figure><p>Pak Blangkon marah, akhirnya beliau berencana melabrak Bupati Wakanda soalnya beliau yang mengatur masalah gaji lurah-lurahnya.</p><p>Di tengah jalan, Pak Blangkon bertemu Didi Dukun. Didi Dukun heran, kenapa wajah Pak Blangkon seperti tidak biasanya, sangat jutek sekali. Pak Blangkon orangnya ember, dia ceritakan kalau gaji dia terpaut jauh dengan Pak Dengklek, beliau tidak terima, mau melabrak Pak Bupati. Didi Dukun penasaran, ia bertanya, memangnya sejauh apa ? Pak Blangkon menyodorkan lembaran kertas tadi.</p><p>Sebagai dukun di Wakanda, Didi ternyata tidak kaget. Didi Dukun yakin ini hanya karena ketidaktelitian Pak Blangkon. Didi menjelaskan bahwa bedanya gaji beliau berdua cuma beda 1 juta, itu pun Pak Dengklek 1 tahun lebih lama masa baktinya daripada Pak Blangkon, dan itu wajar di Wakanda. Didi Dukun jelaskan juga bahwa Pak Blangkon tertipu oleh bar chart ini dan kurang teliti. Karena bar chartnya di-deliver (apa ya, dipersembahkan gitu ? hahaha) bukan dari angka 0, tapi dari 19 juta, jadinya kesannya terlihat beda jauh, padahal kalau dari 0 bedanya sedikit.</p><p>Agar lebih jelas, akhirnya Didi Dukun memperbaiki bar chart tersebut. Dengan bantuan para jin, ini merupakan tugas yang mudah. Dengan bahasa Parsertongue-nya Didi Dukun merapalkan mantra, Pertamina Pertamax! Poof!, bar chart tersebut diperbaiki menjadi seperti ini :</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/551/0*A1ZAMRE-sm2Lxoxi" /></figure><p>Pak Blangkon angguk-angguk, lalu nyengir. Pak Blangkon pun ngaku kalau sebenarnya di tidak teliti. Dia hanya melihat bar-nya saja, tidak melihat angka di samping. Dari barnya kelihatannya dobel, dia kira gaji Pak Dengklek dobel daripada dia.</p><p>Pak Blangkon pulang dengan hati lega.</p><p>Lalu ngepret sekretarisnya.</p><p>Sekretarisnya mencret.</p><p>Masuk rumah sakit.</p><p>Udah ah.</p><p><em>Originally published at </em><a href="https://cacatannya.wordpress.com/2018/06/07/analiesys-bar-chart/"><em>cacatannya.wordpress.com</em></a><em> on June 7, 2018.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5ccc74707bbd" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>