<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>DARSHAN PATEL</title><link>https://mrquantum1915.github.io/</link><description>Recent content on DARSHAN PATEL</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Fri, 20 Mar 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://mrquantum1915.github.io/index.xml" rel="self" type="application/rss+xml"/><item><title>Chasing 99%: Analysis of my experiments...</title><link>https://mrquantum1915.github.io/blogs/digit-recognition-analysis/</link><pubDate>Fri, 20 Mar 2026 00:00:00 +0000</pubDate><guid>https://mrquantum1915.github.io/blogs/digit-recognition-analysis/</guid><description>&lt;p&gt;This blog contains the full analysis of the models I trained on &lt;a href="https://www.kaggle.com/datasets/hojjatk/mnist-dataset"&gt;MNIST Dataset&lt;/a&gt; for digit recognition. Training set contains 60,000 images while test set contains 10,000 images.&lt;/p&gt;
&lt;p&gt;You can follow along with the code in the repo &lt;a href="https://github.com/MrQuantum1915/Digit-Recognition-NN/blob/main/dr_nn.ipynb"&gt;Digit-Recognition-NN&lt;/a&gt;. I have written pretty detailed explanations in that jupyter notebook too about the choices and implementations.&lt;/p&gt;
&lt;h2 id="a-bit-about-implementation"&gt;A bit about implementation&lt;/h2&gt;
&lt;p&gt;I used $n$ Hidden Layer Fully connected Neural Network (not CNN because I don&amp;rsquo;t learn it yet :)).&lt;/p&gt;</description></item></channel></rss>