add paligemma to projects

Author: sarthak247

Diff for: config.yaml

@@ -348,6 +348,21 @@ params:
           - icon: fas fa-blog 
             url: /blogs/attention-is-all-you-need
 
+      - title: "PaliGemma: A versatile 3B VLM for transfer"
+        content: Paper implementation of Paligemma inspired by Umar Jamil. Currently a work in progress
+        image: /projects/paligemma.png
+        # featured:
+        #   name: Demo # Fixit
+        #   link: https://hugo-profile.netlify.app
+        badges:
+          - "VLM"
+          - "Code Follow Along"
+          - "Vision Model"
+          - "Transformers"
+        links:
+          - icon: fab fa-github
+            url: https://github.com/sarthak247/PaliGemma
+
       - title: 🐱 NekoPDF 📖 - Document Interaction with LLM
         content: |-
           A web app to seamlessly interact with your documents using a Large Language Model over the Langchain framework. Currently support OpenAI and a single PDF file at a time but an option for other LLM backends and multiple PDF files will be added soon.

Diff for: public/blogs/word-embeddings/index.html

@@ -547,16 +547,13 @@ <h2 id="tf-idf">TF-IDF</h2>
 <li><strong>Term Frequency (TF)</strong>: This component measures the frequency of a term (word) within a document. It rewards words that appear frequently within a document. It can be calculated as the ratio of the word&rsquo;s occurrences in a document to the total word count in that document.
 l</li>
 </ul>
-$$TF(term)=\frac{\text{Number of times term appears in a document}}{\text{Total number of items in the document}}$$
-<p>For example, consider our previous example <code>I love coding and coding is fun.</code> Here, TF(coding) is 1/6 as the word <code>and</code> has been ignored.</p>
+$$TF(term)=\frac{\text{Number of times term appears in a document}}{\text{Total number of items in the document}}$$<p>For example, consider our previous example <code>I love coding and coding is fun.</code> Here, TF(coding) is 1/6 as the word <code>and</code> has been ignored.</p>
 <ul>
 <li><strong>Inverse Document Frequeny</strong>: This component measures the rarity or importance of a word across all the documents. It is the log of the inverse of the document frequency where document frequency tells us the number of documents which contain a particular word.</li>
 </ul>
-$$DF(term)=\frac{\text{Documents containing our term}}{\text{Total number of documents}}$$
-<p>Thus, DF tells us about the proportion of documents which contain our word of interest. Thus, we inverse it to make sure that the more common a word is, example stopwords, the less score it gets and a logarithm is taken to dampen or reduce it&rsquo;s effect on the final calculation.</p>
+$$DF(term)=\frac{\text{Documents containing our term}}{\text{Total number of documents}}$$<p>Thus, DF tells us about the proportion of documents which contain our word of interest. Thus, we inverse it to make sure that the more common a word is, example stopwords, the less score it gets and a logarithm is taken to dampen or reduce it&rsquo;s effect on the final calculation.</p>
 <!-- raw HTML omitted -->
-$$IDF(term)=\log{\bigg(\frac{\text{Total number of documents}}{\text{Documents containing our term}}\bigg)}$$
-<p>Thus, Inverse Document Frequency (IDF) is a measure of how unique or significant a word is across a collection of documents. It can be computed as the logarithm of the total number of documents divided by the number of documents in which the word occurs, effectively quantifying the word&rsquo;s rarity and importance in the entire document collection.</p>
+$$IDF(term)=\log{\bigg(\frac{\text{Total number of documents}}{\text{Documents containing our term}}\bigg)}$$<p>Thus, Inverse Document Frequency (IDF) is a measure of how unique or significant a word is across a collection of documents. It can be computed as the logarithm of the total number of documents divided by the number of documents in which the word occurs, effectively quantifying the word&rsquo;s rarity and importance in the entire document collection.</p>
 <h3 id="pros-2">Pros</h3>
 <ol>
 <li>

Diff for: public/files/resume.pdf

@@ -2,7 +2,7 @@
 <html>
 
 <head>
-	<meta name="generator" content="Hugo 0.128.2"><script src="/livereload.js?mindelay=10&amp;v=2&amp;port=1313&amp;path=livereload" data-no-instant defer></script><meta charset="UTF-8">
+	<meta name="generator" content="Hugo 0.133.1"><script src="/livereload.js?mindelay=10&amp;v=2&amp;port=1313&amp;path=livereload" data-no-instant defer></script><meta charset="UTF-8">
 <meta name="viewport" content="width=device-width, initial-scale=1.0">
 <meta http-equiv="X-UA-Compatible" content="ie=edge">
 <meta http-equiv="Accept-CH" content="DPR, Viewport-Width, Width">
@@ -1111,6 +1111,43 @@ <h5 class="card-title bg-transparent mt-1">Attention Is All You Need</h5>
                 </div>
             </div>
 
+            <div class="col-lg-4 col-md-6 my-3">
+                <div class="card my-3 h-100" title="PaliGemma: A versatile 3B VLM for transfer">
+                    <div class="card-head">
+                        <img class="card-img-top" src="/projects/paligemma.png" alt="PaliGemma: A versatile 3B VLM for transfer">
+                    </div>
+                    <div class="card-body bg-transparent p-3">
+                        <div class="pb-2 bg-transparent">
+                            
+                            <span class="badge badge-secondary">VLM</span>
+                            
+                            <span class="badge badge-secondary">Code Follow Along</span>
+                            
+                            <span class="badge badge-secondary">Vision Model</span>
+                            
+                            <span class="badge badge-secondary">Transformers</span>
+                            
+                        </div>
+                        <h5 class="card-title bg-transparent mt-1">PaliGemma: A versatile 3B VLM for transfer</h5>
+                        <div class="card-text bg-transparent secondary-font">
+                            Paper implementation of Paligemma inspired by Umar Jamil. Currently a work in progress
+                        </div>
+                    </div>
+                    
+                    <div class="card-footer py-3">
+                        
+                        <span class="m-1 mx-2">
+                            <a href="https://github.com/sarthak247/PaliGemma" target="_blank">
+                                <i class="fab fa-github"></i>
+                            </a>
+                        </span>
+                        
+                        
+                    </div>
+                    
+                </div>
+            </div>
+            
             <div class="col-lg-4 col-md-6 my-3">
                 <div class="card my-3 h-100" title="🐱 NekoPDF 📖 - Document Interaction with LLM">
                     <div class="card-head">