-
Notifications
You must be signed in to change notification settings - Fork 30
/
index.html
73 lines (70 loc) · 4.52 KB
/
index.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>README</title>
<link rel="stylesheet" href="https://stackedit.io/style.css" />
</head>
<body class="stackedit">
<div class="stackedit__html"><h1 id="normalization--regularization">Normalization & Regularization</h1>
<p>Code for the Experiments</p>
<p><code>google-colab link</code>: <a href="https://colab.research.google.com/github/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/06_NormalizationRegularization/06_NormalizationRegularization_Modularized_FINAL.ipynb">06_NormalizationRegularization_Modularized_FINAL</a></p>
<p><code>github link</code>: <a href="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/06_NormalizationRegularization/06_NormalizationRegularization_Modularized_FINAL.ipynb">06_NormalizationRegularization_Modularized_FINAL.ipynb</a></p>
<blockquote>
<p>Dataset: MNIST Digits<br>
Model: CNN - 9960 params<br>
Epochs: 40</p>
</blockquote>
<p>Regularization Values</p>
<pre><code>l1_lambda = 5e-4
l2_lambda = 5e-3
</code></pre>
<p><br>
L1 Regularization</p>
<p><img src="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/06_NormalizationRegularization/l1_reg.png?raw=true" alt="l1_reg.png"></p>
<p>L2 Regularization</p>
<p><img src="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/06_NormalizationRegularization/l2_reg.png?raw=true" alt="l2_reg.png"></p>
<h1 id="results">Results</h1>
<h2 id="no-l1-no-l2">1. No L1 No L2</h2>
<blockquote>
<p><code>Highest Accuracy: 99.47</code></p>
</blockquote>
<h3 id="metrics">Metrics</h3>
<p><img src="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/06_NormalizationRegularization/plots/nol1_nol2_metrics.png?raw=true" alt="nol1_nol2_metrics.png"></p>
<h3 id="misclassifications">Misclassifications</h3>
<p><img src="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/06_NormalizationRegularization/plots/nol1_nol2_misclassifications.png?raw=true" alt="nol1_nol2_misclassifications.png"></p>
<h2 id="l1">2. L1</h2>
<blockquote>
<p><code>Highest Accuracy: 99.44</code></p>
</blockquote>
<h3 id="metrics-1">Metrics</h3>
<p><img src="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/06_NormalizationRegularization/plots/l1_metrics.png?raw=true" alt="l1_metrics.png"></p>
<h3 id="misclassifications-1">Misclassifications</h3>
<p><img src="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/06_NormalizationRegularization/plots/l1_misclassifications.png?raw=true" alt="l1_misclassifications.png"></p>
<h2 id="l2">3. L2</h2>
<blockquote>
<p><code>Highest Accuracy: 99.52</code></p>
</blockquote>
<h3 id="metrics-2">Metrics</h3>
<p><img src="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/06_NormalizationRegularization/plots/l2_metrics.png?raw=true" alt="l2_metrics.png"></p>
<h3 id="misclassifications-2">Misclassifications</h3>
<p><img src="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/06_NormalizationRegularization/plots/l2_misclassifications.png?raw=true" alt="l2_misclassifications.png"></p>
<h2 id="l1--l2">4. L1 & L2</h2>
<blockquote>
<p><code>Highest Accuracy: 98.76</code></p>
</blockquote>
<h3 id="metrics-3">Metrics</h3>
<p><img src="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/06_NormalizationRegularization/plots/l1_l2_metrics.png?raw=true" alt="l1_l2_metrics.png"></p>
<h3 id="misclassifications-3">Misclassifications</h3>
<p><img src="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/06_NormalizationRegularization/plots/l1_l2_misclassifications.png?raw=true" alt="l1_l2_misclassifications.png"></p>
<h1 id="comparison">Comparison</h1>
<h2 id="validation-accuracy">Validation Accuracy</h2>
<p><img src="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/06_NormalizationRegularization/plots/val_acc_compare2.png?raw=true" alt="val_acc_compare2.png"></p>
<p><img src="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/06_NormalizationRegularization/plots/val_acc_compare.png?raw=true" alt="val_acc_compare.png"></p>
<h2 id="validation-loss">Validation Loss</h2>
<p><img src="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/06_NormalizationRegularization/plots/val_loss_compare2.png?raw=true" alt="val_loss_compare2.png"></p>
<p><img src="https://github.com/satyajitghana/TSAI-DeepVision-EVA4.0/blob/master/06_NormalizationRegularization/plots/val_loss_compare.png?raw=true" alt="val_loss_compare.png"></p>
</div>
</body>
</html>