title
Artificial Neural Network Tutorial | Deep Learning With Neural Networks | Edureka

description
( TensorFlow Training - https://www.edureka.co/ai-deep-learning-with-tensorflow ) This Edureka "Neural Network Tutorial" video (Blog: https://goo.gl/4zxMfU) will help you to understand the basics of Neural Networks and how to use it for deep learning. It explains Single layer and Multi layer Perceptron in detail. Below are the topics covered in this tutorial: 1. Why Neural Networks? 2. Motivation Behind Neural Networks 3. What is Neural Network? 4. Single Layer Percpetron 5. Multi Layer Perceptron 6. Use-Case 7. Applications of Neural Networks Subscribe to our channel to get video updates. Hit the subscribe button above. Check our complete Deep Learning With TensorFlow playlist here: https://goo.gl/cck4hE PG in Artificial Intelligence and Machine Learning with NIT Warangal : https://www.edureka.co/post-graduate/machine-learning-and-ai Post Graduate Certification in Data Science with IIT Guwahati - https://www.edureka.co/post-graduate/data-science-program (450+ Hrs || 9 Months || 20+ Projects & 100+ Case studies) - - - - - - - - - - - - - - How it Works? 1. This is 21 hrs of Online Live Instructor-led course. Weekend class: 7 sessions of 3 hours each. 2. We have a 24x7 One-on-One LIVE Technical Support to help you with any problems you might face or any clarifications you may require during the course. 3. At the end of the training you will have to undergo a 2-hour LIVE Practical Exam based on which we will provide you a Grade and a Verifiable Certificate! - - - - - - - - - - - - - - About the Course Edureka's Deep learning with Tensorflow course will help you to learn the basic concepts of TensorFlow, the main functions, operations and the execution pipeline. Starting with a simple “Hello Word” example, throughout the course you will be able to see how TensorFlow can be used in curve fitting, regression, classification and minimization of error functions. This concept is then explored in the Deep Learning world. You will evaluate the common, and not so common, deep neural networks and see how these can be exploited in the real world with complex raw data using TensorFlow. In addition, you will learn how to apply TensorFlow for backpropagation to tune the weights and biases while the Neural Networks are being trained. Finally, the course covers different types of Deep Architectures, such as Convolutional Networks, Recurrent Networks and Autoencoders. Delve into neural networks, implement Deep Learning algorithms, and explore layers of data abstraction with the help of this Deep Learning with TensorFlow course. - - - - - - - - - - - - - - Who should go for this course? The following professionals can go for this course: 1. Developers aspiring to be a 'Data Scientist' 2. Analytics Managers who are leading a team of analysts 3. Business Analysts who want to understand Deep Learning (ML) Techniques 4. Information Architects who want to gain expertise in Predictive Analytics 5. Professionals who want to captivate and analyze Big Data 6. Analysts wanting to understand Data Science methodologies However, Deep learning is not just focused to one particular industry or skill set, it can be used by anyone to enhance their portfolio. - - - - - - - - - - - - - - Why Learn Deep Learning With TensorFlow? TensorFlow is one of the best libraries to implement Deep Learning. TensorFlow is a software library for numerical computation of mathematical expressions, using data flow graphs. Nodes in the graph represent mathematical operations, while the edges represent the multidimensional data arrays (tensors) that flow between them. It was created by Google and tailored for Machine Learning. In fact, it is being widely used to develop solutions with Deep Learning. Machine learning is one of the fastest-growing and most exciting fields out there, and Deep Learning represents its true bleeding edge. Deep learning is primarily a study of multi-layered neural networks, spanning over a vast range of model architectures. Traditional neural networks relied on shallow nets, composed of one input, one hidden layer and one output layer. Deep-learning networks are distinguished from these ordinary neural networks having more hidden layers, or so-called more depth. These kinds of nets are capable of discovering hidden structures within unlabeled and unstructured data (i.e. images, sound, and text), which constitutes the vast majority of data in the world. For more information, please write back to us at sales@edureka.co or call us at IND: 9606058406 / US: 18338555775 (toll-free). Facebook: https://www.facebook.com/edurekaIN/ Twitter: https://twitter.com/edurekain LinkedIn: https://www.linkedin.com/company/edureka

detail
{'title': 'Artificial Neural Network Tutorial | Deep Learning With Neural Networks | Edureka', 'heatmap': [{'end': 661.809, 'start': 593.618, 'weight': 1}, {'end': 1254.126, 'start': 1209.346, 'weight': 0.744}], 'summary': 'Tutorial covers the fundamentals of neural networks and their application in authenticating banknotes, the importance and historical context of neural networks, the decision-making process and model for attending a beer festival, neural network basics, back propagation, achieving 99% accuracy in neural network training, and applications in medicine and business.', 'chapters': [{'end': 267.227, 'segs': [{'end': 72.015, 'src': 'embed', 'start': 17.414, 'weight': 0, 'content': [{'end': 23.236, 'text': "We need to figure out if the banknotes are real or fake, and for that we'll be using artificial neural networks,", 'start': 17.414, 'duration': 5.822}, {'end': 26.417, 'text': 'and obviously we need some sort of data in order to train our network.', 'start': 23.236, 'duration': 3.181}, {'end': 28.632, 'text': 'So let us see how the data set looks like.', 'start': 27.011, 'duration': 1.621}, {'end': 32.253, 'text': "So over here I've taken a screenshot of the data set with few of the rows.", 'start': 29.232, 'duration': 3.021}, {'end': 38.256, 'text': 'In it data were extracted from images that were taken from genuine and forged banknote like specimens.', 'start': 32.934, 'duration': 5.322}, {'end': 43.358, 'text': 'After that wavelet transform tools were used to extract features from those images.', 'start': 39.056, 'duration': 4.302}, {'end': 49.741, 'text': "And these are a few features that I'm highlighting with my cursor and the final column or the last column actually represents the label.", 'start': 43.918, 'duration': 5.823}, {'end': 53.77, 'text': 'So basically label tells us to which class that pattern represents.', 'start': 50.509, 'duration': 3.261}, {'end': 57.411, 'text': 'Whether that pattern represents a fake note or it represents a real note.', 'start': 54.23, 'duration': 3.181}, {'end': 60.112, 'text': 'Let us discuss these features and labels one by one.', 'start': 57.911, 'duration': 2.201}, {'end': 64.913, 'text': 'So the first feature or the first column is nothing but variants of a wavelet transformed image.', 'start': 60.752, 'duration': 4.161}, {'end': 67.214, 'text': 'The second column is about skewness.', 'start': 65.573, 'duration': 1.641}, {'end': 72.015, 'text': 'The third is courtesies of wavelet transformed image and finally fourth one is entropy of the image.', 'start': 67.614, 'duration': 4.401}], 'summary': 'Using artificial neural networks to classify banknotes as real or fake based on image features and labels.', 'duration': 54.601, 'max_score': 17.414, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU17414.jpg'}, {'end': 134.395, 'src': 'embed', 'start': 104.332, 'weight': 2, 'content': [{'end': 108.375, 'text': "After that, we'll use TensorFlow data structures for holding features, labels, et cetera.", 'start': 104.332, 'duration': 4.043}, {'end': 114.76, 'text': 'And TensorFlow is nothing but a Python library that is used in order to implement deep learning models, or you can say neural networks.', 'start': 108.976, 'duration': 5.784}, {'end': 117.703, 'text': "Then we'll write the code in order to implement the model.", 'start': 115.561, 'duration': 2.142}, {'end': 121.245, 'text': 'And once this is done, we will train our model on the training data.', 'start': 118.223, 'duration': 3.022}, {'end': 122.767, 'text': "We'll calculate the error.", 'start': 121.646, 'duration': 1.121}, {'end': 126.87, 'text': 'The error is nothing but your difference between the model output and the actual output.', 'start': 123.327, 'duration': 3.543}, {'end': 128.994, 'text': "and we'll try to reduce this error.", 'start': 127.594, 'duration': 1.4}, {'end': 134.395, 'text': "And once this error becomes minimum, we'll make prediction on the test data and we'll calculate the final accuracy.", 'start': 129.494, 'duration': 4.901}], 'summary': 'Using tensorflow to implement deep learning models, training and testing with prediction and accuracy calculation.', 'duration': 30.063, 'max_score': 104.332, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU104332.jpg'}, {'end': 248.679, 'src': 'embed', 'start': 213.642, 'weight': 4, 'content': [{'end': 218.024, 'text': "Then we'll focus on what is the motivation behind neural networks and what exactly are neural networks.", 'start': 213.642, 'duration': 4.382}, {'end': 222.125, 'text': "After that we'll understand what is single layer perceptron and multi-layer perceptron.", 'start': 218.724, 'duration': 3.401}, {'end': 226.647, 'text': "And then I'll tell you how to implement the use case that I was talking about in the beginning.", 'start': 222.905, 'duration': 3.742}, {'end': 230.088, 'text': "And finally we'll discuss various applications of neural networks.", 'start': 227.227, 'duration': 2.861}, {'end': 232.369, 'text': 'So I hope we all are clear with the agenda.', 'start': 230.808, 'duration': 1.561}, {'end': 234.61, 'text': 'If you have any questions or doubts you can go ahead and ask me.', 'start': 232.389, 'duration': 2.221}, {'end': 237.513, 'text': "So there's a question from Shivani.", 'start': 236.152, 'duration': 1.361}, {'end': 240.434, 'text': "she's asking do we need any prior knowledge on TensorFlow??", 'start': 237.513, 'duration': 2.921}, {'end': 241.755, 'text': 'Definitely, Shivani.', 'start': 240.834, 'duration': 0.921}, {'end': 246.097, 'text': "there is a separate tutorial on TensorFlow where I've covered all the fundamentals of TensorFlow.", 'start': 241.755, 'duration': 4.342}, {'end': 248.679, 'text': 'So you could go through it if you want to understand the code.', 'start': 246.497, 'duration': 2.182}], 'summary': 'Introduction to neural networks, single/multi-layer perceptron, use case implementation, and diverse applications discussed.', 'duration': 35.037, 'max_score': 213.642, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU213642.jpg'}], 'start': 0.089, 'title': 'Neural networks in banknote authentication', 'summary': 'Covers the fundamentals of neural networks and their application in authenticating banknotes using artificial neural networks, extracting features from images, and classifying them as real or fake. it involves tensorflow implementation, error calculation, and accuracy prediction.', 'chapters': [{'end': 87.401, 'start': 0.089, 'title': 'Neural networks fundamentals & banknote authentication', 'summary': 'Covers the fundamentals of neural networks and their application in authenticating banknotes using artificial neural networks, extracting features from images, and classifying them as real or fake, with the dataset containing features like wavelet transformed image variants, skewness, courtesies, and entropy, and labels indicating real or fake notes.', 'duration': 87.312, 'highlights': ["The dataset contains features like wavelet transformed image variants, skewness, courtesies, and entropy, and labels indicating real or fake notes, with '1' representing real notes and '0' representing fake notes.", 'The chapter covers the fundamentals of neural networks and their application in authenticating banknotes using artificial neural networks and extracting features from images.', 'Data were extracted from images of genuine and forged banknotes, and wavelet transform tools were used to extract features from those images.']}, {'end': 267.227, 'start': 88.442, 'title': 'Neural network fundamentals', 'summary': 'Covers the process of implementing a neural network model using tensorflow, including data set reading, feature and label definition, encoding of dependent variable, model implementation, training, error calculation, and accuracy prediction.', 'duration': 178.785, 'highlights': ['The process involves implementing a neural network model using TensorFlow, including data set reading, feature and label definition, encoding of dependent variable, model implementation, training, error calculation, and accuracy prediction. The chapter covers the process of implementing a neural network model using TensorFlow, including data set reading, feature and label definition, encoding of dependent variable, model implementation, training, error calculation, and accuracy prediction.', 'Explanation of code and fundamentals of neural networks will be covered at the end of the session. Explanation of code and fundamentals of neural networks will be covered at the end of the session.', 'The agenda includes understanding the motivation behind neural networks, single layer perceptron, multi-layer perceptron, use case implementation, and various applications of neural networks. The agenda includes understanding the motivation behind neural networks, single layer perceptron, multi-layer perceptron, use case implementation, and various applications of neural networks.', 'Prior knowledge on TensorFlow is required, and there is a separate tutorial covering the basics of TensorFlow. Prior knowledge on TensorFlow is required, and there is a separate tutorial covering the basics of TensorFlow.']}], 'duration': 267.138, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU89.jpg', 'highlights': ['The chapter covers the fundamentals of neural networks and their application in authenticating banknotes using artificial neural networks and extracting features from images.', "The dataset contains features like wavelet transformed image variants, skewness, courtesies, and entropy, and labels indicating real or fake notes, with '1' representing real notes and '0' representing fake notes.", 'The process involves implementing a neural network model using TensorFlow, including data set reading, feature and label definition, encoding of dependent variable, model implementation, training, error calculation, and accuracy prediction.', 'Data were extracted from images of genuine and forged banknotes, and wavelet transform tools were used to extract features from those images.', 'The agenda includes understanding the motivation behind neural networks, single layer perceptron, multi-layer perceptron, use case implementation, and various applications of neural networks.', 'Prior knowledge on TensorFlow is required, and there is a separate tutorial covering the basics of TensorFlow.']}, {'end': 715.215, 'segs': [{'end': 368.182, 'src': 'embed', 'start': 340.645, 'weight': 0, 'content': [{'end': 343.266, 'text': 'You cannot program them to perform a specific task.', 'start': 340.645, 'duration': 2.621}, {'end': 346.207, 'text': 'They will learn from their examples, from their experience.', 'start': 343.846, 'duration': 2.361}, {'end': 350.128, 'text': "So you don't need to provide all the instructions to perform a specific task.", 'start': 346.687, 'duration': 3.441}, {'end': 353.149, 'text': 'And your network will learn on its own with its own experience.', 'start': 350.348, 'duration': 2.801}, {'end': 355.87, 'text': 'All right, so this is what basically neural network does.', 'start': 353.809, 'duration': 2.061}, {'end': 362.152, 'text': "So, even if you don't know how to solve a problem, you can train your network in such a way that with experience,", 'start': 356.704, 'duration': 5.448}, {'end': 364.055, 'text': 'it can actually learn how to solve the problem.', 'start': 362.152, 'duration': 1.903}, {'end': 368.182, 'text': 'So that was a major reason why neural networks came into existence.', 'start': 364.796, 'duration': 3.386}], 'summary': 'Neural networks learn from experience, solving problems without explicit instructions. this led to their creation.', 'duration': 27.537, 'max_score': 340.645, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU340645.jpg'}, {'end': 437.025, 'src': 'embed', 'start': 409.881, 'weight': 2, 'content': [{'end': 414.103, 'text': "So we'll move forward and we'll understand what is the motivation behind neural networks.", 'start': 409.881, 'duration': 4.222}, {'end': 419.586, 'text': 'So these neural networks are basically inspired by neurons, which are nothing but your brain cells,', 'start': 414.923, 'duration': 4.663}, {'end': 422.407, 'text': 'and the exact working of the human brain is still a mystery, though.', 'start': 419.586, 'duration': 2.821}, {'end': 427.397, 'text': "So as I've told you earlier as well that neural networks work like human brain and so the name.", 'start': 423.274, 'duration': 4.123}, {'end': 435.024, 'text': 'And similar to a newborn human baby as he or she learns from his or her experience, we want a network to do that as well.', 'start': 428.438, 'duration': 6.586}, {'end': 437.025, 'text': 'But we want it to do very quickly.', 'start': 435.324, 'duration': 1.701}], 'summary': 'Neural networks mimic the human brain and aim to learn quickly.', 'duration': 27.144, 'max_score': 409.881, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU409881.jpg'}, {'end': 524.797, 'src': 'embed', 'start': 468.849, 'weight': 1, 'content': [{'end': 472.272, 'text': "Now let's understand what exactly are artificial neural networks.", 'start': 468.849, 'duration': 3.423}, {'end': 480.029, 'text': 'It is basically a computing system that is designed to simulate the way the human brain analyzes and process the information.', 'start': 473.305, 'duration': 6.724}, {'end': 487.894, 'text': 'Artificial neural networks has self-learning capabilities that enable it to produce better results as more data becomes available.', 'start': 480.71, 'duration': 7.184}, {'end': 490.716, 'text': 'So if you train your network on more data, it should be more accurate.', 'start': 487.914, 'duration': 2.802}, {'end': 493.678, 'text': 'So these neural networks, they actually learn by example.', 'start': 491.396, 'duration': 2.282}, {'end': 497.72, 'text': 'And you can configure your neural network for specific applications.', 'start': 494.378, 'duration': 3.342}, {'end': 501.863, 'text': 'It can be pattern recognition or it can be data classification, anything like that, all right?', 'start': 497.8, 'duration': 4.063}, {'end': 506.162, 'text': 'So because of neural networks we see a lot of new technology has evolved.', 'start': 502.759, 'duration': 3.403}, {'end': 513.888, 'text': 'From translating web pages to other languages, to having a virtual assistant to order groceries online, to conversing with chatbots.', 'start': 506.842, 'duration': 7.046}, {'end': 517.551, 'text': 'All of these things are possible because of neural networks.', 'start': 514.467, 'duration': 3.084}, {'end': 524.797, 'text': 'So in a nutshell, if I need to tell you, artificial neural network is nothing but a network of various artificial neurons.', 'start': 518.472, 'duration': 6.325}], 'summary': 'Artificial neural networks simulate human brain, with self-learning capabilities, enabling better results with more data, leading to various technological advancements.', 'duration': 55.948, 'max_score': 468.849, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU468849.jpg'}, {'end': 661.809, 'src': 'heatmap', 'start': 563.027, 'weight': 5, 'content': [{'end': 572.653, 'text': 'but still it can identify certain features of the dogs that we have trained on and it can match those features with the dog that is there in this particular image and it can identify that dog.', 'start': 563.027, 'duration': 9.626}, {'end': 575.574, 'text': 'So this happens all because of neural networks.', 'start': 573.133, 'duration': 2.441}, {'end': 579.196, 'text': 'So this is just an example to show you how important our neural networks.', 'start': 575.894, 'duration': 3.302}, {'end': 582.999, 'text': 'Now I know you all must be thinking how neural networks work.', 'start': 579.797, 'duration': 3.202}, {'end': 587.001, 'text': "So for that we'll move forward and understand how it actually works.", 'start': 583.699, 'duration': 3.302}, {'end': 592.812, 'text': "So over here I'll begin by first explaining a single artificial neuron that is called as perceptron.", 'start': 588.001, 'duration': 4.811}, {'end': 595.698, 'text': 'So this is an example of a perceptron.', 'start': 593.618, 'duration': 2.08}, {'end': 599.739, 'text': 'Over here we have multiple inputs, x1, x2, dash, dash, dash, till xn.', 'start': 596.258, 'duration': 3.481}, {'end': 605.08, 'text': 'And we have corresponding weights as well, w1 for x1, w2 for x2, similarly, wn for xn.', 'start': 600.259, 'duration': 4.821}, {'end': 609.541, 'text': 'Then what happens, we calculated the weighted sum of these inputs.', 'start': 605.7, 'duration': 3.841}, {'end': 613.122, 'text': 'And after doing that, we pass it through an activation function.', 'start': 610.181, 'duration': 2.941}, {'end': 616.862, 'text': 'This activation function is nothing but it provides a threshold value.', 'start': 613.682, 'duration': 3.18}, {'end': 620.863, 'text': "So above that value, my neuron will fire, else it won't fire.", 'start': 617.382, 'duration': 3.481}, {'end': 623.394, 'text': 'So this is basically an artificial neuron.', 'start': 621.513, 'duration': 1.881}, {'end': 630.719, 'text': 'So when I talk about a neural network, it involves a lot of these artificial neurons with their own activation function and their processing element.', 'start': 623.514, 'duration': 7.205}, {'end': 637.703, 'text': "Now we'll move forward and we'll actually understand various modes of this perceptron or single artificial neuron.", 'start': 632.179, 'duration': 5.524}, {'end': 639.847, 'text': 'So there are two modes in a perceptron.', 'start': 638.365, 'duration': 1.482}, {'end': 641.328, 'text': 'One is training, another is using mode.', 'start': 639.927, 'duration': 1.401}, {'end': 646.253, 'text': 'In training mode, the neuron can be trained to fire for particular input patterns,', 'start': 641.809, 'duration': 4.444}, {'end': 653.1, 'text': "which means that we'll actually train our neuron to fire on certain set of inputs and to not fire on the other set of inputs.", 'start': 646.253, 'duration': 6.847}, {'end': 654.782, 'text': "That's what basically training mode is.", 'start': 653.42, 'duration': 1.362}, {'end': 661.809, 'text': 'When I talk about using mode, it means that when a taught input pattern is detected at the input, its associated output becomes the current output.', 'start': 655.202, 'duration': 6.607}], 'summary': 'Neural networks identify dog features, using perceptron as an artificial neuron for training and using modes.', 'duration': 42.053, 'max_score': 563.027, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU563027.jpg'}, {'end': 653.1, 'src': 'embed', 'start': 623.514, 'weight': 7, 'content': [{'end': 630.719, 'text': 'So when I talk about a neural network, it involves a lot of these artificial neurons with their own activation function and their processing element.', 'start': 623.514, 'duration': 7.205}, {'end': 637.703, 'text': "Now we'll move forward and we'll actually understand various modes of this perceptron or single artificial neuron.", 'start': 632.179, 'duration': 5.524}, {'end': 639.847, 'text': 'So there are two modes in a perceptron.', 'start': 638.365, 'duration': 1.482}, {'end': 641.328, 'text': 'One is training, another is using mode.', 'start': 639.927, 'duration': 1.401}, {'end': 646.253, 'text': 'In training mode, the neuron can be trained to fire for particular input patterns,', 'start': 641.809, 'duration': 4.444}, {'end': 653.1, 'text': "which means that we'll actually train our neuron to fire on certain set of inputs and to not fire on the other set of inputs.", 'start': 646.253, 'duration': 6.847}], 'summary': 'Neural networks involve artificial neurons with distinct training and using modes.', 'duration': 29.586, 'max_score': 623.514, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU623514.jpg'}], 'start': 267.548, 'title': 'Importance of neural networks', 'summary': 'Discusses the need for neural networks, comparing the limitations of conventional computers with the advantages of neural networks, emphasizing their ability to learn and adapt, and exploring their historical context and technological advancements. it also delves into understanding artificial neural networks, highlighting their self-learning capabilities, applications in technology, and the working of a single artificial neuron.', 'chapters': [{'end': 468.089, 'start': 267.548, 'title': 'Importance of neural networks', 'summary': 'Discusses the need for neural networks by comparing the limitations of conventional computers with the advantages of neural networks, highlighting their ability to learn from examples and adapt to solve problems, with a historical context and technological advancements contributing to their feasibility.', 'duration': 200.541, 'highlights': ['Neural networks enable computers to learn from examples and experiences, allowing them to solve problems for which specific instructions are unknown, expanding problem-solving capabilities beyond what is already understood. Neural networks enable computers to solve problems without the need for specific instructions.', "The term 'neural network' was coined in 1943, but technological limitations hindered their practical implementation until the availability of GPUs and the generation of large amounts of data from sources such as the Internet of Things and social media. Technological advancements, including GPUs and data generation from various sources, have made practical implementation of neural networks feasible.", "Neural networks are motivated by the functioning of neurons in the human brain, aiming to replicate the learning process of human beings, albeit at a faster pace. Neural networks are inspired by the human brain's neurons and aim to replicate human learning processes in a quicker manner."]}, {'end': 715.215, 'start': 468.849, 'title': 'Understanding artificial neural networks', 'summary': 'Explores artificial neural networks, emphasizing their self-learning capabilities, applications in technology, and the importance of neural networks through examples, and also delves into the working of a single artificial neuron, including its modes and activation functions.', 'duration': 246.366, 'highlights': ['Artificial neural networks have self-learning capabilities that enable them to produce better results as more data becomes available, making them more accurate with increased data. Artificial neural networks possess self-learning capabilities, leading to improved accuracy as more data is utilized for training.', 'Applications of neural networks include translating web pages, virtual assistants for online shopping, and conversing with chatbots, showcasing the impact of neural networks on various technological advancements. Neural networks have facilitated technological advancements such as web page translation, virtual assistants for online shopping, and chatbot interactions.', 'The importance of neural networks is demonstrated through an example where a machine trained on specific dogs can still identify certain features of an untrained dog, showcasing the significance of neural networks in pattern recognition. A demonstration highlights the significance of neural networks in pattern recognition, as a machine trained on specific dogs can still identify features of an untrained dog.', 'The chapter delves into the working of a single artificial neuron, explaining the perceptron and its components, including inputs, weights, weighted sum calculation, and activation function, providing a foundational understanding of the building block of neural networks. A detailed explanation of the perceptron, including its components such as inputs, weights, weighted sum calculation, and activation function, provides a foundational understanding of the building block of neural networks.', 'The chapter also discusses the modes of a perceptron, involving training and using modes, and explores various activation functions available for artificial neurons, providing insights into the learning and functioning of single artificial neurons within a network. The chapter explores the modes of a perceptron, encompassing training and using modes, and discusses various activation functions available for artificial neurons, offering insights into their learning and functioning within a network.']}], 'duration': 447.667, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU267548.jpg', 'highlights': ['Neural networks enable computers to solve problems without the need for specific instructions.', 'Technological advancements, including GPUs and data generation from various sources, have made practical implementation of neural networks feasible.', "Neural networks are inspired by the human brain's neurons and aim to replicate human learning processes in a quicker manner.", 'Artificial neural networks possess self-learning capabilities, leading to improved accuracy as more data is utilized for training.', 'Neural networks have facilitated technological advancements such as web page translation, virtual assistants for online shopping, and chatbot interactions.', 'A demonstration highlights the significance of neural networks in pattern recognition, as a machine trained on specific dogs can still identify features of an untrained dog.', 'A detailed explanation of the perceptron, including its components such as inputs, weights, weighted sum calculation, and activation function, provides a foundational understanding of the building block of neural networks.', 'The chapter explores the modes of a perceptron, encompassing training and using modes, and discusses various activation functions available for artificial neurons, offering insights into their learning and functioning within a network.']}, {'end': 1025.954, 'segs': [{'end': 771.534, 'src': 'embed', 'start': 735.051, 'weight': 0, 'content': [{'end': 738.654, 'text': "So that's why I've chosen this particular analogy so that all of you can relate to it.", 'start': 735.051, 'duration': 3.603}, {'end': 740.135, 'text': 'All right jokes apart.', 'start': 739.394, 'duration': 0.741}, {'end': 741.195, 'text': 'So fine guys.', 'start': 740.615, 'duration': 0.58}, {'end': 748.5, 'text': "So there's a beer festival happening near your house and you want to badly go there but your decision actually depends on three factors.", 'start': 741.235, 'duration': 7.265}, {'end': 751.162, 'text': 'First is how is the weather whether it is good or bad.', 'start': 749.061, 'duration': 2.101}, {'end': 754.084, 'text': 'Second is your wife or husband is going with you or not.', 'start': 751.822, 'duration': 2.262}, {'end': 757.086, 'text': 'And the third one is any public transport is available.', 'start': 754.504, 'duration': 2.582}, {'end': 761.368, 'text': 'So on these three factors your decision will depend whether you will go or not.', 'start': 757.686, 'duration': 3.682}, {'end': 769.913, 'text': "So we'll consider these three factors as inputs to our perceptron and we'll consider our decision of going or not going to the beer festival as our output.", 'start': 761.909, 'duration': 8.004}, {'end': 771.534, 'text': 'So let us move forward with that.', 'start': 770.374, 'duration': 1.16}], 'summary': 'An analogy of decision-making using weather, company, and transport as factors for a perceptron model.', 'duration': 36.483, 'max_score': 735.051, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU735051.jpg'}], 'start': 715.515, 'title': 'Beer festival decision-making process and model', 'summary': 'Explores the decision-making process and model for attending a beer festival, incorporating three factors - weather, company, and transportation - that influence the decision, and demonstrates the use of weights and threshold to determine the decision in various scenarios.', 'chapters': [{'end': 789.119, 'start': 715.515, 'title': 'Beer festival decision-making process', 'summary': 'Explains the decision-making process using an analogy of attending a beer festival, where three factors - weather, company, and transportation - influence the decision, reflecting inputs to a perceptron model.', 'duration': 73.604, 'highlights': ['The decision-making process for attending a beer festival is explained using an analogy, with three factors - weather, company, and transportation - determining the decision.', "The analogy of attending a beer festival is used to relate the decision-making process, appealing to the audience's interest in beer and humor.", 'Inputs to a perceptron model are represented by the three factors influencing the decision to attend the beer festival.']}, {'end': 1025.954, 'start': 789.84, 'title': 'Beer festival decision model', 'summary': 'Explains a decision model for going to a beer festival based on weather, company, and public transport, assigning weights to factors and using a threshold to determine the decision, with a demonstration of different scenarios.', 'duration': 236.114, 'highlights': ['The decision to attend the beer festival is based on weather, company, and public transport, with a weight of 6 assigned to weather and 2 to each of the other factors. The decision to attend the beer festival is determined by the weather, company, and public transport, with a weight of 6 assigned to weather and 2 to each of the other factors, impacting the decision-making process.', 'A threshold value of 5 is set, where if the weighted sum of the inputs is greater than 5, the decision to attend the beer festival is made. A threshold value of 5 is set, where if the weighted sum of the inputs is greater than 5, the decision to attend the beer festival is made, demonstrating the significance of the weighted sum in the decision-making process.', 'Demonstration of the decision-making process in different scenarios by varying the threshold, showing how the decision changes based on the threshold value and the inputs. Demonstration of the decision-making process in different scenarios by varying the threshold, showing how the decision changes based on the threshold value and the inputs, providing insights into the impact of threshold values on the decision outcome.']}], 'duration': 310.439, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU715515.jpg', 'highlights': ['The decision-making process for attending a beer festival is explained using an analogy, with three factors - weather, company, and transportation - determining the decision.', 'Demonstration of the decision-making process in different scenarios by varying the threshold, showing how the decision changes based on the threshold value and the inputs, providing insights into the impact of threshold values on the decision outcome.', 'The decision to attend the beer festival is based on weather, company, and public transport, with a weight of 6 assigned to weather and 2 to each of the other factors, impacting the decision-making process.']}, {'end': 1475.672, 'segs': [{'end': 1123.707, 'src': 'embed', 'start': 1084.318, 'weight': 3, 'content': [{'end': 1087.641, 'text': 'All right, and in real life problems are not that easy.', 'start': 1084.318, 'duration': 3.323}, {'end': 1091.004, 'text': 'They are very very complex problems that we actually face.', 'start': 1087.961, 'duration': 3.043}, {'end': 1094.727, 'text': 'So in order to solve those problems a single neuron is definitely not enough.', 'start': 1091.484, 'duration': 3.243}, {'end': 1102.193, 'text': "So we need networks of neuron and that's where artificial neural network or you can say multi-layer perception comes into the picture.", 'start': 1094.787, 'duration': 7.406}, {'end': 1103.755, 'text': 'Now, let us discuss that.', 'start': 1102.734, 'duration': 1.021}, {'end': 1106.801, 'text': 'multi-layer perceptron or artificial neural network.', 'start': 1104.52, 'duration': 2.281}, {'end': 1109.802, 'text': 'So this is how an artificial neural network actually looks like.', 'start': 1107.441, 'duration': 2.361}, {'end': 1113.223, 'text': 'So over here we have multiple neurons present in different layers.', 'start': 1110.162, 'duration': 3.061}, {'end': 1115.764, 'text': 'The first layer is always your input layer.', 'start': 1113.743, 'duration': 2.021}, {'end': 1118.605, 'text': 'This is where you actually feed in all of your inputs.', 'start': 1116.224, 'duration': 2.381}, {'end': 1123.707, 'text': 'Then we have the first hidden layer, then we have second hidden layer, and then we have the output layer.', 'start': 1119.165, 'duration': 4.542}], 'summary': 'Real-life problems require complex solutions; artificial neural networks with multiple layers are necessary for addressing them.', 'duration': 39.389, 'max_score': 1084.318, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU1084318.jpg'}, {'end': 1200.142, 'src': 'embed', 'start': 1173.362, 'weight': 5, 'content': [{'end': 1176.785, 'text': "So over here I'll take an example of image recognition using neural networks.", 'start': 1173.362, 'duration': 3.423}, {'end': 1181.308, 'text': 'So over here what happens, we feed in a lot of images to our input layer.', 'start': 1177.425, 'duration': 3.883}, {'end': 1185.911, 'text': 'Now this input layer will actually detect the patterns of local contrast.', 'start': 1181.908, 'duration': 4.003}, {'end': 1189.294, 'text': "And then we'll feed that to the next layer which is hidden layer one.", 'start': 1186.432, 'duration': 2.862}, {'end': 1193.217, 'text': 'So in this hidden layer one, the face features will be recognized.', 'start': 1189.894, 'duration': 3.323}, {'end': 1196.539, 'text': 'We recognize eyes, nose, ears, things like that.', 'start': 1193.637, 'duration': 2.902}, {'end': 1200.142, 'text': 'And then that will be again fed as input to the next hidden layer.', 'start': 1197.079, 'duration': 3.063}], 'summary': 'Image recognition using neural networks: input layer detects local contrast patterns, hidden layer recognizes facial features.', 'duration': 26.78, 'max_score': 1173.362, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU1173362.jpg'}, {'end': 1254.126, 'src': 'heatmap', 'start': 1209.346, 'weight': 0.744, 'content': [{'end': 1215.308, 'text': 'So if you notice here with every layer we are trying to get a more abstract version or the generalized version of the input.', 'start': 1209.346, 'duration': 5.962}, {'end': 1219.05, 'text': 'So this is how basically an artificial neural network how it works.', 'start': 1215.848, 'duration': 3.202}, {'end': 1223.672, 'text': "All right, and there's a lot of training and learning which is involved that I'll show you now.", 'start': 1219.63, 'duration': 4.042}, {'end': 1225.432, 'text': 'training a neural network.', 'start': 1224.349, 'duration': 1.083}, {'end': 1227.296, 'text': 'So how we actually train a neural network.', 'start': 1225.452, 'duration': 1.844}, {'end': 1231.847, 'text': 'So basically the most common algorithm for training a network is called back propagation.', 'start': 1227.938, 'duration': 3.909}, {'end': 1238.76, 'text': 'So what happens in back propagation, after the weighted sum of inputs and passing through an activation function and getting the output?', 'start': 1232.677, 'duration': 6.083}, {'end': 1242.041, 'text': 'we compare that output to the actual output that we already know.', 'start': 1238.76, 'duration': 3.281}, {'end': 1247.183, 'text': 'we figure out how much is the difference, we calculate the error and, based on that error, what we do.', 'start': 1242.041, 'duration': 5.142}, {'end': 1251.545, 'text': "we propagate backwards and we'll see what happens when we change the weight.", 'start': 1247.183, 'duration': 4.362}, {'end': 1254.126, 'text': 'will the error decrease or will it increase?', 'start': 1251.545, 'duration': 2.581}], 'summary': 'Neural network training involves back propagation to minimize error and adjust weights.', 'duration': 44.78, 'max_score': 1209.346, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU1209346.jpg'}, {'end': 1254.126, 'src': 'embed', 'start': 1227.938, 'weight': 0, 'content': [{'end': 1231.847, 'text': 'So basically the most common algorithm for training a network is called back propagation.', 'start': 1227.938, 'duration': 3.909}, {'end': 1238.76, 'text': 'So what happens in back propagation, after the weighted sum of inputs and passing through an activation function and getting the output?', 'start': 1232.677, 'duration': 6.083}, {'end': 1242.041, 'text': 'we compare that output to the actual output that we already know.', 'start': 1238.76, 'duration': 3.281}, {'end': 1247.183, 'text': 'we figure out how much is the difference, we calculate the error and, based on that error, what we do.', 'start': 1242.041, 'duration': 5.142}, {'end': 1251.545, 'text': "we propagate backwards and we'll see what happens when we change the weight.", 'start': 1247.183, 'duration': 4.362}, {'end': 1254.126, 'text': 'will the error decrease or will it increase?', 'start': 1251.545, 'duration': 2.581}], 'summary': 'Back propagation algorithm adjusts weights based on error to decrease error.', 'duration': 26.188, 'max_score': 1227.938, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU1227938.jpg'}, {'end': 1425.096, 'src': 'embed', 'start': 1393.8, 'weight': 2, 'content': [{'end': 1396.181, 'text': 'So how we determine that all right fine guys.', 'start': 1393.8, 'duration': 2.381}, {'end': 1399.621, 'text': 'This is how basically a computer decide whether it has to increase the weight or decrease away.', 'start': 1396.201, 'duration': 3.42}, {'end': 1402.782, 'text': 'So what happens here? This is a graph of square error versus weight.', 'start': 1400.042, 'duration': 2.74}, {'end': 1406.803, 'text': 'So what here what happens suppose your square error is somewhere here.', 'start': 1403.462, 'duration': 3.341}, {'end': 1408.764, 'text': 'and your computer.', 'start': 1407.603, 'duration': 1.161}, {'end': 1413.946, 'text': 'it starts increasing the weight in order to reduce the square error and it notices that whenever it increases the weight,', 'start': 1408.764, 'duration': 5.182}, {'end': 1415.207, 'text': 'square error is actually decreasing.', 'start': 1413.946, 'duration': 1.261}, {'end': 1419.429, 'text': "So it'll keep on increasing until the square error reaches a minimum value.", 'start': 1415.907, 'duration': 3.522}, {'end': 1425.096, 'text': 'And after that, when it tries to still increase the weight, the square error will increase.', 'start': 1420.373, 'duration': 4.723}], 'summary': 'Computer increases weight to minimize square error in a graph.', 'duration': 31.296, 'max_score': 1393.8, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU1393800.jpg'}], 'start': 1025.994, 'title': 'Neural network basics and back propagation', 'summary': 'Covers the concept of assigning weights in a perceptron, the functionality of a multi-layer perceptron, and the process of training a neural network using back propagation, involving iterative weight adjustments to minimize error.', 'chapters': [{'end': 1208.785, 'start': 1025.994, 'title': 'Neural network basics', 'summary': 'Explains the concept of assigning weights and prioritizing inputs in a perceptron and delves into the structure and functionality of a multi-layer perceptron, emphasizing the need for networks of neurons in solving complex real-life problems.', 'duration': 182.791, 'highlights': ['Artificial neural network structure The chapter provides an overview of the structure of an artificial neural network, consisting of multiple layers with different functions and the flow of inputs and outputs through these layers.', 'Importance of networks of neurons in solving complex problems It emphasizes the need for networks of neurons, such as multi-layer perceptrons, in solving complex real-life problems due to the inadequacy of a single neuron.', 'Example of image recognition using neural networks An example of image recognition using neural networks is explained, detailing the process of feeding images to the input layer, detecting patterns of local contrast, and recognizing facial features in the hidden layers to properly identify a face.']}, {'end': 1475.672, 'start': 1209.346, 'title': 'Back propagation in neural networks', 'summary': 'Explains the process of training an artificial neural network using back propagation, involving iterative weight adjustments to minimize error, illustrated with a specific example and a graphical representation of the error-weight relationship.', 'duration': 266.326, 'highlights': ['The process of training a neural network involves back propagation, where iterative weight adjustments are made to minimize error by comparing the model output with the desired output until the error becomes minimum after a lot of iterations. training process, back propagation, iterative weight adjustments, minimizing error, comparison of model output and desired output, many iterations', 'An example is presented to illustrate the back propagation process, involving the initialization of weight values, calculation of errors, and iterative adjustments of the weight to minimize the error, demonstrating how the computer learns to produce the desired output. example illustration, weight initialization, error calculation, iterative weight adjustments, learning process of the computer', 'A graphical representation of the square error versus weight is used to explain how the computer determines whether to increase or decrease the weight, showing that the weight value where the square error is minimized is the final weight value. graphical representation, error-weight relationship, determining weight adjustment, minimizing square error, final weight value']}], 'duration': 449.678, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU1025993.jpg', 'highlights': ['The process of training a neural network involves back propagation, where iterative weight adjustments are made to minimize error by comparing the model output with the desired output until the error becomes minimum after a lot of iterations.', 'An example is presented to illustrate the back propagation process, involving the initialization of weight values, calculation of errors, and iterative adjustments of the weight to minimize the error, demonstrating how the computer learns to produce the desired output.', 'A graphical representation of the square error versus weight is used to explain how the computer determines whether to increase or decrease the weight, showing that the weight value where the square error is minimized is the final weight value.', 'Artificial neural network structure The chapter provides an overview of the structure of an artificial neural network, consisting of multiple layers with different functions and the flow of inputs and outputs through these layers.', 'Importance of networks of neurons in solving complex problems It emphasizes the need for networks of neurons, such as multi-layer perceptrons, in solving complex real-life problems due to the inadequacy of a single neuron.', 'Example of image recognition using neural networks An example of image recognition using neural networks is explained, detailing the process of feeding images to the input layer, detecting patterns of local contrast, and recognizing facial features in the hidden layers to properly identify a face.']}, {'end': 2193.791, 'segs': [{'end': 1529.246, 'src': 'embed', 'start': 1496.336, 'weight': 4, 'content': [{'end': 1499.778, 'text': 'So over here, what we do, we import the first important libraries which are required.', 'start': 1496.336, 'duration': 3.442}, {'end': 1505.722, 'text': 'Matplotlib is used for visualization TensorFlow, we know, in order to implement the neural networks, NumPy for arrays,', 'start': 1500.018, 'duration': 5.704}, {'end': 1510.386, 'text': 'Pandas for reading the data set, similarly sklearn for label encoding, as well as for shuffling,', 'start': 1505.722, 'duration': 4.664}, {'end': 1513.428, 'text': 'and also to split the data set into training and testing paths.', 'start': 1510.386, 'duration': 3.042}, {'end': 1514.709, 'text': 'All right, fine guys.', 'start': 1514.068, 'duration': 0.641}, {'end': 1519.274, 'text': "So we'll begin by first reading the data set as I've told you earlier as well when I was explaining the steps.", 'start': 1515.029, 'duration': 4.245}, {'end': 1524.6, 'text': "So what I'll do, I'll use Pandas in order to read the CSV file which has the data set.", 'start': 1519.855, 'duration': 4.745}, {'end': 1526.803, 'text': "After that, I'll define features and labels.", 'start': 1525.261, 'duration': 1.542}, {'end': 1529.246, 'text': 'So X will be my feature and Y will contain my label.', 'start': 1526.843, 'duration': 2.403}], 'summary': 'Imported essential libraries for data visualization, neural network implementation, and data manipulation using numpy, pandas, matplotlib, tensorflow, and sklearn.', 'duration': 32.91, 'max_score': 1496.336, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU1496336.jpg'}, {'end': 1848.034, 'src': 'embed', 'start': 1816.879, 'weight': 1, 'content': [{'end': 1818.5, 'text': "So let's go ahead and execute this, guys.", 'start': 1816.879, 'duration': 1.621}, {'end': 1822.372, 'text': 'All right, so training is done.', 'start': 1821.291, 'duration': 1.081}, {'end': 1825.375, 'text': 'And this is the graph we have got for accuracy versus epoch.', 'start': 1822.913, 'duration': 2.462}, {'end': 1828.938, 'text': 'This is accuracy, y-axis represents accuracy, whereas this is epochs.', 'start': 1825.795, 'duration': 3.143}, {'end': 1832.522, 'text': 'We have taken 100 epochs, and our accuracy has reached somewhere around 99%.', 'start': 1828.958, 'duration': 3.564}, {'end': 1839.448, 'text': 'So with every epoch, it is actually increasing, apart from a couple of instances, it is actually keep on increasing.', 'start': 1832.522, 'duration': 6.926}, {'end': 1843.272, 'text': "So the more data you train your model on, it'll be more accurate.", 'start': 1839.868, 'duration': 3.404}, {'end': 1844.572, 'text': 'Let me just close it.', 'start': 1843.772, 'duration': 0.8}, {'end': 1848.034, 'text': 'So now the model has also been saved where I wanted it to be.', 'start': 1845.073, 'duration': 2.961}], 'summary': 'Model training achieved 99% accuracy in 100 epochs.', 'duration': 31.155, 'max_score': 1816.879, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU1816879.jpg'}, {'end': 1895.951, 'src': 'embed', 'start': 1871.463, 'weight': 0, 'content': [{'end': 1878.345, 'text': 'So all the values in the row of 754 and 768 will be fed to our model and our model will make prediction on that.', 'start': 1871.463, 'duration': 6.882}, {'end': 1880.166, 'text': 'So let us go ahead and run this.', 'start': 1878.865, 'duration': 1.301}, {'end': 1887.368, 'text': "So when I'm restoring my model, it seems that my model is hundred percent accurate for the values that I fed it.", 'start': 1882.106, 'duration': 5.262}, {'end': 1895.951, 'text': "So, whatever values that I have actually given as input to my model, it has correctly identified its class, whether it's a fake note or a real note,", 'start': 1887.888, 'duration': 8.063}], 'summary': 'Model achieves 100% accuracy in identifying fake or real banknotes.', 'duration': 24.488, 'max_score': 1871.463, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU1871463.jpg'}, {'end': 1973.796, 'src': 'embed', 'start': 1942.411, 'weight': 2, 'content': [{'end': 1948.374, 'text': 'And currently the research is mostly on modeling parts of human body and recognizing diseases from various scans.', 'start': 1942.411, 'duration': 5.963}, {'end': 1952.317, 'text': 'For example it can be cardiograms, CAT scans, ultrasonic scans, et cetera.', 'start': 1948.414, 'duration': 3.903}, {'end': 1956.706, 'text': 'And currently the research is going mostly on two major areas.', 'start': 1953.084, 'duration': 3.622}, {'end': 1959.407, 'text': 'First is modeling and diagnosing the cardiovascular system.', 'start': 1956.766, 'duration': 2.641}, {'end': 1964.15, 'text': 'So neural networks are used experimentally to model the human cardiovascular system.', 'start': 1960.168, 'duration': 3.982}, {'end': 1973.796, 'text': 'Diagnosis can be achieved by building a model of the cardiovascular system of an individual and comparing it with the real time physiological measurements taken from the patient.', 'start': 1964.929, 'duration': 8.867}], 'summary': 'Research focuses on modeling and diagnosing the cardiovascular system using neural networks and real-time physiological measurements.', 'duration': 31.385, 'max_score': 1942.411, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU1942411.jpg'}, {'end': 2094.659, 'src': 'embed', 'start': 2067.661, 'weight': 3, 'content': [{'end': 2073.165, 'text': 'and was trained using back propagation to assist the marketing control of airline seat allocation.', 'start': 2067.661, 'duration': 5.504}, {'end': 2077.068, 'text': 'So it has wide applications in marketing as well.', 'start': 2073.866, 'duration': 3.202}, {'end': 2079.71, 'text': 'Now the second area is credit evaluation.', 'start': 2077.629, 'duration': 2.081}, {'end': 2080.992, 'text': "Now I'll give you an example here.", 'start': 2079.751, 'duration': 1.241}, {'end': 2084.594, 'text': 'The HNC company has developed several neural network applications,', 'start': 2081.032, 'duration': 3.562}, {'end': 2091.877, 'text': 'and one of them is a credit scoring system which increases the profitability of existing model up to 27%.', 'start': 2084.594, 'duration': 7.283}, {'end': 2094.659, 'text': "So these are few applications that I'm telling you guys.", 'start': 2091.877, 'duration': 2.782}], 'summary': 'Neural networks improve airline seat allocation and credit scoring by up to 27%.', 'duration': 26.998, 'max_score': 2067.661, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU2067661.jpg'}], 'start': 1476.373, 'title': 'Implementing node classification and neural network applications', 'summary': 'Focuses on implementing a use case to determine node authenticity, including steps such as data processing and model training. it also discusses achieving 99% accuracy in neural network training, model testing, and applications in medicine and business.', 'chapters': [{'end': 1756.604, 'start': 1476.373, 'title': 'Implementing a node classification use case', 'summary': 'Focuses on implementing a use case to determine whether a node is fake or real, involving steps such as importing libraries, reading and encoding the data set, splitting the data set into training and testing, defining learning rate and epochs, initializing variables, and creating a saver object.', 'duration': 280.231, 'highlights': ['The chapter discusses the use of important libraries such as Matplotlib for visualization, TensorFlow for neural networks, NumPy for arrays, Pandas for reading the data set, and sklearn for label encoding and data set splitting.', 'It explains the process of reading the CSV file using Pandas, defining features and labels, and encoding the dependent variable, which is crucial for the use case implementation.', "The chapter covers the optional steps of splitting the data set into training and testing, including the printing of the shape of the training and test data, providing insights into the data set's structure.", 'It details the definition of learning rate, epochs, course history, endim (x shape of axis one), number of classes, and the model path for saving the model, essential parameters for the neural network implementation.', 'The chapter provides a comprehensive explanation of defining the neural network, including parameters like hidden layers, number of neurons, placeholders for X and Y, weight and bias initialization, and the creation of a multi-layer perceptron function for modeling.', 'It highlights the process of defining the weights and biases using truncated normal distributions and the importance of initializing variables before running them, as discussed in the TensorFlow tutorial.']}, {'end': 2193.791, 'start': 1757.004, 'title': 'Neural network applications', 'summary': "Explains the training process of a neural network model, achieving 99% accuracy after 100 epochs, restoring and testing the model's accuracy, and discusses the applications of neural networks in medicine and business.", 'duration': 436.787, 'highlights': ['The model achieved 99% accuracy after 100 epochs of training. The accuracy of the model reached around 99% after 100 epochs of training.', 'The model showed 100% accuracy in predicting the class of test data values. The model displayed 100% accuracy in correctly identifying the class of test data values.', 'Neural networks are extensively researched for applications in medicine, particularly in modeling and diagnosing the cardiovascular system. Research on neural networks in medicine focuses on modeling and diagnosing the cardiovascular system, aiding in the early detection of potential harmful medical conditions.', 'Neural networks have potential applications in business, including marketing and credit evaluation. Neural networks can be applied in business areas like marketing and credit evaluation, leading to improved profitability and resource allocation.']}], 'duration': 717.418, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/fv6Qll3laUU/pics/fv6Qll3laUU1476373.jpg', 'highlights': ['The model showed 100% accuracy in predicting the class of test data values.', 'The model achieved 99% accuracy after 100 epochs of training.', 'Neural networks are extensively researched for applications in medicine, particularly in modeling and diagnosing the cardiovascular system.', 'Neural networks have potential applications in business, including marketing and credit evaluation.', 'The chapter discusses the use of important libraries such as Matplotlib, TensorFlow, NumPy, Pandas, and sklearn for label encoding and data set splitting.', 'It explains the process of reading the CSV file using Pandas, defining features and labels, and encoding the dependent variable, crucial for the use case implementation.']}], 'highlights': ['The model achieved 99% accuracy after 100 epochs of training.', 'The chapter covers the fundamentals of neural networks and their application in authenticating banknotes using artificial neural networks and extracting features from images.', "The dataset contains features like wavelet transformed image variants, skewness, courtesies, and entropy, and labels indicating real or fake notes, with '1' representing real notes and '0' representing fake notes.", 'The process involves implementing a neural network model using TensorFlow, including data set reading, feature and label definition, encoding of dependent variable, model implementation, training, error calculation, and accuracy prediction.', 'The model showed 100% accuracy in predicting the class of test data values.', 'The decision-making process for attending a beer festival is explained using an analogy, with three factors - weather, company, and transportation - determining the decision.', 'The process of training a neural network involves back propagation, where iterative weight adjustments are made to minimize error by comparing the model output with the desired output until the error becomes minimum after a lot of iterations.', 'Neural networks enable computers to solve problems without the need for specific instructions.', 'Artificial neural networks possess self-learning capabilities, leading to improved accuracy as more data is utilized for training.', 'Neural networks have facilitated technological advancements such as web page translation, virtual assistants for online shopping, and chatbot interactions.', 'The chapter discusses the use of important libraries such as Matplotlib, TensorFlow, NumPy, Pandas, and sklearn for label encoding and data set splitting.', 'Neural networks are extensively researched for applications in medicine, particularly in modeling and diagnosing the cardiovascular system.', 'Neural networks have potential applications in business, including marketing and credit evaluation.']}