title
Complete Road Map To Prepare For Deep Learning🔥🔥🔥🔥
description
In this video I am going to discuss about the complete road map to prepare for deep learning which will be definitely helpful for preparing for interviews
Complete DL Playlist :https://www.youtube.com/playlist?list=PLZoTAELRMXVPGU70ZGsckrMdr0FteeRUi
⭐ Kite is a free AI-powered coding assistant that will help you code faster and smarter. The Kite plugin integrates with all the top editors and IDEs to give you smart completions and documentation while you’re typing. I've been using Kite for a few months and I love it! https://www.kite.com/get-kite/?utm_medium=referral&utm_source=youtube&utm_campaign=krishnaik&utm_content=description-only
All Playlist In My channel
Complete ML Playlist :https://www.youtube.com/playlist?list=PLZoTAELRMXVPBTrWtJkn3wWQxZkmTXGwe
Complete NLP Playlist:https://www.youtube.com/playlist?list=PLZoTAELRMXVMdJ5sqbCK2LiM0HhQVWNzm
Docker End To End Implementation: https://www.youtube.com/playlist?list=PLZoTAELRMXVNKtpy0U_Mx9N26w8n0hIbs
Live stream Playlist: https://www.youtube.com/playlist?list=PLZoTAELRMXVNxYFq_9MuiUdn2YnlFqmMK
Machine Learning Pipelines: https://www.youtube.com/playlist?list=PLZoTAELRMXVNKtpy0U_Mx9N26w8n0hIbs
Pytorch Playlist: https://www.youtube.com/playlist?list=PLZoTAELRMXVNxYFq_9MuiUdn2YnlFqmMK
Feature Engineering :https://www.youtube.com/playlist?list=PLZoTAELRMXVPwYGE2PXD3x0bfKnR0cJjN
Live Projects :https://www.youtube.com/playlist?list=PLZoTAELRMXVOFnfSwkB_uyr4FT-327noK
Kaggle competition :https://www.youtube.com/playlist?list=PLZoTAELRMXVPiKOxbwaniXjHJ02bdkLWy
Mongodb with Python :https://www.youtube.com/playlist?list=PLZoTAELRMXVN_8zzsevm1bm6G-plsiO1I
MySQL With Python :https://www.youtube.com/playlist?list=PLZoTAELRMXVMd3RF7p-u7ezEysGaG9JmO
Deployment Architectures:https://www.youtube.com/playlist?list=PLZoTAELRMXVOPzVJiSJAn9Ly27Fi1-8ac
Amazon sagemaker :https://www.youtube.com/playlist?list=PLZoTAELRMXVONh5mHrXowH6-dgyWoC_Ew
Please donate if you want to support the channel through GPay UPID,
Gpay: krishnaik06@okicici
Discord Server Link: https://discord.gg/tvAJuuy
Telegram link: https://t.me/joinchat/N77M7xRvYUd403DgfE4TWw
Please join as a member in my channel to get additional benefits like materials in Data Science, live streaming for Members and many more
https://www.youtube.com/channel/UCNU_lfiiWBdtULKOw6X0Dig/join
Please do subscribe my other channel too
https://www.youtube.com/channel/UCjWY5hREA6FFYrthD0rZNIw
Connect with me here:
Twitter: https://twitter.com/Krishnaik06
Facebook: https://www.facebook.com/krishnaik06
instagram: https://www.instagram.com/krishnaik06
#deeplearning
#dl
#deeplearninginterviews
detail
{'title': 'Complete Road Map To Prepare For Deep Learning🔥🔥🔥🔥', 'heatmap': [{'end': 132.612, 'start': 83.227, 'weight': 0.737}, {'end': 266.111, 'start': 242.493, 'weight': 0.752}, {'end': 357.223, 'start': 331.451, 'weight': 0.864}, {'end': 787.481, 'start': 767.009, 'weight': 0.744}, {'end': 876.734, 'start': 854.339, 'weight': 0.84}], 'summary': 'Provides a comprehensive roadmap for preparing for deep learning, covering the increasing popularity of deep learning, fundamental concepts like neural networks and back propagation, implementation of convolutional neural networks and transfer learning, and ai interview topics including optimizers, loss functions, and activation functions, with emphasis on nlp and computer vision, and offering a deep learning roadmap with key libraries like pytorch, keras, and tensorflow, along with 53 videos on deep learning techniques and algorithms.', 'chapters': [{'end': 64.083, 'segs': [{'end': 48.196, 'src': 'embed', 'start': 15.535, 'weight': 0, 'content': [{'end': 20.697, 'text': "so that you'll be able to cover almost most of the things, so that will be helpful for you for interviews.", 'start': 15.535, 'duration': 5.162}, {'end': 22.338, 'text': 'And one more thing, guys.', 'start': 21.357, 'duration': 0.981}, {'end': 30.444, 'text': 'yes, deep learning is becoming much more popular because there were many questions put up by many of my subscribers saying that is machine learning enough to get into jobs?', 'start': 22.338, 'duration': 8.106}, {'end': 31.484, 'text': 'with respect to data science?', 'start': 30.444, 'duration': 1.04}, {'end': 36.748, 'text': "I'll say, guys, nowadays it would be difficult, because companies are looking for both deep learning and machine learning techniques.", 'start': 32.205, 'duration': 4.543}, {'end': 38.049, 'text': 'It is good to have both of them.', 'start': 36.808, 'duration': 1.241}, {'end': 40.19, 'text': 'And that is what is the trend.', 'start': 38.869, 'duration': 1.321}, {'end': 46.295, 'text': "If you go three years back, I think I would definitely say that if you know machine learning, you'll be able to get the jobs itself.", 'start': 40.29, 'duration': 6.005}, {'end': 48.196, 'text': "So let's understand this roadmap.", 'start': 46.835, 'duration': 1.361}], 'summary': 'Companies now seek both deep learning and machine learning for jobs in data science.', 'duration': 32.661, 'max_score': 15.535, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c15535.jpg'}], 'start': 0.029, 'title': 'Roadmap to deep learning', 'summary': 'Discusses the increasing popularity of deep learning, the necessity of learning both deep learning and machine learning techniques, and provides a roadmap for effective learning to prepare for interviews and job opportunities in data science.', 'chapters': [{'end': 64.083, 'start': 0.029, 'title': 'Roadmap to deep learning', 'summary': 'Discusses the increasing popularity of deep learning, the necessity of learning both deep learning and machine learning techniques, and provides a roadmap for learning deep learning effectively to prepare for interviews and job opportunities in data science.', 'duration': 64.054, 'highlights': ['Deep learning and machine learning techniques are both necessary for job opportunities in data science, as companies are increasingly looking for candidates with knowledge in both areas, reflecting the current trend.', 'The popularity of deep learning has increased, making it essential to learn in order to be competitive in the job market, especially in comparison to three years ago when machine learning knowledge was sufficient for job opportunities.', 'The chapter provides a roadmap for learning deep learning effectively, aiming to cover a comprehensive set of topics to prepare individuals for interviews and job opportunities in data science.']}], 'duration': 64.054, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c29.jpg', 'highlights': ['The popularity of deep learning has increased, making it essential to learn in order to be competitive in the job market, especially in comparison to three years ago when machine learning knowledge was sufficient for job opportunities.', 'Deep learning and machine learning techniques are both necessary for job opportunities in data science, as companies are increasingly looking for candidates with knowledge in both areas, reflecting the current trend.', 'The chapter provides a roadmap for learning deep learning effectively, aiming to cover a comprehensive set of topics to prepare individuals for interviews and job opportunities in data science.']}, {'end': 295.373, 'segs': [{'end': 132.612, 'src': 'heatmap', 'start': 64.602, 'weight': 0, 'content': [{'end': 71.085, 'text': 'Now, one of the most important thing is that why did deep learning come into existence and what is the main aim of the neural networks,', 'start': 64.602, 'duration': 6.483}, {'end': 72.945, 'text': 'which is basically used in deep learning?', 'start': 71.085, 'duration': 1.86}, {'end': 74.765, 'text': 'It is used to mimic the human brain.', 'start': 73.005, 'duration': 1.76}, {'end': 82.827, 'text': 'So scientists, researchers, actually thought that can we make the machine learn in such a way that like how we human being learn right?', 'start': 75.205, 'duration': 7.622}, {'end': 89.748, 'text': 'So because of that, they came up with the concept of perceptron neural networks, and it was Mr. Jeffrey Hinton Right.', 'start': 83.227, 'duration': 6.521}, {'end': 97.49, 'text': 'Because of him, because of his paper that was on back propagation algorithm that led to the invention of all these techniques that are right now.', 'start': 89.868, 'duration': 7.622}, {'end': 102.331, 'text': 'Yes So let us go ahead and try to start with how we should start with deep learning itself.', 'start': 98.09, 'duration': 4.241}, {'end': 108.633, 'text': "Now, guys, in the base over here, in the bottom part, you can see that I've actually written introduction to neural network.", 'start': 102.852, 'duration': 5.781}, {'end': 115.535, 'text': "Then I've written loss function optimizes gradient descent, stochastic gradient descent, AdaGrad, RMS prop and Adam.", 'start': 109.093, 'duration': 6.442}, {'end': 121.44, 'text': 'Apart from this, there are also activations functions like ReLU, TanH, Sigmoid activation functions.', 'start': 116.495, 'duration': 4.945}, {'end': 124.984, 'text': 'So that all are the base for everything that we will try to learn.', 'start': 121.46, 'duration': 3.524}, {'end': 127.346, 'text': 'Because on training right?', 'start': 125.425, 'duration': 1.921}, {'end': 132.612, 'text': "Suppose, if we are training our images in convolution neural network indirectly, you'll be using some kind of optimizers,", 'start': 127.387, 'duration': 5.225}], 'summary': "Deep learning aims to mimic human brain using neural networks, pioneered by jeffrey hinton's back propagation algorithm.", 'duration': 37.729, 'max_score': 64.602, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c64602.jpg'}, {'end': 152.192, 'src': 'embed', 'start': 121.46, 'weight': 3, 'content': [{'end': 124.984, 'text': 'So that all are the base for everything that we will try to learn.', 'start': 121.46, 'duration': 3.524}, {'end': 127.346, 'text': 'Because on training right?', 'start': 125.425, 'duration': 1.921}, {'end': 132.612, 'text': "Suppose, if we are training our images in convolution neural network indirectly, you'll be using some kind of optimizers,", 'start': 127.387, 'duration': 5.225}, {'end': 134.234, 'text': "you'll be using some kind of loss functions.", 'start': 132.612, 'duration': 1.622}, {'end': 137.237, 'text': "Apart from that, you'll also be using some kind of activation functions.", 'start': 134.654, 'duration': 2.583}, {'end': 143.743, 'text': 'So in short to begin with guys all the topics that I have actually mentioned over here in the base are pretty much important.', 'start': 137.837, 'duration': 5.906}, {'end': 145.065, 'text': 'You need to learn this.', 'start': 144.104, 'duration': 0.961}, {'end': 152.192, 'text': 'you need to know the maths, how it actually works, because implementation of this is just like hardly a single line of code and training.', 'start': 145.065, 'duration': 7.127}], 'summary': 'Understanding base concepts like optimizers, loss functions, and activation functions is crucial for training in convolutional neural networks.', 'duration': 30.732, 'max_score': 121.46, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c121460.jpg'}, {'end': 218.372, 'src': 'embed', 'start': 192.935, 'weight': 2, 'content': [{'end': 198.457, 'text': 'This Gradient Descent, if you understand, I think Stochastic Gradient Descent will also become easy because there is some minor change.', 'start': 192.935, 'duration': 5.522}, {'end': 200.238, 'text': 'There was some disadvantage in Gradient Descent.', 'start': 198.477, 'duration': 1.761}, {'end': 206.142, 'text': 'So they came up with SGD, then they came up with Adagrad, then they came up with RMS prop and then, finally,', 'start': 200.578, 'duration': 5.564}, {'end': 208.584, 'text': 'which is a state of art which is called as Adam optimizer.', 'start': 206.142, 'duration': 2.442}, {'end': 213.668, 'text': 'Nowadays, everybody is using Adam optimizer, because it is nothing, but it is.', 'start': 209.245, 'duration': 4.423}, {'end': 215.53, 'text': 'it is also able to change the momentum.', 'start': 213.668, 'duration': 1.862}, {'end': 218.372, 'text': 'that is basically a learning rate, as your training is actually going on.', 'start': 215.53, 'duration': 2.842}], 'summary': 'Evolution of optimization algorithms: gradient descent to adam optimizer, widely used for its adaptive learning rate and momentum.', 'duration': 25.437, 'max_score': 192.935, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c192935.jpg'}, {'end': 271.477, 'src': 'heatmap', 'start': 242.493, 'weight': 0.752, 'content': [{'end': 247.375, 'text': "Once you're able to understand all these things, I think directly you'll be able to implement artificial neural network.", 'start': 242.493, 'duration': 4.882}, {'end': 252.557, 'text': 'Now in machine learning you will be having a data set like a regression or classification problem.', 'start': 247.816, 'duration': 4.741}, {'end': 256.178, 'text': 'that similar problem statement you can solve with the help of artificial neural network.', 'start': 252.557, 'duration': 3.621}, {'end': 260.964, 'text': 'OK, now, when you implement artificial neural network, you can do it in a local machine.', 'start': 256.678, 'duration': 4.286}, {'end': 266.111, 'text': "You can you should also have the knowledge of Google Colab because there there'll be a usage of GPUs.", 'start': 260.985, 'duration': 5.126}, {'end': 271.477, 'text': "You'll get an idea how to work in the GPU, how to execute the programs and see that how the training actually happens.", 'start': 266.471, 'duration': 5.006}], 'summary': 'Understanding artificial neural networks leads to implementation, involving datasets, local machines, and google colab with gpus.', 'duration': 28.984, 'max_score': 242.493, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c242493.jpg'}, {'end': 300.935, 'src': 'embed', 'start': 275.661, 'weight': 4, 'content': [{'end': 283.265, 'text': 'try to deploy some projects by using some web frameworks like flask or in the heroku cloud, in the aws cloud, azure cloud,', 'start': 275.661, 'duration': 7.604}, {'end': 284.506, 'text': 'any type of cloud that you want.', 'start': 283.265, 'duration': 1.241}, {'end': 286.607, 'text': 'again, you can check my deployment playlist.', 'start': 284.506, 'duration': 2.101}, {'end': 292.491, 'text': 'you should also try to dockerize this whole model and try to dockerize this whole web application.', 'start': 286.607, 'duration': 5.884}, {'end': 293.512, 'text': 'try to deploy that.', 'start': 292.491, 'duration': 1.021}, {'end': 295.373, 'text': 'that will also be a very, very good idea.', 'start': 293.512, 'duration': 1.861}, {'end': 300.935, 'text': "now, as i said you, that whatever problem statement that you're solving with machine learning can also definitely solve it.", 'start': 296.073, 'duration': 4.862}], 'summary': 'Deploy projects using web frameworks in various clouds, dockerize web application for deployment, and solve ml problems.', 'duration': 25.274, 'max_score': 275.661, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c275661.jpg'}], 'start': 64.602, 'title': 'Deep learning fundamentals', 'summary': 'Delves into the origin of deep learning, focusing on neural networks and the back propagation algorithm by mr. jeffrey hinton. it also covers fundamental concepts including loss functions, optimizers like gradient descent, sgd, adagrad, rms prop, and adam, and activation functions such as relu, tanh, and sigmoid.', 'chapters': [{'end': 102.331, 'start': 64.602, 'title': 'Origin of deep learning', 'summary': 'Explains the origin of deep learning, highlighting the concept of neural networks aiming to mimic the human brain, particularly focusing on the role of perceptron neural networks and the back propagation algorithm by mr. jeffrey hinton in the development of deep learning techniques.', 'duration': 37.729, 'highlights': ['The concept of neural networks aims to mimic the human brain, sparking the development of deep learning techniques.', 'The role of perceptron neural networks and the back propagation algorithm by Mr. Jeffrey Hinton in the development of deep learning techniques.']}, {'end': 295.373, 'start': 102.852, 'title': 'Neural network fundamentals', 'summary': 'Covers the fundamental concepts of neural networks, including loss functions, optimizers such as gradient descent, sgd, adagrad, rms prop, and adam, as well as activation functions like relu, tanh, and sigmoid, emphasizing their importance in understanding the implementation and training process of artificial neural networks.', 'duration': 192.521, 'highlights': ['Understanding the fundamental concepts of neural networks, including loss functions, optimizers, and activation functions, is crucial before delving into artificial neural networks, convolutional neural networks, or recurrent neural networks. importance of understanding basic concepts before progressing to advanced neural network models', 'Exploration of loss functions and optimizers like Gradient Descent, SGD, Adagrad, RMS prop, and Adam, and their significance in training and achieving optimal minima. importance of loss functions and optimizers in training neural networks', 'Advising the audience to further their knowledge by learning about the deployment of projects using web frameworks like Flask, cloud platforms such as AWS, Azure, and Dockerization of models and web applications. recommendation to explore project deployment using web frameworks and cloud platforms']}], 'duration': 230.771, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c64602.jpg', 'highlights': ['The role of perceptron neural networks and the back propagation algorithm by Mr. Jeffrey Hinton in the development of deep learning techniques.', 'The concept of neural networks aims to mimic the human brain, sparking the development of deep learning techniques.', 'Exploration of loss functions and optimizers like Gradient Descent, SGD, Adagrad, RMS prop, and Adam, and their significance in training and achieving optimal minima.', 'Understanding the fundamental concepts of neural networks, including loss functions, optimizers, and activation functions, is crucial before delving into artificial neural networks, convolutional neural networks, or recurrent neural networks.', 'Advising the audience to further their knowledge by learning about the deployment of projects using web frameworks like Flask, cloud platforms such as AWS, Azure, and Dockerization of models and web applications.']}, {'end': 653.417, 'segs': [{'end': 324.146, 'src': 'embed', 'start': 296.073, 'weight': 7, 'content': [{'end': 300.935, 'text': "now, as i said you, that whatever problem statement that you're solving with machine learning can also definitely solve it.", 'start': 296.073, 'duration': 4.862}, {'end': 302.416, 'text': 'artificial neural network.', 'start': 300.935, 'duration': 1.481}, {'end': 306.017, 'text': 'there are also some concepts in artificial neural network, like weight initialization.', 'start': 302.416, 'duration': 3.601}, {'end': 307.498, 'text': 'how do you do the weight initialization?', 'start': 306.017, 'duration': 1.481}, {'end': 309.659, 'text': 'what are the different parameters?', 'start': 307.498, 'duration': 2.161}, {'end': 315.181, 'text': 'how you can actually perform hyper parameter tuning with respect to artificial neural network, right like, how do you decide, like,', 'start': 309.659, 'duration': 5.522}, {'end': 316.622, 'text': 'how many number of hidden layers are there?', 'start': 315.181, 'duration': 1.441}, {'end': 319.023, 'text': 'how many number of neurons should i take in the hidden layer?', 'start': 316.622, 'duration': 2.401}, {'end': 324.146, 'text': 'that all can be done by the help of hyper parameter tuning, And for that you can use Keras tuner.', 'start': 319.023, 'duration': 5.123}], 'summary': 'Machine learning problems can be solved using artificial neural network with techniques like weight initialization and hyper parameter tuning using keras tuner.', 'duration': 28.073, 'max_score': 296.073, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c296073.jpg'}, {'end': 357.223, 'src': 'heatmap', 'start': 331.451, 'weight': 0.864, 'content': [{'end': 336.594, 'text': "So once you're comfortable with this, the type of libraries that you can use is PyTorch, Keras, TensorFlow.", 'start': 331.451, 'duration': 5.143}, {'end': 342.379, 'text': 'I would suggest you to go with Keras, because it will be very, very easy if you are trying to use Keras with TensorFlow 2.0,', 'start': 336.975, 'duration': 5.404}, {'end': 344.14, 'text': 'because Keras is already included in that.', 'start': 342.379, 'duration': 1.761}, {'end': 346.44, 'text': 'okay, so go with that.', 'start': 344.6, 'duration': 1.84}, {'end': 348.441, 'text': "and also pytorch videos also i'm trying to.", 'start': 346.44, 'duration': 2.001}, {'end': 350.922, 'text': "i i've been uploading in the other playlist.", 'start': 348.441, 'duration': 2.481}, {'end': 352.622, 'text': 'now coming to the second pillar.', 'start': 350.922, 'duration': 1.7}, {'end': 357.223, 'text': 'now with respect to the second pillar, you have convolution neural network.', 'start': 352.622, 'duration': 4.601}], 'summary': 'Recommended using keras with tensorflow 2.0 for ease, along with pytorch videos. also, mentions convolutional neural network.', 'duration': 25.772, 'max_score': 331.451, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c331451.jpg'}, {'end': 365.025, 'src': 'embed', 'start': 336.975, 'weight': 6, 'content': [{'end': 342.379, 'text': 'I would suggest you to go with Keras, because it will be very, very easy if you are trying to use Keras with TensorFlow 2.0,', 'start': 336.975, 'duration': 5.404}, {'end': 344.14, 'text': 'because Keras is already included in that.', 'start': 342.379, 'duration': 1.761}, {'end': 346.44, 'text': 'okay, so go with that.', 'start': 344.6, 'duration': 1.84}, {'end': 348.441, 'text': "and also pytorch videos also i'm trying to.", 'start': 346.44, 'duration': 2.001}, {'end': 350.922, 'text': "i i've been uploading in the other playlist.", 'start': 348.441, 'duration': 2.481}, {'end': 352.622, 'text': 'now coming to the second pillar.', 'start': 350.922, 'duration': 1.7}, {'end': 357.223, 'text': 'now with respect to the second pillar, you have convolution neural network.', 'start': 352.622, 'duration': 4.601}, {'end': 365.025, 'text': 'again, the important concepts like loss function optimizers like gradient descent, sgd, adagrad, rms from adam, activation functions will be used.', 'start': 357.223, 'duration': 7.802}], 'summary': 'Use keras with tensorflow 2.0 for ease, also exploring pytorch. covering cnn and key concepts like loss functions and optimizers.', 'duration': 28.05, 'max_score': 336.975, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c336975.jpg'}, {'end': 442.714, 'src': 'embed', 'start': 410.652, 'weight': 0, 'content': [{'end': 411.473, 'text': 'what are filters?', 'start': 410.652, 'duration': 0.821}, {'end': 412.733, 'text': 'what are strides?', 'start': 411.473, 'duration': 1.26}, {'end': 413.674, 'text': 'what is the formula?', 'start': 412.733, 'duration': 0.941}, {'end': 420.517, 'text': 'how the image will get reduced when you apply a filter of 3, cross 3, 5, cross, 5, if your image is of 224, 224,', 'start': 413.674, 'duration': 6.843}, {'end': 422.619, 'text': 'what is the next layer that will actually happen?', 'start': 420.517, 'duration': 2.102}, {'end': 429.063, 'text': 'all those explanation is again given in my uh videos, guys, it is pretty much important to understand.', 'start': 422.619, 'duration': 6.444}, {'end': 434.447, 'text': "okay, apart from that, guys, i've also started with the transfer learning techniques and i have also uploaded a lot of videos.", 'start': 429.063, 'duration': 5.384}, {'end': 442.714, 'text': "i've also done end-to-end projects with respect to transfer learning, like vg 16, lx net, you have inception v3, you have resnet, everything,", 'start': 434.447, 'duration': 8.267}], 'summary': 'Explains filters, strides, image reduction, and transfer learning techniques, with end-to-end projects on vgg16, alexnet, inceptionv3, and resnet.', 'duration': 32.062, 'max_score': 410.652, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c410652.jpg'}, {'end': 497.599, 'src': 'embed', 'start': 462.551, 'weight': 1, 'content': [{'end': 469.557, 'text': 'in object detection, first of all, start with rcnn, then you have master rcnn, then you have shd single shot detector right.', 'start': 462.551, 'duration': 7.006}, {'end': 471.038, 'text': 'then you have the yolo algorithm.', 'start': 469.557, 'duration': 1.481}, {'end': 473.501, 'text': 'so yolo is pretty much handy for the object detection.', 'start': 471.038, 'duration': 2.463}, {'end': 477.664, 'text': 'it is also pretty much fast when compared to RCNN and masked RCNN.', 'start': 473.501, 'duration': 4.163}, {'end': 479.805, 'text': 'Masked RCNN is like object detection.', 'start': 477.764, 'duration': 2.041}, {'end': 481.066, 'text': "They'll also be doing masking.", 'start': 479.826, 'duration': 1.24}, {'end': 484.029, 'text': 'A lot of research are also going on with respect to this.', 'start': 481.106, 'duration': 2.923}, {'end': 486.571, 'text': "So this all topics I've actually mentioned guys.", 'start': 484.529, 'duration': 2.042}, {'end': 491.454, 'text': "And again, I'm not telling you as a fresher, you should directly go understanding object detection.", 'start': 486.611, 'duration': 4.843}, {'end': 495.357, 'text': 'So you start with convolution neural network, then go with transfer learning techniques.', 'start': 491.534, 'duration': 3.823}, {'end': 497.599, 'text': 'Once, you are very good with transfer learning techniques.', 'start': 495.557, 'duration': 2.042}], 'summary': 'Yolo algorithm is fast and handy for object detection, compared to rcnn and masked rcnn.', 'duration': 35.048, 'max_score': 462.551, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c462551.jpg'}, {'end': 544.622, 'src': 'embed', 'start': 513.99, 'weight': 3, 'content': [{'end': 518.852, 'text': "once you're able to create again, do the deployment in different platforms like Heroku, AWS,", 'start': 513.99, 'duration': 4.862}, {'end': 523.236, 'text': 'Azure and just try to check that how that particular application actually works, okay?', 'start': 518.852, 'duration': 4.384}, {'end': 529.998, 'text': "And once you're comfortable with all the transfer learning techniques, then move into object detection and try to cover this five algorithms,", 'start': 523.635, 'duration': 6.363}, {'end': 534.779, 'text': "or four algorithms that I've actually mentioned, because these are some amazing algorithms itself, right?", 'start': 529.998, 'duration': 4.781}, {'end': 537.2, 'text': 'Now, once this is done, this is your second pillar.', 'start': 535.199, 'duration': 2.001}, {'end': 544.622, 'text': "Again, the most popular pillar out of this, I'll suggest that it is the NLP part or you can say the RNN part.", 'start': 537.68, 'duration': 6.942}], 'summary': 'Learn deployment on multiple platforms, then move to object detection with 5 algorithms, and finally focus on nlp or rnn.', 'duration': 30.632, 'max_score': 513.99, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c513990.jpg'}, {'end': 623.104, 'src': 'embed', 'start': 586.592, 'weight': 4, 'content': [{'end': 593.922, 'text': "What is the advantages of attention models? Then we are coming to Transformers, BERT, right? So Transformers, again, I've taken the live session.", 'start': 586.592, 'duration': 7.33}, {'end': 597.747, 'text': 'Only the things is about BERT, but is also pretty much amazing.', 'start': 594.342, 'duration': 3.405}, {'end': 604.157, 'text': 'And if I talk about BERT, guys, all the NLP transfer learning techniques can be actually done with the help of BERT and Transformers, right?', 'start': 597.787, 'duration': 6.37}, {'end': 608.579, 'text': 'and the libraries that are actually used are called as hugging face and k train.', 'start': 604.617, 'duration': 3.962}, {'end': 612.02, 'text': 'now you will be thinking, krish, what all videos you have not uploaded.', 'start': 608.579, 'duration': 3.441}, {'end': 616.201, 'text': 'guys, i need to upload this object detection, all the videos, and i need to upload,', 'start': 612.02, 'duration': 4.181}, {'end': 623.104, 'text': "uh birth theoretical videos and with the help of hugging face and k train, i'll be showing you a lot of practical application.", 'start': 616.201, 'duration': 6.903}], 'summary': 'Transformers like bert enable nlp transfer learning, utilizing libraries such as hugging face and k train for practical applications.', 'duration': 36.512, 'max_score': 586.592, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c586592.jpg'}], 'start': 296.073, 'title': 'Neural networks, convolution, transfer learning, and object detection', 'summary': 'Covers artificial neural networks, hyperparameter tuning, and convolutional neural networks, emphasizing keras implementation and understanding convolution layer math concepts. it also discusses transfer learning, object detection, and the importance of focusing on these techniques before moving on to nlp, with mention of various algorithms and platforms, as well as plans for future video uploads.', 'chapters': [{'end': 429.063, 'start': 296.073, 'title': 'Neural networks and convolution', 'summary': 'Discusses the concepts of artificial neural networks, hyperparameter tuning, and convolutional neural networks, emphasizing the importance of keras for easy implementation and the significance of understanding convolution layer math concepts for image processing.', 'duration': 132.99, 'highlights': ['The importance of using Keras with TensorFlow 2.0 for easy implementation is emphasized, with Keras tuner and Auto Keras recommended for hyperparameter tuning.', 'The significance of understanding convolution layer math concepts for image processing, including filters, strides, and image reduction, is highlighted.', 'Concepts in artificial neural networks, such as weight initialization and hyperparameter tuning, are discussed, with Keras, PyTorch, and TensorFlow suggested as applicable libraries.']}, {'end': 653.417, 'start': 429.063, 'title': 'Transfer learning and object detection', 'summary': 'Discusses the importance of transfer learning and object detection in machine learning, emphasizing the need to focus on these techniques before moving on to nlp, with mention of various algorithms and platforms, as well as plans for future video uploads.', 'duration': 224.354, 'highlights': ['The importance of focusing on transfer learning and object detection as foundational techniques before delving into NLP, with specific recommendations on algorithms to learn, like YOLO for object detection.', 'The mention of specific platforms for deployment, including Heroku, AWS, and Azure, as well as plans for future video uploads related to object detection and NLP.', 'The emphasis on preparing materials for Hugging Face, including practical applications and custom techniques using the Hugging Face and K-Train library.', 'The mention of various NLP techniques and models like RNN, LHTM, GRU, word embedding, BERT, and Transformers, with an emphasis on their significance in NLP transfer learning.', 'The recommendation for beginners to start with convolutional neural networks before progressing to transfer learning techniques and end-to-end projects, followed by a transition into object detection and NLP.', 'The recommendation for experienced individuals to also start with object detection, particularly mentioning algorithms like RCNN, Masked RCNN, and YOLO, and highlighting the speed and effectiveness of YOLO for object detection.']}], 'duration': 357.344, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c296073.jpg', 'highlights': ['The importance of understanding convolution layer math concepts for image processing is emphasized, including filters, strides, and image reduction.', 'The recommendation for beginners to start with convolutional neural networks before progressing to transfer learning techniques and end-to-end projects, followed by a transition into object detection and NLP.', 'The recommendation for experienced individuals to also start with object detection, particularly mentioning algorithms like RCNN, Masked RCNN, and YOLO, and highlighting the speed and effectiveness of YOLO for object detection.', 'The mention of specific platforms for deployment, including Heroku, AWS, and Azure, as well as plans for future video uploads related to object detection and NLP.', 'The emphasis on preparing materials for Hugging Face, including practical applications and custom techniques using the Hugging Face and K-Train library.', 'The mention of various NLP techniques and models like RNN, LHTM, GRU, word embedding, BERT, and Transformers, with an emphasis on their significance in NLP transfer learning.', 'The importance of using Keras with TensorFlow 2.0 for easy implementation is emphasized, with Keras tuner and Auto Keras recommended for hyperparameter tuning.', 'Concepts in artificial neural networks, such as weight initialization and hyperparameter tuning, are discussed, with Keras, PyTorch, and TensorFlow suggested as applicable libraries.']}, {'end': 982.442, 'segs': [{'end': 681.649, 'src': 'embed', 'start': 653.977, 'weight': 0, 'content': [{'end': 659.739, 'text': "Again, I'm also planning to make some deployment projects end to end so that you'll be able to accommodate all these things.", 'start': 653.977, 'duration': 5.762}, {'end': 661.74, 'text': 'Now, these all are pretty important.', 'start': 660.199, 'duration': 1.541}, {'end': 668.662, 'text': 'You have to go again the base and the most favorite interview questions will be revolving around this base.', 'start': 661.8, 'duration': 6.862}, {'end': 672.983, 'text': 'things like optimizers, you know, loss functions, activation functions.', 'start': 668.662, 'duration': 4.321}, {'end': 674.604, 'text': 'things about convolution, neural network.', 'start': 672.983, 'duration': 1.621}, {'end': 677.766, 'text': 'again, they directly will not ask you about object detection, you know.', 'start': 674.984, 'duration': 2.782}, {'end': 681.649, 'text': "they'll just ask you that, whether you know or not, give us some basic idea, something like that.", 'start': 677.766, 'duration': 3.883}], 'summary': 'Plan to create end-to-end deployment projects, focus on key topics like optimizers, loss functions, convolution, and neural network.', 'duration': 27.672, 'max_score': 653.977, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c653977.jpg'}, {'end': 745.527, 'src': 'embed', 'start': 687.714, 'weight': 1, 'content': [{'end': 694.92, 'text': 'they may talk about bi-directional, but out of this, uh, nlp part, which is this one right, is much, is becoming much more popular.', 'start': 687.714, 'duration': 7.206}, {'end': 698.062, 'text': "second, i'll say that it is, uh, convolution neural network.", 'start': 694.92, 'duration': 3.142}, {'end': 699.423, 'text': 'then you have, ann.', 'start': 698.062, 'duration': 1.361}, {'end': 705.548, 'text': 'apart from this, guys, in convolution neural network there is also a separate topic which is called as computer vision, Computer vision.', 'start': 699.423, 'duration': 6.125}, {'end': 715.431, 'text': 'try to learn in such a way that you will be able to capture the frames from your webcam and try to use it and detect it through your convolution neural networks.', 'start': 705.548, 'duration': 9.883}, {'end': 721.653, 'text': 'something like it may be face recognition, it may be different kind of image classification, live image classification and many more things.', 'start': 715.431, 'duration': 6.222}, {'end': 724.955, 'text': "Like you'll be creating some bounding boxes and many more things.", 'start': 721.774, 'duration': 3.181}, {'end': 727.976, 'text': 'And those kind of video has also been uploaded in my playlist.', 'start': 725.455, 'duration': 2.521}, {'end': 732.539, 'text': 'So this, in short, is the whole roadmap to prepare the deep learning.', 'start': 728.556, 'duration': 3.983}, {'end': 739.263, 'text': 'Again, remember, the important libraries are PyTorch, Keras, TensorFlow, then you have Hovering Phase, then you have Kteran libraries.', 'start': 733.079, 'duration': 6.184}, {'end': 743.625, 'text': "Now let's go ahead and I'll try to show you from where you can learn from my playlist.", 'start': 739.803, 'duration': 3.822}, {'end': 745.527, 'text': "So let's go over here, guys.", 'start': 744.546, 'duration': 0.981}], 'summary': 'Deep learning roadmap: nlp gaining popularity, cnn, ann, computer vision, important libraries pytorch, keras, tensorflow.', 'duration': 57.813, 'max_score': 687.714, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c687714.jpg'}, {'end': 796.088, 'src': 'heatmap', 'start': 767.009, 'weight': 0.744, 'content': [{'end': 768.07, 'text': 'now i have planned it.', 'start': 767.009, 'duration': 1.061}, {'end': 770.451, 'text': "i'm preparing the material as soon as that will be available.", 'start': 768.07, 'duration': 2.381}, {'end': 774.353, 'text': "it will be available to you all also and i'll try to upload it as soon as possible.", 'start': 770.451, 'duration': 3.902}, {'end': 777.975, 'text': 'guys, the other thing that you should follow is basically a keras blog.', 'start': 774.353, 'duration': 3.622}, {'end': 779.956, 'text': 'this keras blog is pretty much amazing.', 'start': 777.975, 'duration': 1.981}, {'end': 782.238, 'text': 'if you really want to know about losses, you can click away.', 'start': 779.956, 'duration': 2.282}, {'end': 784.939, 'text': 'what are the different loss technique like with regression?', 'start': 782.238, 'duration': 2.701}, {'end': 787.481, 'text': 'losses will be mean squared error, mean absolute error.', 'start': 784.939, 'duration': 2.542}, {'end': 789.222, 'text': 'i think we have also done this in machine learning.', 'start': 787.481, 'duration': 1.741}, {'end': 796.088, 'text': "right, In classification problem you'll be having binary cross entropy, then category cross entropy, sparse category cross entropy,", 'start': 789.222, 'duration': 6.866}], 'summary': 'Preparing material, uploading soon. follow keras blog for loss techniques in regression and classification.', 'duration': 29.079, 'max_score': 767.009, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c767009.jpg'}, {'end': 884.838, 'src': 'heatmap', 'start': 854.339, 'weight': 0.84, 'content': [{'end': 856.28, 'text': "You'll be able to see that how many parameters are there.", 'start': 854.339, 'duration': 1.941}, {'end': 861.484, 'text': 'Like recently, if you know about GPT-3, it has 175 billion parameters, I guess.', 'start': 857.401, 'duration': 4.083}, {'end': 865.787, 'text': 'Now here, if you take an example of VG16, it has 138 million parameters.', 'start': 861.904, 'duration': 3.883}, {'end': 869.329, 'text': "So billion and million, right? It's a huge gap.", 'start': 866.807, 'duration': 2.522}, {'end': 874.372, 'text': 'So it is said that GPT-3 is one of the state of the art algorithm and probably it should be available.', 'start': 869.689, 'duration': 4.683}, {'end': 876.734, 'text': "I don't know whether it will be available to everyone or not.", 'start': 874.532, 'duration': 2.202}, {'end': 879.476, 'text': 'I still requested for the API, but I have not got it yet.', 'start': 877.194, 'duration': 2.282}, {'end': 881.217, 'text': 'So that I start exploring things.', 'start': 879.896, 'duration': 1.321}, {'end': 884.838, 'text': 'so this was the complete roadmap.', 'start': 882.177, 'duration': 2.661}], 'summary': 'Gpt-3 has 175 billion parameters, vg16 has 138 million parameters, highlighting the vast gap between billion and million.', 'duration': 30.499, 'max_score': 854.339, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c854339.jpg'}, {'end': 980.819, 'src': 'embed', 'start': 955.786, 'weight': 4, 'content': [{'end': 961.747, 'text': 'I may be knowing this base, right? But on top of it, if I really want to implement anything, I also have to put that much effort.', 'start': 955.786, 'duration': 5.961}, {'end': 965.068, 'text': 'So that is the reason why I tell you that you have to practice things.', 'start': 962.227, 'duration': 2.841}, {'end': 967.108, 'text': "You don't have to leave those things.", 'start': 965.408, 'duration': 1.7}, {'end': 970.869, 'text': 'Whenever you have an interest on something, start working on it.', 'start': 967.128, 'duration': 3.741}, {'end': 975.31, 'text': 'okay?. If you have some amazing ideas, start working on it, because the base is ready for you.', 'start': 970.869, 'duration': 4.441}, {'end': 977.393, 'text': 'So I hope you like this particular video.', 'start': 976.01, 'duration': 1.383}, {'end': 979.396, 'text': 'Please do subscribe to the channel if you have not already subscribed.', 'start': 977.413, 'duration': 1.983}, {'end': 980.819, 'text': "And I'll see you all in the next video.", 'start': 979.697, 'duration': 1.122}], 'summary': 'Encouragement to practice and implement ideas with a strong base.', 'duration': 25.033, 'max_score': 955.786, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c955786.jpg'}], 'start': 653.977, 'title': 'Ai interview topics and deep learning roadmap', 'summary': 'Covers ai interview topics including optimizers, loss functions, activation functions, and neural network types, emphasizing nlp and computer vision. it also provides a deep learning roadmap with key libraries like pytorch, keras, and tensorflow, and 53 videos on deep learning techniques and algorithms, emphasizing the importance of practice.', 'chapters': [{'end': 724.955, 'start': 653.977, 'title': 'Key interview topics for ai roles', 'summary': 'Outlines the importance of understanding key ai concepts such as optimizers, loss functions, activation functions, and neural network types (ann, cnn, rnn) for interviews, emphasizing the growing significance of natural language processing (nlp) and computer vision.', 'duration': 70.978, 'highlights': ['Understanding of key AI concepts like optimizers, loss functions, and activation functions is crucial for interviews, with a focus on ANN, CNN, and RNN.', 'Emphasizing the growing importance of natural language processing (NLP) in AI roles.', 'The significance of understanding computer vision, including topics like capturing frames from a webcam, face recognition, and image classification.']}, {'end': 982.442, 'start': 725.455, 'title': 'Deep learning roadmap & techniques', 'summary': 'Presents a detailed roadmap for deep learning preparation, including key libraries like pytorch, keras, and tensorflow, along with 53 uploaded videos covering various deep learning techniques and state-of-the-art algorithms, emphasizing the importance of practice and exploration for skill development.', 'duration': 256.987, 'highlights': ['The chapter presents a detailed roadmap for deep learning preparation, including key libraries like PyTorch, Keras, and TensorFlow, along with 53 uploaded videos covering various deep learning techniques and state-of-the-art algorithms. The roadmap includes important libraries such as PyTorch, Keras, and TensorFlow, and covers 53 videos on deep learning techniques and state-of-the-art algorithms.', 'Emphasizes the importance of practice and exploration for skill development. The importance of practice and exploration is stressed for skill development in deep learning, encouraging viewers to work on applying the knowledge to real-world applications.']}], 'duration': 328.465, 'thumbnail': 'https://coursnap.oss-ap-southeast-1.aliyuncs.com/video-capture/9jA0KjS7V_c/pics/9jA0KjS7V_c653977.jpg', 'highlights': ['Understanding of key AI concepts like optimizers, loss functions, and activation functions is crucial for interviews, with a focus on ANN, CNN, and RNN.', 'Emphasizing the growing importance of natural language processing (NLP) in AI roles.', 'The significance of understanding computer vision, including topics like capturing frames from a webcam, face recognition, and image classification.', 'The chapter presents a detailed roadmap for deep learning preparation, including key libraries like PyTorch, Keras, and TensorFlow, along with 53 uploaded videos covering various deep learning techniques and state-of-the-art algorithms.', 'Emphasizes the importance of practice and exploration for skill development, encouraging viewers to work on applying the knowledge to real-world applications.']}], 'highlights': ['The popularity of deep learning has increased, making it essential to learn in order to be competitive in the job market, especially in comparison to three years ago when machine learning knowledge was sufficient for job opportunities.', 'Deep learning and machine learning techniques are both necessary for job opportunities in data science, as companies are increasingly looking for candidates with knowledge in both areas, reflecting the current trend.', 'The chapter provides a roadmap for learning deep learning effectively, aiming to cover a comprehensive set of topics to prepare individuals for interviews and job opportunities in data science.', 'The role of perceptron neural networks and the back propagation algorithm by Mr. Jeffrey Hinton in the development of deep learning techniques.', 'The concept of neural networks aims to mimic the human brain, sparking the development of deep learning techniques.', 'Exploration of loss functions and optimizers like Gradient Descent, SGD, Adagrad, RMS prop, and Adam, and their significance in training and achieving optimal minima.', 'Understanding the fundamental concepts of neural networks, including loss functions, optimizers, and activation functions, is crucial before delving into artificial neural networks, convolutional neural networks, or recurrent neural networks.', 'Advising the audience to further their knowledge by learning about the deployment of projects using web frameworks like Flask, cloud platforms such as AWS, Azure, and Dockerization of models and web applications.', 'The importance of understanding convolution layer math concepts for image processing is emphasized, including filters, strides, and image reduction.', 'The recommendation for beginners to start with convolutional neural networks before progressing to transfer learning techniques and end-to-end projects, followed by a transition into object detection and NLP.', 'The recommendation for experienced individuals to also start with object detection, particularly mentioning algorithms like RCNN, Masked RCNN, and YOLO, and highlighting the speed and effectiveness of YOLO for object detection.', 'The mention of specific platforms for deployment, including Heroku, AWS, and Azure, as well as plans for future video uploads related to object detection and NLP.', 'The emphasis on preparing materials for Hugging Face, including practical applications and custom techniques using the Hugging Face and K-Train library.', 'The mention of various NLP techniques and models like RNN, LHTM, GRU, word embedding, BERT, and Transformers, with an emphasis on their significance in NLP transfer learning.', 'The importance of using Keras with TensorFlow 2.0 for easy implementation is emphasized, with Keras tuner and Auto Keras recommended for hyperparameter tuning.', 'Concepts in artificial neural networks, such as weight initialization and hyperparameter tuning, are discussed, with Keras, PyTorch, and TensorFlow suggested as applicable libraries.', 'Understanding of key AI concepts like optimizers, loss functions, and activation functions is crucial for interviews, with a focus on ANN, CNN, and RNN.', 'Emphasizing the growing importance of natural language processing (NLP) in AI roles.', 'The significance of understanding computer vision, including topics like capturing frames from a webcam, face recognition, and image classification.', 'The chapter presents a detailed roadmap for deep learning preparation, including key libraries like PyTorch, Keras, and TensorFlow, along with 53 uploaded videos covering various deep learning techniques and state-of-the-art algorithms.', 'Emphasizes the importance of practice and exploration for skill development, encouraging viewers to work on applying the knowledge to real-world applications.']}