"🔥Caltech Post Graduate Program In AI And Machine Learning - www.simplilearn.com/artificial-intelligence-masters-program-training-course?Jy9-aGMB_TE&Comments&RUclips 🔥IITK - Professional Certificate Course in Generative AI and Machine Learning (India Only) - www.simplilearn.com/iitk-professional-certificate-course-ai-machine-learning?Jy9-aGMB_TE&Comments&RUclips 🔥Purdue - Post Graduate Program in AI and Machine Learning - www.simplilearn.com/pgp-ai-machine-learning-certification-training-course?Jy9-aGMB_TE&Comments&RUclips 🔥IITG - Professional Certificate Program in Generative AI and Machine Learning (India Only) - www.simplilearn.com/iitg-generative-ai-machine-learning-program?Jy9-aGMB_TE&Comments&RUclips 🔥Caltech - AI & Machine Learning Bootcamp (US Only) - www.simplilearn.com/ai-machine-learning-bootcamp?Jy9-aGMB_TE&Comments&RUclips"
Machine Learning is the Future and yours can begin today. Comment below with you email to get our latest Machine Learning Career Guide. Let your journey begin. Do you have any questions on this topic? Please share your feedback in the comment section below and we'll have our experts answer it for you. Thanks for watching the video. Cheers!
hey , i was training the model for 1500 times instead of 500 and i reached an accuracy of .6780.. how do u explain thay accuracy and also when predicting the label of a particular image it is giving a totally random label instead of right one....so what do u think is the problem ...well also i am initializing the variables again during testing of the particular image...is that why i am getting totally wrong output?
Hi Yogeesh, 0.67 is a pretty good accuracy if you attain it with just 500 iterations. Try running it for 1000 iterations. The precision will certainly increase. Please don't initialize the variables repeatedly while testing an image.
@@SimplilearnOfficial hey , i got this accuracy after 1500 epochs... and also how to predict an image's label after training it? cause i fed an image into the network and a label and tried running the network but it showed the error as "attempting to use uninitialized variables." so my question is how to load or use the trained model weights and predict image since i am not using keras ?
Hi. First of all thank you very much for such a great video. The question is how yo chose the filter. At the begining you said that you will talk about it, but you didnot. May be i missed. ??!?
Please initialize the variable you want to predict and then run it. Use Jupyter notebook and TensorFlow to do it. Keras is now built on top of TensorFlow.
Hello how do we obatain the filters I mean how does backpropagation happens to get the best filter or weights please help I am not able to understand this
Hi, really a great tutorial on CNN, small doubt suppose i download a bird image from the net and reshape it into 32X32, how can i feed that into this model to get the classified value.
At 5:26 , a = [5,3,7,5,9,7] but in within the blue frame it shows a = [5,3,2,5,9,7] ie the 3rd number is different. Is that a typo? if not, how did 7 change to 2?
Hi Ashna, you can check out this link to learn about : www.researchgate.net/publication/308809184_CNN_based_common_approach_to_handwritten_character_recognition_of_multiple_scripts: CNN based common approach to handwritten character recognition.
HI simplilearn i found this video educated and relevant please can you let have the codes and other relevant materials for the lecture. please how can i process a sequence of geospatial images?
Hey, thank you for appreciating our work. We are glad to have helped. Do check out our other tutorial videos and subscribe to us to stay connected. Cheers :)
Hello Sir. Thanks for the great video. I am facing problem whenever i develop CNN network it shows accuracy around 90 to 95%.But when i reused it with same data its accuracy decreases. It doesn't shows same result. Will you kindly help me?
A ReLU activation function provides non-linearity in convolutional layer. In the feature map, ReLU replaces all the negative pixels to zero. This will convert all the black pixels to non-negative values.
I have started exploring deep learning recently. To my surprise, I can only find tutorials about deep learning/CNN/LSTM/ RNN on youtube that are meant only for image classification. How does one implement it for structured data? For example, suppose I have a .csv file having data matrix of 200000 observations and some 1000 - 2000 features, how does one implement such a problem?
Hi Shweta, We do have videos on Deep Learning where the demo is based on structured data. Refer to the below links: ruclips.net/video/E8n_k6HNAgs/видео.html ruclips.net/video/_NMI8peAmNA/видео.html
Excellent lecture. Highly appreciated. Can you please add a lecture on depth prediction from a single RGB image using CNN? And various kinds of filters that are used in CNN for feature extraction specifically for depth prediction of various obstacles in a single RGB image. Regards
Hi Usama, thanks for the kind comment. We are glad to have helped. We acknowledge your suggestion and we will definitly transfer this to the relevant team. Do show your love by subscribing our channel using this link: ruclips.net/user/Simplilearn and don't forget to hit the like button as well. Cheers!
Hi Woodruff, thanks for the kind comment. We are glad to have helped. You can use our content for your presentation without editing the slides and it should be strictly for non-commercial use. Hope that helps!
In the first layer, we are applying a convolution filter operation using the function "convolution_layer()". Here, we are providing the size of the filter, the different channels and the number of pixels. To the same 1st layer, we also apply a max pooling layer. Hope this helps!
Hi Patrick, thanks for appreciating our work. We have shared the dataset to your mail ID. Do show your love by subscribing our channel using this link: ruclips.net/user/Simplilearn and don't forget to hit the like button as well. Cheers!
"Hi Krishna, CNN is mainly used for image classification and recognition task. The convolution layer in CNN identifies and extracts the features in the dataset. ANN is not a algorithm to solve a problem."
Outstanding video! One question: Is it possible to analyze whole RUclips videos frame by frame with CNNs? What do you think about the analysis of video data with ML approaches? Do you thinik there are nice applications besides Autonomous driving?
"Hi Grenke, It is possible to analyze any video. There are lot of other applications in Deep Learning apart from Self driving cars. Watch this video to know more: ruclips.net/video/1LxmmF88fDw/видео.html
Convolutional neural network works on gray scale images.Hope so.Raw Rgb image is given as input to train neural network,then rgb is converted to gray scale images.Then how the color is trained.if it an image, it learn the features of the image.It is simply only colors.How it works??will you clarify my doubt
"Hi Gopikha, In CNN, the images are fed in the form of color pixels. The model trains itself based on these pixels and identifies each part from the image. These pixels in variable intensity and size."
Thank You very much for a fantastic detailed CNN walkthrough. Would you mind sharing the code and dataset to my mailid. What are the other possible activation functions apart from RELu, Did I miss to notice the Filter Stride explanation.
Hello Mengop, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that also. Hope that helps.
Video is awesome to understand CNN, but in the practical part, the first 7 input and output is not there in the video. Kindly help me out, how to import data, from where data can be downloaded?
Hello, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we can send the requested dataset promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that too. Hope that helps.
To predict your own image using CNN, you need to have a huge collection of your pictures (around 40k). Also, the image sizes should not be very large, otherwise it would take a lot of time to process the images. Make sure the dimensions of the images do not very much. Then, use any simple CNN model (MNIST preferably) to build the model. Hope that helps!
Who (and how) designs the filters in a convnet ? Can filter design be automated ? It is said that AlphaZero Chess trained itself without any human interference , but if Alpha uses CNN those filters must have been created by Alpha itself … like treating a filter as another parameter.
Hi, I am a newbie on Machine Learning, let alone Deep Learning.. Thank you very much for the tutorial.. I really learnt a lot watching it.. I'm just wondering if you try to input a 32x32 image of a thing not classified as one of the 10 labels to the model, what would happen? And how do you do that?
Thanks by the video.I am a ecuadorian ingenier, I am starting to study CNN, I need classify images teen and no teen, Do you have any example?, help me please...
Hello Sergio, thanks for viewing our tutorial and we hope it is helpful. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly.
Hello Karthikeyan, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we could send the requested source code of CNN promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that also. Hope that helps.
good explanation. Thank you so much.. I have one doubt, in line 29, conv2 layer, how do you determine 64 ? is there any calculation over there or we can give any ? pls explain.
Hello, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we can send the requested dataset promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that too. Hope that helps.
Hi Kapil, you can check out the slides using this link: www.slideshare.net/Simplilearn/convolutional-neural-network-tutorial-cnn-how-cnn-works-deep-learning-tutorial-simplilearn.
Hi Payal, thanks for watching our tutorial. It would be great if you could provide your email ID so that we can send the source code and screenshots of the steps where you need our assistance. Thanks.
Hi Abhilash, thanks for watching our tutorial. We have sent the requested dataset to your mail ID. Do subscribe to our channel and get our new video updates directly into your email. If you have any questions related to these videos, you can post in the comments section, we will clear your queries/doubts.
Hi Mohammad, thanks for watching our tutorial. We have sent the requested source code to your mail ID. Do subscribe to our channel and get our new video updates directly into your email. If you have any questions related to these videos, you can post in the comments section, we will clear your queries/doubts.
Hello Russo, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that also. Hope that helps.
Hi and thanks for the great video! I'm having trouble understanding what's being done in the ReLU layer. If i base myself on the convolved matrix in the convolution layer demo, what happens to it in the ReLU layer?
ReLU layer in CNN uses a Rectified Linear Unit function to increase the non-linearity in the images you are training. An image contains a lot of non-linear features such as transition between the pixels, the borders and colors. ReLU enhances these features to indentify the images better.
Hello Jose, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that also. Hope that helps.
Hello, thanks for viewing our tutorial and we hope it is helpful. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly.
@@SimplilearnOfficial sir i need a more information about an neural network sir ,can I send my mail id sir? Another question is how to develop project in neutral network (deep learning)
Hi, it would be really helpful if you can take up a course on Deep learning using Tensorflow. For more info, click here: www.simplilearn.com/deep-learning-course-with-tensorflow-training.
Hey Prasen, thank you for appreciating our work. We are glad to have helped. Do check out our other tutorial videos and subscribe to us to stay connected. Cheers :)
Hello Santosh, thanks for viewing our tutorial and we hope it is helpful. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly.
Hi - awesome lecture, thanks! I was wondering if you could email me the data set and Jupyter files for this project, I'd love to test it out myself. Thanks!
Hello Noah, thanks for viewing our tutorial and we hope it is helpful. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly.
Hi, thanks for watching our video. We have sent the requested dataset to your mail ID. Do show your love by subscribing to our channel using this link: ruclips.net/user/Simplilearn and don't forget to hit the like button as well. Cheers!
Thanks, great video, I would like to understand in deep the CNN. Do you belive is posible to share the source and the data set? With the source code for sure I could debug and then go line by line to understand the details, for sure the video introduce to me in the best way, however I would like to go more deeper. BR, Saul Alquicira
Hello Saul, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that also. Hope that helps.
Such a great lecture but am i the only one who have problems with audio idk I don't mean to hurt the person who's explaining but may be something wrong with equipment the voice is kind of not going well with my ears
Hi, thank you for watching our video and for the honest feedback. We will definitely look into this. Do subscribe, like and share to stay connected with us. Cheers :)
Hello Risna, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that also. Hope that helps.
Hi Risna, thanks for sharing your email ID. We have forwarded the requested dataset to your mail ID. Do show your love by subscribing our channel using this link: ruclips.net/user/Simplilearn and don't forget to hit the like button as well. Cheers!
Hello Sir, Best explanation of CNN concepts. I'm doing my research in CNN expression recognition. It would be a great help if you send me the object detection code using CNN in Matlab. I am using real time videos as input to my classifier.
WooHoo! We are so happy you love our videos. Please do keep checking back in. We put up new videos every week on all your favorite topics. All the best for your research and please share your email ID to get the dataset. Cheers!
Thanks for your exciting lectures, but in your code in Jupyter notebook, you just from 1- to 8 but the 1-7 is most important aspect to majority of of who are new in the feed, but all the same thanks, we would figure it out how to recap the 1-7 but we can from slide
Hi Benya, thanks for checking out our tutorial and for sharing the information. Do subscribe to our channel by clicking this link ruclips.net/user/Simplilearn and stay posted for upcoming tutorials. Cheers!
Hi Suleyol, thanks for watching our video. We have sent the requested source code to your mail ID. Do show your love by subscribing our channel using this link: ruclips.net/user/Simplilearn and don't forget to hit the like button as well. Cheers!
Great tutorial ! Thanks a lot. That would be great to show how to use the model (make a prediction with an image). And eventually how to save the model for reuse
We are glad you found our video helpful, Emmanuel. Like and share our video with your peers and also do not forget to subscribe to our channel for not missing video updates. We will be coming up with more such videos. Cheers!
Hello Nani, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that also. Hope that helps.
Hi Piyush, thanks for checking out our tutorial and for sharing the feedback. Do subscribe to our channel to stay posted on upcoming tutorials. Cheers!
Hello Avinav, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we could send the requested dataset & source code promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that also. Hope that helps.
I know it's a long shot but can you actually show us how to make a CNN from scratch? I mean, making all the layers and connections by yourself instead of using external libraries and I would really prefer if you did it in C++. I have yet to see anyone try to implement a CNN by themselves using just C++ and not any external libraries like tensorflow etc. It seems as if it is just too difficult even though the theoretical portion makes it look like it can easily be done using an object-oriented approach but no one seems to even try it.
Hi Sarim, thanks for watching our video and sharing your suggestion. We will definitely look into this and share it with our team. Do subscribe, like and share to stay connected with us. Cheers :)
Great tutorial but I'm still confused about how we determine which matrix filter to use (in the convolution step). Is it randomly initialized then optimized as you train the network? Any chance you could shed some light on that? Additionally if you could email the code/dataset to torlinskig@gmail.com that would be a great help. Cheers!
Matrix filter is not applied randomly. In the convolution layer, max pool filter is applied to extract the maximum pixel value from a given input image. Please send the code and the dataset to the mentioned email id: kennet.rajan@simplilearn.net.
Hello Shruthi, thanks for viewing our tutorial and we hope it is helpful. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly.
Hello, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we can send the requested dataset promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that too. Hope that helps.
"🔥Caltech Post Graduate Program In AI And Machine Learning - www.simplilearn.com/artificial-intelligence-masters-program-training-course?Jy9-aGMB_TE&Comments&RUclips
🔥IITK - Professional Certificate Course in Generative AI and Machine Learning (India Only) - www.simplilearn.com/iitk-professional-certificate-course-ai-machine-learning?Jy9-aGMB_TE&Comments&RUclips
🔥Purdue - Post Graduate Program in AI and Machine Learning - www.simplilearn.com/pgp-ai-machine-learning-certification-training-course?Jy9-aGMB_TE&Comments&RUclips
🔥IITG - Professional Certificate Program in Generative AI and Machine Learning (India Only) - www.simplilearn.com/iitg-generative-ai-machine-learning-program?Jy9-aGMB_TE&Comments&RUclips
🔥Caltech - AI & Machine Learning Bootcamp (US Only) - www.simplilearn.com/ai-machine-learning-bootcamp?Jy9-aGMB_TE&Comments&RUclips"
Machine Learning is the Future and yours can begin today. Comment below with you email to get our latest Machine Learning Career Guide. Let your journey begin.
Do you have any questions on this topic? Please share your feedback in the comment section below and we'll have our experts answer it for you.
Thanks for watching the video. Cheers!
hey , i was training the model for 1500 times instead of 500 and i reached an accuracy of .6780.. how do u explain thay accuracy and also when predicting the label of a particular image it is giving a totally random label instead of right one....so what do u think is the problem ...well also i am initializing the variables again during testing of the particular image...is that why i am getting totally wrong output?
Hi Yogeesh, 0.67 is a pretty good accuracy if you attain it with just 500 iterations. Try running it for 1000 iterations. The precision will certainly increase. Please don't initialize the variables repeatedly while testing an image.
@@SimplilearnOfficial hey , i got this accuracy after 1500 epochs... and also how to predict an image's label after training it? cause i fed an image into the network and a label and tried running the network but it showed the error as "attempting to use uninitialized variables." so my question is how to load or use the trained model weights and predict image since i am not using keras ?
Hi. First of all thank you very much for such a great video. The question is how yo chose the filter. At the begining you said that you will talk about it, but you didnot. May be i missed. ??!?
Please initialize the variable you want to predict and then run it. Use Jupyter notebook and TensorFlow to do it. Keras is now built on top of TensorFlow.
sir it was a great video and was extremely helpful for me to learn cnn from scratch
Glad to hear that! Do subscribe to our channel and stay tuned for more.
Hello how do we obatain the filters I mean how does backpropagation happens to get the best filter or weights please help I am not able to understand this
Hi, really a great tutorial on CNN, small doubt suppose i download a bird image from the net and reshape it into 32X32, how can i feed that into this model to get the classified value.
At 5:26 , a = [5,3,7,5,9,7] but in within the blue frame it shows a = [5,3,2,5,9,7] ie the 3rd number is different. Is that a typo? if not, how did 7 change to 2?
Hi, Thanks for bringing this to our notice. This is certainly a typo error.
@@SimplilearnOfficial I too noticed this, thought typo.
@@mengop Good eye! Our mistake.
I'm doing handwritten alphabet recognition using CNN.could you help me in this?
Hi Ashna, you can check out this link to learn about : www.researchgate.net/publication/308809184_CNN_based_common_approach_to_handwritten_character_recognition_of_multiple_scripts: CNN based common approach to handwritten character recognition.
HI simplilearn i found this video educated and relevant please can you let have the codes and other relevant materials for the lecture. please how can i process a sequence of geospatial images?
"Hi Hassan,
Here is a link that will help you understand about CNN www.simplilearn.com/tutorials/deep-learning-tutorial/convolutional-neural-network"
greate tutorial ,thank you very much! how to use CNN for face recognition please give me matlab code
Hey, thank you for appreciating our work. We are glad to have helped. Do check out our other tutorial videos and subscribe to us to stay connected. Cheers :)
Hello Sir. Thanks for the great video. I am facing problem whenever i develop CNN network it shows accuracy around 90 to 95%.But when i reused it with same data its accuracy decreases. It doesn't shows same result. Will you kindly help me?
Glad you enjoyed our video! We have a ton more videos like this on our channel. We hope you will join our community!
in x.reshape(train_len,3,32,32).transpose(0,2,3,1) 32*32 is the size of the actual image? and what if the size of my image is 32*64 or something else?
@12:20 What does negative pixels refer to, since the pixels are only 1 and 0?
A ReLU activation function provides non-linearity in convolutional layer. In the feature map, ReLU replaces all the negative pixels to zero. This will convert all the black pixels to non-negative values.
I have started exploring deep learning recently. To my surprise, I can only find tutorials about deep learning/CNN/LSTM/ RNN on youtube that are meant only for image classification.
How does one implement it for structured data?
For example, suppose I have a .csv file having data matrix of 200000 observations and some 1000 - 2000 features, how does one implement such a problem?
Hi Shweta,
We do have videos on Deep Learning where the demo is based on structured data. Refer to the below links:
ruclips.net/video/E8n_k6HNAgs/видео.html
ruclips.net/video/_NMI8peAmNA/видео.html
Excellent lecture. Highly appreciated. Can you please add a lecture on depth prediction from a single RGB image using CNN? And various kinds of filters that are used in CNN for feature extraction specifically for depth prediction of various obstacles in a single RGB image.
Regards
Hi Usama, thanks for the kind comment. We are glad to have helped. We acknowledge your suggestion and we will definitly transfer this to the relevant team. Do show your love by subscribing our channel using this link: ruclips.net/user/Simplilearn and don't forget to hit the like button as well. Cheers!
Hi friend, which library you are using for this tutorial?
Hi Nelson, we use tensorflow library for this tutorial.
Great lecture, Next week i will do a presentation about CNN on powerpoint can i use screen capture for some illustrations ?
Hi Woodruff, thanks for the kind comment. We are glad to have helped. You can use our content for your presentation without editing the slides and it should be strictly for non-commercial use. Hope that helps!
Can you please explain how the filters are obtained in the first step of convolution layer?
In the first layer, we are applying a convolution filter operation using the function "convolution_layer()". Here, we are providing the size of the filter, the different channels and the number of pixels. To the same 1st layer, we also apply a max pooling layer. Hope this helps!
thank you so much for the tutorial sir helped me a lot in understanding CNN.
Hi Patrick, thanks for appreciating our work. We have shared the dataset to your mail ID. Do show your love by subscribing our channel using this link: ruclips.net/user/Simplilearn and don't forget to hit the like button as well. Cheers!
why do we use convolution ??? why not just simple ANN in case of image ?? main question is what is need of convolution in CNN?? please Answer....
"Hi Krishna,
CNN is mainly used for image classification and recognition task. The convolution layer in CNN identifies and extracts the features in the dataset. ANN is not a algorithm to solve a problem."
Outstanding video! One question: Is it possible to analyze whole RUclips videos frame by frame with CNNs? What do you think about the analysis of video data with ML approaches?
Do you thinik there are nice applications besides Autonomous driving?
"Hi Grenke,
It is possible to analyze any video. There are lot of other applications in Deep Learning apart from Self driving cars. Watch this video to know more: ruclips.net/video/1LxmmF88fDw/видео.html
Awesome Covers all my doubts in this lecture
Convolutional neural network works on gray scale images.Hope so.Raw Rgb image is given as input to train neural network,then rgb is converted to gray scale images.Then how the color is trained.if it an image, it learn the features of the image.It is simply only colors.How it works??will you clarify my doubt
"Hi Gopikha,
In CNN, the images are fed in the form of color pixels. The model trains itself based on these pixels and identifies each part from the image. These pixels in variable intensity and size."
Thank You very much for a fantastic detailed CNN walkthrough. Would you mind sharing the code and dataset to my mailid. What are the other possible activation functions apart from RELu, Did I miss to notice the Filter Stride explanation.
Hello Mengop, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that also. Hope that helps.
where did "data_batch1" at 24:27 come from?? did i miss a part of a video??
He skips over it
Video is awesome to understand CNN, but in the practical part, the first 7 input and output is not there in the video. Kindly help me out, how to import data, from where data can be downloaded?
Hello, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we can send the requested dataset promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that too. Hope that helps.
Hi there! I want to predict my own image after training CNN . How to do that? please help!
You'll obviously need a lot of your own images to train the CNN.
To predict your own image using CNN, you need to have a huge collection of your pictures (around 40k). Also, the image sizes should not be very large, otherwise it would take a lot of time to process the images. Make sure the dimensions of the images do not very much. Then, use any simple CNN model (MNIST preferably) to build the model. Hope that helps!
Thanks for your reply, Waqar.
Who (and how) designs the filters in a convnet ? Can filter design be automated ?
It is said that AlphaZero Chess trained itself without any human interference , but if Alpha uses CNN those filters must have been created by Alpha itself … like treating a filter as another parameter.
Hi, I am a newbie on Machine Learning, let alone Deep Learning.. Thank you very much for the tutorial.. I really learnt a lot watching it.. I'm just wondering if you try to input a 32x32 image of a thing not classified as one of the 10 labels to the model, what would happen? And how do you do that?
"Hi Windia,
The CNN model will classify it as a random label out of the 10 labels if that images has not been trained."
Thanks by the video.I am a ecuadorian ingenier, I am starting to study CNN, I need classify images teen and no teen, Do you have any example?, help me please...
Hello Sergio, thanks for viewing our tutorial and we hope it is helpful. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly.
Dope tutorial. Is it possible to train and predict colors using convolutional neural networks??
Yes, it is possible!
I like this video and the way it has been explained. But can someone help me with the code in the video to try with CIFAR-8 dataset.
Hello Karthikeyan, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we could send the requested source code of CNN promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that also. Hope that helps.
good explanation. Thank you so much.. I have one doubt, in line 29, conv2 layer, how do you determine 64 ? is there any calculation over there or we can give any ? pls explain.
Hi Shobana, when we use a 3x3 filter, we usually provide 32 or 64 nodes. There is no mandatory rule for this. Hope that helps!
Great video. 27:55 that is an emu lol...
Haha! You got a great vision!
The bird at 21.04 is a malabar hornbill?
Yes, you are right!
sir from where should i download dataset
Hi Dharavath, we have sent the requested dataset to your mail ID. Thanks.
Thank you for this very interesting cours, can you please send me the source code to play with it ? thank you :)
Hello, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we can send the requested dataset promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that too. Hope that helps.
@@SimplilearnOfficial my email ID karim.yahiatene.1987@gmail.com
Please upload the link of the ppt you used in it. I have my presentation. I will thank your for life.
Hi Kapil, you can check out the slides using this link: www.slideshare.net/Simplilearn/convolutional-neural-network-tutorial-cnn-how-cnn-works-deep-learning-tutorial-simplilearn.
Hello. Excelente. Does the code run in tf 2?
Yes it does!
Kindly share step 1, 2 and 3 of CIFAR 10 dataset preprocessing step of implementation of CNN. I will be very thankful to you.
Hi Payal, thanks for watching our tutorial. It would be great if you could provide your email ID so that we can send the source code and screenshots of the steps where you need our assistance. Thanks.
@@SimplilearnOfficial Thanks for this video! Please send me the code for it to my email id: abhilashpandurangan@gmail.com
Hi Abhilash, thanks for watching our tutorial. We have sent the requested dataset to your mail ID. Do subscribe to our channel and get our new video updates directly into your email. If you have any questions related to these videos, you can post in the comments section, we will clear your queries/doubts.
Simplilearn Thanks for this great video... Could you please send me the dataset. My email ID is mohammad.ghanim1982@gmail.com
Hi Mohammad, thanks for watching our tutorial. We have sent the requested source code to your mail ID. Do subscribe to our channel and get our new video updates directly into your email. If you have any questions related to these videos, you can post in the comments section, we will clear your queries/doubts.
Hey thank you for this lecture! I’ve subscribed and liked! Anyway I can get the python file? That would really be appreciated!
Hello, thanks for viewing our tutorial. You can find your requested dataset in the video description. Hope that helps.
Great tutorial, thank you very much! Could you please send me the code and dataset for it? I'd greatly appreciate it.
Hello Russo, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that also. Hope that helps.
Thanks for your nice video
We are so grateful for your kind words. Also, subscribe to our channel and stay tuned for more videos. Cheers!
Hi and thanks for the great video! I'm having trouble understanding what's being done in the ReLU layer. If i base myself on the convolved matrix in the convolution layer demo, what happens to it in the ReLU layer?
ReLU layer in CNN uses a Rectified Linear Unit function to increase the non-linearity in the images you are training. An image contains a lot of non-linear features such as transition between the pixels, the borders and colors. ReLU enhances these features to indentify the images better.
This is great! Could you please send me the dataset?
Hello Jose, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that also. Hope that helps.
Awesome
Thank you! Please subscribe to Simplilearn channel and stay updated with all the latest trending technologies.
Any course materials are available sir
Hello, thanks for viewing our tutorial and we hope it is helpful. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly.
@@SimplilearnOfficial sir i need a more information about an neural network sir ,can I send my mail id sir? Another question is how to develop project in neutral network (deep learning)
Hi, it would be really helpful if you can take up a course on Deep learning using Tensorflow. For more info, click here: www.simplilearn.com/deep-learning-course-with-tensorflow-training.
@@SimplilearnOfficial thank you sir this is very use full for me sir
@@SimplilearnOfficial this course is very useful sir
Excellent work sir, you explained each and everything perfectly. Thanks a lot
Hey Prasen, thank you for appreciating our work. We are glad to have helped. Do check out our other tutorial videos and subscribe to us to stay connected. Cheers :)
Hey, can you please give the dataset??
Hi Iliyas, could you share your email ID so that we can send the dataset promptly? Cheers!
hi. simplilearn
can u send me datasets and the code of this video??
Hello Santosh, thanks for viewing our tutorial and we hope it is helpful. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly.
@@SimplilearnOfficial email :: santoshtmp7@gmail.com
Hi - awesome lecture, thanks! I was wondering if you could email me the data set and Jupyter files for this project, I'd love to test it out myself. Thanks!
Hello Noah, thanks for viewing our tutorial and we hope it is helpful. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly.
@@SimplilearnOfficial "noah.sealy@dal.ca" (quotations not included in email), thanks!
Hi, thanks for watching our video. We have sent the requested dataset to your mail ID. Do show your love by subscribing to our channel using this link: ruclips.net/user/Simplilearn and don't forget to hit the like button as well. Cheers!
Thanks, great video, I would like to understand in deep the CNN.
Do you belive is posible to share the source and the data set?
With the source code for sure I could debug and then go line by line to understand the details, for sure the video introduce to me in the best way, however I would like to go more deeper.
BR,
Saul Alquicira
Hello Saul, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that also. Hope that helps.
@@SimplilearnOfficial my email is emilianato19@gmail.com
thanks
Such a great lecture but am i the only one who have problems with audio idk I don't mean to hurt the person who's explaining but may be something wrong with equipment the voice is kind of not going well with my ears
Hi, thank you for watching our video and for the honest feedback. We will definitely look into this. Do subscribe, like and share to stay connected with us. Cheers :)
may I get the dataset and source code?
Hello Risna, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that also. Hope that helps.
@@SimplilearnOfficial risnahasanah456@gmail.com
Hi Risna, thanks for sharing your email ID. We have forwarded the requested dataset to your mail ID. Do show your love by subscribing our channel using this link: ruclips.net/user/Simplilearn and don't forget to hit the like button as well. Cheers!
very good
Hi, we appreciate the kind comment! enjoy!
Hello Sir, Best explanation of CNN concepts. I'm doing my research in CNN expression recognition. It would be a great help if you send me the object detection code using CNN in Matlab. I am using real time videos as input to my classifier.
WooHoo! We are so happy you love our videos. Please do keep checking back in. We put up new videos every week on all your favorite topics. All the best for your research and please share your email ID to get the dataset. Cheers!
Thanks for your exciting lectures, but in your code in Jupyter notebook, you just from 1- to 8 but the 1-7 is most important aspect to majority of of who are new in the feed, but all the same thanks, we would figure it out how to recap the 1-7 but we can from slide
Hi Benya, thanks for checking out our tutorial and for sharing the information. Do subscribe to our channel by clicking this link ruclips.net/user/Simplilearn and stay posted for upcoming tutorials. Cheers!
@@SimplilearnOfficial Thank you sir. please kindly share the source code to me: wyjoe2@yahoo.co.uk
Hi Suleyol, thanks for watching our video. We have sent the requested source code to your mail ID. Do show your love by subscribing our channel using this link: ruclips.net/user/Simplilearn and don't forget to hit the like button as well. Cheers!
Could you please send me code at hereisswati@gmail.com
Great tutorial ! Thanks a lot. That would be great to show how to use the model (make a prediction with an image). And eventually how to save the model for reuse
We are glad you found our video helpful, Emmanuel. Like and share our video with your peers and also do not forget to subscribe to our channel for not missing video updates. We will be coming up with more such videos. Cheers!
would be send dis program
Hello Nani, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that also. Hope that helps.
Your voice sounds as if you are talking on phone, kindly consider replacing your microphone
Hi Piyush, thanks for checking out our tutorial and for sharing the feedback. Do subscribe to our channel to stay posted on upcoming tutorials. Cheers!
hello sir ..can i have the dataset and source code at my mail....
Hello Avinav, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we could send the requested dataset & source code promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that also. Hope that helps.
I know it's a long shot but can you actually show us how to make a CNN from scratch? I mean, making all the layers and connections by yourself instead of using external libraries and I would really prefer if you did it in C++. I have yet to see anyone try to implement a CNN by themselves using just C++ and not any external libraries like tensorflow etc. It seems as if it is just too difficult even though the theoretical portion makes it look like it can easily be done using an object-oriented approach but no one seems to even try it.
Hi Sarim, thanks for watching our video and sharing your suggestion. We will definitely look into this and share it with our team. Do subscribe, like and share to stay connected with us. Cheers :)
The mic sounds like shit. I think you may need to rerecord it.
We are sorry about that Ahmed, we will share the feedback with the relevant department.
actually the accuracy is to bad it is 50% it means that the prediction may running under randomness fool
Can you provide the dataset and notebook code to me
Great tutorial but I'm still confused about how we determine which matrix filter to use (in the convolution step). Is it randomly initialized then optimized as you train the network? Any chance you could shed some light on that? Additionally if you could email the code/dataset to torlinskig@gmail.com that would be a great help. Cheers!
Matrix filter is not applied randomly. In the convolution layer, max pool filter is applied to extract the maximum pixel value from a given input image.
Please send the code and the dataset to the mentioned email id: kennet.rajan@simplilearn.net.
😮. 😮
Thank you for this amazing lecture!!! Could you please mail me the notes of CNN @simplilearn
Hello Shruthi, thanks for viewing our tutorial and we hope it is helpful. It would be helpful if you will provide your email ID to us so that we could send the requested dataset promptly.
Great tutorial, thank you very much! Could you please send me the code and dataset for it? I'd greatly appreciate it.
Hello, thanks for viewing our tutorial. It would be helpful if you will provide your email ID to us so that we can send the requested dataset promptly. On the off chance that you need your email ID to be kept hidden from others, we can do that too. Hope that helps.