Thank you so much for this insightful tutorial on hyperparameter tuning using Keras Tuner! Your clear explanations and step-by-step guidance made the entire process easy to understand and implement. I've struggled with optimizing my models before, but your video has provided me with the tools and confidence to tackle hyperparameter tuning effectively. Keep up the fantastic work! 🙌
Hi sir, I recently come accross your 100days ML playlist. It looks lots of informative, I just wanted to ask is this your complete playlist? As I am seeing 66days and some maybe 5-6 other videos.
Hello Nitish! Great video. Only 1 doubt. If we add BatchNormalization() to the layers, it is giving a consolidated set of hyperparameters (not for every layer but one set for all the layers) which is somehow performing better without BatchNormalization. Wanted to know how can we introduce BatchNormalization and still get hyperparameters for every layer. Thank you!
Hello. What if I want to optimize the learning rate, the number of epochs or callbacks such as ReduceLROnPlateau? Can I use keras tuner for that? Or is it already included in the search procedure?
Hello dear, it is requested that kindly make video regarding Deep learning or pretrained models based ensemble techniques such as stacking , bagging, and boosting. Thanks dear
Sir pls make a video on k fold cross validation Sir maine abhut saara video dekha par usme acche se explain nahi Kiya gaya hai Aap ek video banao aur statistics par bhi video banao Apne playlist me dusre ka video daala hai Sir it's very important for me I am watching only your video , getting understand with your video only Humble request from honest student to best teacher
the keras model is not running completely in colab and triggering EXIT (we can see it in the video 59:41 ) and it is triggering exit, that's the reason why we are getting lesser accuracy sometimes
Thanks. I love ur teaching style. One question: from the majority results of keras tuner, it appears that it is not very useful as the validation accuracy continously decreased till the end of the video. So, if I am right, then what might be the rule of thumb for "starting architecture" of neural network. Best,
Hello Sir , I am really fan of your teaching . I referred many videos but nothing can match your teaching . just have a doubt with respect to this video shouldn't feature scaling be done after train test split
Thanks for a very insightful video. Very helpful. One question I would like to understand - the initial model with 2 layers and small number of nodes performed close to or better than the recommendation by the Keras Hypeparameter Tuner. The network recommended by Keras Hyperparameter is much bigger and more complex. Why is that? After all that effort, our initial model performed better. Any thoughts?
![CAMPปลิ้น | EP.83[2/2] ประสบการณ์ขนหัวลุกจากตำนานวงการผี!](http://i.ytimg.com/vi/T_kFecQw7Ds/mqdefault.jpg)
You are the real data scientist. you are providing the best content regarding data science as compared to another channels.
i was afraid of learning Deep learning. because of you it become possible. Thank you!
You are just awesom. I've learned many things from your videos. Thank you so much. Lot of Love from London (UK). 👌👌👌👌👌
london mai wathman hai kya?
stop bragging abt yrself
The Scientist term fits you, thoroughly knowledgeable and well-researched information. Thank you 🙏
you are the best in teaching sir, you are gift to us sir long live
Till now all your lectures all awesome in Deep learning series, but this video can be improved lot..
simple and better explanation , keep up the good work please 👍
Thank you so much for this insightful tutorial on hyperparameter tuning using Keras Tuner! Your clear explanations and step-by-step guidance made the entire process easy to understand and implement. I've struggled with optimizing my models before, but your video has provided me with the tools and confidence to tackle hyperparameter tuning effectively. Keep up the fantastic work! 🙌
Thank you! Best teacher ever..
Hi , By using this I got an accuracy of 79 percent and val_Accuracy of 80.5 percent , And really i learned something different.
yeah i am also confused here ,at 26:00
wha twas your previous accuracy/?
Sadly, this is called underfitting
Hey, I am learning so much from all these videos and implementing them as much as I can. Will try this technique too.
There is one alternative of you, the real gems.
You're doing gods work Nitish Sir.
Bhayya you teach very well ❤❤❤
Thank you Nitish !! 😃
wow great sir. God bless you sir.
Hi sir,
I recently come accross your 100days ML playlist. It looks lots of informative, I just wanted to ask is this your complete playlist? As I am seeing 66days and some maybe 5-6 other videos.
Hello Nitish! Great video. Only 1 doubt.
If we add BatchNormalization() to the layers, it is giving a consolidated set of hyperparameters (not for every layer but one set for all the layers) which is somehow performing better without BatchNormalization. Wanted to know how can we introduce BatchNormalization and still get hyperparameters for every layer.
Thank you!
Hey when will you start RNN and CNN. Also which one will you start first, really hoping to study these topics from your channel!!!
Hello.
What if I want to optimize the learning rate, the number of epochs or callbacks such as ReduceLROnPlateau? Can I use keras tuner for that? Or is it already included in the search procedure?
Sir plz make a video series on UMAP
feature Extraction technique
Amazing Explanation sir
I've learned many things from your videos
Hello dear, it is requested that kindly make video regarding Deep learning or pretrained models based ensemble techniques such as stacking , bagging, and boosting. Thanks dear
Sir mujhe iloc samjh nahi aa raha aapne konse video mein padaya hai ek baar pata dijiye🙏
Sir pls make a video on k fold cross validation
Sir maine abhut saara video dekha par usme acche se explain nahi Kiya gaya hai
Aap ek video banao aur statistics par bhi video banao
Apne playlist me dusre ka video daala hai
Sir it's very important for me I am watching only your video , getting understand with your video only
Humble request from honest student to best teacher
@CampusX Sir, where can I get the code written in this video? Have you posted it anywhere?
the keras model is not running completely in colab and triggering EXIT (we can see it in the video 59:41 ) and it is triggering exit, that's the reason why we are getting lesser accuracy sometimes
Thanks. I love ur teaching style.
One question: from the majority results of keras tuner, it appears that it is not very useful as the validation accuracy continously decreased till the end of the video. So, if I am right, then what might be the rule of thumb for "starting architecture" of neural network.
Best,
Please create a complete video on Time Series Forcasting using ARIMA Model🙏
brief info :
1. BEFORE TUNING our model(like with random neurons, activation function, optimizer), results were as follow:
loss: 0.4201 - accuracy: 0.7964 - val_loss: 0.4568 - val_accuracy: 0.8052
2. AFTER TUNING,
loss: 0.4576 - accuracy: 0.7883 - val_loss: 0.4793 - val_accuracy: 0.7922
😂😂
May be we can apply some early stopping technique as well, as we are running for so many epochs, causing overfitting
@@rajatchauhan4410 no bro here overfitting is not occuring so why you apply early stopping.
last me early stopping bhi laga skte hai na: after applying dropout and running model at last when keras_tuner is used to get layers and neurons both?
Sir, Good video & well explained, Thanks a lot.
kindly share the code of this, i am not getting expected output. Thank you
It is really helping me lot.
Thanks for the wonderful explanation.
Hello Sir , I am really fan of your teaching . I referred many videos but nothing can match your teaching . just have a doubt with respect to this video
shouldn't feature scaling be done after train test split
every thing possible 😎
from where can i learn hyper parameter tuning in PyTorch?
Thanks for a very insightful video. Very helpful. One question I would like to understand - the initial model with 2 layers and small number of nodes performed close to or better than the recommendation by the Keras Hypeparameter Tuner. The network recommended by Keras Hyperparameter is much bigger and more complex. Why is that? After all that effort, our initial model performed better. Any thoughts?
This was for small dataset but when working with large dataset keras turner help
Majedar
Amazing video, nice learning keras tuner
When i first did it gave me rms prop now its giving adam😮
Keep Going Sir
Nice video pls provide DL one notes
finished watching
we can use gridsearchcv also
Love it sir
brother if i used the same model for regression mean output layer is linear I am getting ccuracy as 0.0 what to do?
I guess we need to run it for more trails
is the procedure same for CNN?
If the optimum number of layers is getting set as 3 so why there are values till units8..it should be till units3 right?
Bro next video kb upload hogi ?
Sir, please update ML, DL and NLP playlists
Bro ANN kab tak complete hoga ? This month will it complete ?
Yes
God NITISH
Sir next video please jaldi upload karo
Thank You Sir.
you are the best
you did a great job
nice explanation, Thanks
Hey, how can I contact you. I am struggling with some guidence issue in starting my career
At the very end we get to 75% accuracy what else we can try to improve it further
thank you
tune search is not working . can anyone help.get hyperparameter shows list indx out of range
most of viewer not getting expected output (i also try more then 5 time ). sir please share this notebook code
Sir , does campusX provide any paid course ?
Awesome
mujhe lgta h 5 epoches bahut kam h checking k liye 10 - 12 epochs use karo fir better hyperparameter tuning hogi
ISI dataset par kaam Kiya may nay aik project par
Without 'hp' is giving best result 82%
Thanks sir
sir please share the link of this notebook
Admist mediocrity filled in the name of Data Science on Indian Internet, you stand out as an exception. Khush raho, bs ye playlist poori krdo!!!
Nice
Waha bhaiya harry potter ka book rakhee ho
59:52 {'units_0': 56, 'activation_0': 'relu', 'optimizer': 'adam'}
only this comes ! any help is much appreciated
Where is the notebook link sir
can you plz share your colab notebook
In the no of layers fine tuning i got 99% accuracy while in the last fine tuning i got only 65 %
Can someone explain this
best
can u provide colab file link so that we can practice
Bro, where can i find the code?
can you please share the code
Please provide the code as well.
where is code for this ?
Plz provide google colab notebook for cde
sir may you share this notebook
notebook link
sir please provide the source code. I got some error(output not correct) althrough the code is same
sie plz share the code of this video. its request to you.
Where is the code link 🙄
best best
Can you please add link of source code in description of the video
Hit like if you 👍👍👍👍👍
🙏
Sir can you provide the code?
where's source code?
Bro share file link
source code
before learnint AI and DEEP learning...Learn Data Analysis
Put hindi on the video title jeez
Wish i know your channel 2 years ago🥺🥺