Save and load a Keras model

🕒🦎 VIDEO SECTIONS 🦎🕒

00:30 Help deeplizard add video timestamps - See example in the description
06:51 Collective Intelligence and the DEEPLIZARD HIVEMIND

💥🦎 DEEPLIZARD COMMUNITY RESOURCES 🦎💥

👋 Hey, we're Chris and Mandy, the creators of deeplizard!

👉 Check out the website for more learning material:

💻 ENROLL TO GET DOWNLOAD ACCESS TO CODE FILES

🧠 Support collective intelligence, join the deeplizard hivemind:

🧠 Use code DEEPLIZARD at checkout to receive 15% off your first Neurohacker order
👉 Use your receipt from Neurohacker to get a discount on deeplizard courses

👀 CHECK OUT OUR VLOG:

❤️🦎 Special thanks to the following polymaths of the deeplizard hivemind:
Tammy
Mano Prime
Ling Li

🚀 Boost collective intelligence by sharing this video on social media!

👀 Follow deeplizard:

🎓 Deep Learning with deeplizard:

🎓 Other Courses:

🛒 Check out products deeplizard recommends on Amazon:

🎵 deeplizard uses music by Kevin MacLeod

❤️ Please use the knowledge gained from deeplizard content for good, not evil.

[leilajalali]

without a doubt you are one of the best instructor I have ever seen. I absolutely love your channel. well done! and thank you so much for your great help. You saved my life

[vibekdutta]

Wow! In love with ya videos, so brief and distinct! Awesome

[saltywizardflyingonafish]

Neatly organized explanation.
Explanation contains only what’s necessary to learn.

Perfect just perfect, you did well 👍🏻👍🏻👍🏻

[bobcrunch]

I wish I had found your videos a year ago. It would have saved me a lot of frustration!

[phitchaphalertsiravarameth]

This is very helpful, indeed. Thank you !!!

[mochamadwahyuhidayat]

Hey, newbie here. That was well explained and I thank you for the video! You've just got +1 subscriber.

[MuhammadArnaldo]

Thank you for the good content, it's really helpful...

[fatimaezzahra]

thank you soo much for this amazing video, i want to know if i can save to model and use it with a lot of data

[fossin]

Hey Great Videos very Simple way of explanation Videos are engaging and short
In this video, we are saving model architecture in json_string variable but we are not actually saving it to ROM
this way we will be only able to use that variable in a current notebook only.

[mohamedberrimi]

Thank you sooo much !!
Please i have noticed that the train acc is lower than the valid_acc, does this effect the model? and why is this happening ?
Thanks Again <3

[awaisahmad]

Thanks a lot

Very great explanation

[divya]

Hey this is a great video. I have a question I scaled my training features and saved my model and loaded into new notebook. I want to scale my new data features for prediction using the saved model but it is not allowing. What should I do?

[sergiocarrasco]

Hi,

Great Video, when I'm loading a model saved using the method outlined in the video my network seems to make different predictions for my test set compared to the predictions made when the same test set was presented at the end of training. Any idea what might be causing this?

[krishnaik]

I hope the when we are saving the weights it wont work with different weights right?

[JimmyCheng]

Another amazing video as always. Btw, I have a question regarding how to use the saved weights on a new model. The new model architecture has to be the same as the old one right? otherwise how does the weights even apply. In that case, what is the use case for saving only the weights? Cuz seems retaining the architecture is almost necessary.

[allieubisse]

I'm glad I've found your videos before studying any material for deep learning

[pranavravuri]

quick question, can we pause the training and continue it by loading the .h5 file later? is it possible? how?

[RedShipsofSpainAgain]

Is there a reason you'd only want to save just the weights, or just the architecture, or just the model, as opposed to saving the full model (i.e. architecture, weights, loss function, optimizer, optimizer state)?

Is there a storage/size consideration for large models that would give one pause to save/store the full model? Thanks, great video. Very helpful

[politicalcanadian]

I am confused about scaling when it comes to saving and loading a model. Do you have a video on that? I am loading and running my model. Its output is scaled. Do i inverse the scale before saving the model or after loading the model, or after running on the test data. It must be earlier in the sequence, otherwise, how does the model know the original scale of the features?

[janiobachmann]

Thanks for the tutorial! I have one question, what if we have 2 hidden layers in our model and I want to save the weights of only one hidden layer and pass it to one of the hidden layers of my new model. Is there someway for calling the weights of one specific hidden layer? I know we are able to do this in TensorFlow, I just wanted to know if it is possible in Keras.