Этот курс научит вас использовать Keras, API нейронной сети, написанный на Python и интегрированный с TensorFlow. Мы научимся подготавливать и обрабатывать данные для искусственных нейронных сетей, строить и обучать искусственные нейронные сети с нуля, строить и обучать сверточные нейронные сети (CNN), выполнять тонкую настройку и передавать обучение и многое другое! ⭐️? СОДЕРЖАНИЕ КУРСА ?⭐️ ⌨️ (00:00:00) Добро пожаловать на этот курс ⌨️ (00:00:16) Введение в курс Keras ⌨️ (00:00:50) Предварительные требования к курсу ⌨️ (00:01:33) DEEPLIZARD Deep Learning Путь ⌨️ (00:01:45) Ресурсы курса ⌨️ (00:02:30) О Keras ⌨️ (00:06:41) Keras с TensorFlow — обработка данных для обучения нейронной сети ⌨️ (00:18:39) Создание искусственного Нейронная сеть с API Keras от TensorFlow ⌨️ (00:24:36) Обучение искусственной нейронной сети с API Keras от TensorFlow ⌨️ (00:30:07) Создание проверочного набора с API Keras от TensorFlow ⌨️ (00:39:28) Прогнозы нейронной сети с API Keras от TensorFlow ⌨️ (00:47:48) Создание матрицы путаницы для прогнозов нейронных сетей ⌨️ (00:52:29) Сохранение и загрузка модели с API Keras от TensorFlow ⌨️ (01:01:25) Подготовка изображения для CNN с Keras API от TensorFlow ⌨️ (01:19:22) Создание и обучение CNN с помощью Keras API от TensorFlow ⌨️ (01:28:42) Прогнозирование CNN с помощью Keras API от TensorFlow ⌨️ (01:37:05) Создание точно настроенной нейронной сети с TensorFlow Keras API ⌨️ (01:48:19) Обучение точно настроенной нейронной сети с помощью TensorFlow Keras API ⌨️ (01:52:39) Прогнозирование с помощью точно настроенной нейронной сети с помощью TensorFlow Keras API ⌨️ (01:57:50) MobileNet Классификация изображений с помощью TensorFlow Keras API ⌨️ (02:11:18) Обработка изображений для точной настройки MobileNet с помощью TensorFlow Keras API ⌨️ (02:24:24) Тонкая настройка MobileNet на пользовательском наборе данных с помощью TensorFlow Keras API ⌨️ (02:38) :59) Увеличение данных с помощью TensorFlow’ Keras API ⌨️ (02:47:24) Коллективный разум и DEEPLIZARD HIVEMIND ⭐️? РЕСУРСЫ СООБЩЕСТВА DEEPLIZARD ?⭐️ ? Ознакомьтесь с постом в блоге и другими ресурсами для этого курса: ?
? СКАЧАТЬ ДОСТУП К ФАЙЛАМ КОДА ? Доступно для членов коллективного разума deeplizard: ?
? Поддержите коллективный разум, присоединяйтесь к коллективному разуму дипящеров: ?
? Привет, мы Крис и Мэнди, создатели Deeplizard! ? СМОТРИТЕ НАШ ВЛОГ: ?
? Подписывайтесь на deeplizard: YouTube:
? deeplizard использует музыку Кевина Маклауда ?
❤️ Пожалуйста, используйте знания, полученные из контента Deeplizard, во благо, а не во зло. — Научитесь программировать бесплатно и получите работу разработчика:
Прочтите сотни статей по программированию:
И подписывайтесь на новые видео о технологиях каждый день:
source
According to researches by the adsmember team
Hi everyone! Hope you all learn and gain from this course! Come check out the other deep learning courses available on our channel! ❤️?
darkmode pls
1:04:54 / 2:47:54
I am getting validation accuracy on the 1st data set that is clinical trial as , 64.84% exactly for 6 epochs and then jumps to 100%. no stepwise increase . Any idea why?. Rerunning didn't change anything
I can't concentrate because I'm looking at your (beautiful) freckles 🙂
Thanks!
Clear communicator. Interesting lessons. Good vid
I guess i am learning keras now…
25:53
Hello, can you put your personal github address in the description section of the YouTube channel?
when i use fit focntion it's return nen
As I go through some readings, they say that the test folder should not contain several sub-folders and the images under test should not be labeled. In my cases, I am having some IndexError but still I could not figure it out. Anyone has any idea? I have basically 4 classes and I am creating for different sub-folders by labeling the images and files under test file. In this video, she does the same things. Interestingly, when I run model.fit, it says "IndexError: index 1 is out of bounds for axis 2 with size 1"
Hey, can you please tell how to assign sample_weights in model.compile for imbalanced data?
Thankssssssssssssssssss!!!!!!!!
Thx
HI Please update the code:
model.fit(x = train_batches, validation_data = valid_batches, epochs = 10, verbose = 2) # conv fit
Error: TypeError: __array__() takes 1 positional argument but 2 were given
Not only was I not distracted by her beautifulness, I was actually able to understand everything she said. Thank you!
02:46:04 Nice.
I guess I am sick but… Am I the only that sees the weird vibes of the set up? why the bed? Is this made on purpose?
The machine learning tutorial space is a literal landfill of garbage, thousands of slightly different versions of this and that.
Given the fact that these models can be created with 10 lines of code, you would think there would be a really simple start to finish showing data to results in 10 mins, but all there is is 3 hour mess videos like this.
I cant describe in words how much this video helped me with my research project! You are a great teacher, Thankyou so much!
Nice
Hi Mandy, I tried training neural network for the clinical trials example @28:47. I am getting Nan for loss and 0.5 for accuracy. I ran the exact lines of code you did. What is the issue? BTW, I am not using the three lines of code to use the GPU.
too much repeated content, but good side of repeat is it enforces memory. so overall pretty good
There's a great book out there that does an awesome job at explaining every single line of python code for deep learning and artificial intelligence. It's called "Artificial Intelligence and Deep Learning with Python Every Line of Code Explained "
Abella dangers
The bed as a background for DL tutorial is not the best idea.
@ 1:25:00 I am getting error Error "when checking input: expected conv2d_3_input to have shape (224, 224, 3) but got array with shape (244, 244, 3)" How to resolve this?
Hi, I'm new to the subject and eager to understand. ☺️ Watching the episode Train an Artificial Neural Network (circa 29:00) and I understand the training but don't really understand whether the model there was just trained "theoretically", so to say, or was there any actual data input present, as I don't remember that being discussed. I understand that information is being added gradually and I can live with not knowing everything at once ?, but this seems important – can I run a model without any actual input to check how it works or is it just not being shown. Sorry if I'm misunderstanding everything ??
Why the test samples has to be in the same format as train samples ?
Test samples can be random nah
Such a detailed and amazingly designed course. Covered every question I had in mind!.
Finally, a southern accent.
awesome quality video!!
Crossentropy
One of the best videos on Keras Deep Learning. Thanks for your wonderful teaching.
Dear Lady DeepLizard,
Thank you so much for the energy, time and thought you've putting this course! I have benefited a lot from your channel,
Thanks for a well prepared, well organized, professional presentation. GREATLY appreciated
Where can I download resources?
I love her deep freckles. ?
I am not sure why but… somehow I feel like she is an AI.
"alright, that's it for the manual labor" at the one hour mark haha… i love it.
completely off topic, but mandy looks a lot like Gal Gadot. The course was awesome though.
Can't get the fine tuning to work. The accuracy and validation accuracy are quite high for the vgg16 model but when I get it to predict and plug it into the confusion matrix plotting function, it shows something completely off.
This tutorial is great but it's too focused on image data.
hi, ho do i change the input shape in vgg16?
Awesome, Mandy you rock
ثثثث ياسبوكة
u accetly look like abbela danger 1000000000000000%
Sorry for being little critical here even though it's helpful tutorial some things were just left hanging…
To the people that actually try to repeat the tutorial (because I don't think there many) I have to say you have to use your logic to figure things out and not take everything for granted because it will not work she just messed up some parts…
For example: when she shuffles the data they have to be shuffled in union with the answers because your model will be accurate in training but 50% accurate on the unseen answers since it's like you lost the answers… Don't ask me if I messed that part by blindly following and wasting half an hour…
So you have to find your way to do this she leaves that part out and I don't know how she gets results…
Maybe keras at the time did it by default and currently not? Or I missed something idk…
Also some of the functions don't work if you follow on your computer's compiler… Reshape, Radom. I mean they work but you have to specify the module. I know some python masters will say you just don't know how to use python modules and I agree but this tutorial is not for the people knowing only less than two programming languages and starting out with keras.
Sorry, but it still helped to start somehow but it wasn't the best experience…
Thanks!