You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
You now can train a really good image classification model on 25k images in less than 2 hours, entirely for free, on a Colab GPU. And it's dead easy. What a time to be alive I remember when you had to write your own CUDA and C++ and you needed a hardware budget
Here's a Colab that does just that. Download a zip file with 25k images. Get a model training in seconds, with all the best practices baked in. One hour and a half to convergence on the free Colab GPU.https://colab.research.google.com/drive/1umJnCp8tZ7UDTYSQsuWdKRhqbHts38AC …
This is great!
Things are moving in an extraordinarily fast way. We don't realize it but we are living in a very hot period of science.
I'm still learning this stuff trying to understand - so I have two questions about the model: 1. When you first enter the for loop you have an activation in series with another activation. What is the point of that? 2. When you leave the for loop there is no activation? Why not?
Thanks for this. Question: how would you go about class balancing?
Can the used model architecture be used on other datasets?
Ran this program on Windows laptop with GPU. Got error. Any info to fix this error please? @aureliengeron
train_ds = tf.keras.preprocessing.image_dataset_from_directory(
AttributeError: module 'tensorflow_core.keras.preprocessing' has no attribute 'image_dataset_from_directory'
Noob question but if you wanted to try Xception on CIFAR-10 what would you change? Would you upscale from 32x32 or change the model?
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
