In this part of the lesson, we'll learn how to define the training function for our model, as well as connecting to a GPU engine FOR FREE! I've also included lots of helpful code inside the starter notebook to account for checkpoint saving, checkpoint loading, loss plotting, and prediction, so please make sure you copy/download it from the link below! Even though the model is not nearly as accurate as it can be, we will test it at the end of the video and check how logical our predicted story is. Feel free to experiment with different weights, batch size, embedding dimensions, linear transformations, activation functions, losses, optimizers and etc. as I'll be doing this on my end as well. Once I reach a great accuracy with my StorryTeller, we'll go over this model in much more detail and tackle all the functions we didn't have a chance to build together (including validation and pre-processing). In the meanwhile, you can use this code to troubleshoot your own models or build new models based-off the same principles and hopefully, this will give you a good idea about some of the processes involved in AI. #pytorch #python #neuralnetwork #ai #gpu #cuda Starter Notebook and Files: https://drive.google.com/drive/folders/1J9Am97Kyeviofr7wEDSpCiv3bSxZ60Tg?usp=sharing Timestamps: 0:00 - Introduction 0:36 - Train Function 1:19 - Check if GPU is available / Cuda 3:21 - Iterating over epochs 4: 25 - Iterating over the features and target 4:40 - Convert data to tensors 5:43 - Set gradients to zero 5:53 - Step Forward 6:30 - Calculating Loss 6:55 - Backpropagation 7:04 - Update the parameters/weights 7:13 - Update Running Loss 7:41 - Print epochs and running loss while training 10:21 - Fix indentation and typos 11:07 - Set up training parameters 11:35 - Set the loss / criterion 11:59 - Set the optimizer 12:45 - Set the epochs 13:05 - Set the device 13:33 - Call the training function 14:18 - Fixing Errors 14:57 - Training on CPU 15:36 - Save the checkpoint 15:59 - Connect to GPU 16:40 - Training on GPU 17:03 - Load Checkpoint 17:40 - Plot a graph of the loss 17:55 - Prediction 18:36 - Prediction Results 19:53 - Thanks for watching! Connect on LinkedIn: https://www.linkedin.com/in/mariyasha888/ Follow me on Facebook: https://www.facebook.com/MariyaSha888/ Follow me on Instagram: https://www.instagram.com/mariyasha888/ Check out my Github: https://github.com/MariyaSha