In this 2-hour long guided-project course, you will load a pretrained state of the art model CNN and you will train in PyTorch to classify facial expressions. The data that you will use, consists of 48 x 48 pixel grayscale images of faces and there are seven targets (angry, disgust, fear, happy, sad, surprise, neutral). Furthermore, you will apply augmentation for classification task to augment images. Moreover, you are going to create train and evaluator function which will be helpful to write training loop. Lastly, you will use best trained model to classify expression given any input image.
This course is good for practing python scripts by creating a facial recognition AI. The course offers an exercise in python, nothing more.
It is a good approach to create a facial expression regonition and code explanation is very well, i am happy to learn
Showing: 11 of 12
I dislike this course. Things weren't explained at all and the course finished without even testing the model.
1-start because the video for last section is missing. Otherwise the project is OK.
Instructor never complete the last part (inference) of this project. Learner have to find their own solution in order to complete the task. Irresponsible instructor. Not good.
code didn't work. Error in last step: Expected input batch_size (32) to match target batch_size (21). Also, running on cuda didn't work. I had to run on cpu. Please verify code works before offering this class on coursera. Also, can you provide a sample of the final code for the students to download?
It is a good approach to create a facial expression regonition and code explanation is very well, i am happy to learn
Excellent. Learnt a lot. Really
Πολύ καλό μάθημα.
good
This course is good for practing python scripts by creating a facial recognition AI. The course offers an exercise in python, nothing more.
This course very good and great
last video is missing