In this 2-hour project-based course, you will be able to : - Understand the Segmentation Dataset and you will write a custom dataset class for Image-mask dataset. Additionally, you will apply segmentation augmentation to augment images as well as its masks. For image-mask augmentation you will use albumentation library. You will plot the image-Mask pair. - Load a pretrained state of the art convolutional neural network for segmentation problem(for e.g, Unet) using segmentation model pytorch library. - Create train function and evaluator function which will helpful to write training loop. Moreover, you will use training loop to train the model.
Very nicely taught. Code must be made available in text file.
I am extremely grateful for the chance to work on this Course project as it will undoubtedly help me advance in my career. Thank you for entrusting me with this opportunity!
Great explanation and hands-on experience for image segmentation model. I took more than an hour (because I like to look up things), and it was worth the time.
There are some typos and some things that are not shown in the video but if you worked with python before you should be able to solve them easily. Overall a great exercise.
Very good teaching style and well explained project
Showing: 20 of 30
The inference part of the code doesn't work. The instructor doesn't respond to queries in the discussion forum either. After spending the whole time, I realised that it didn't work. Disappointed. "AttributeError: module 'helper' has no attribute 'show_image' " (I tried installing the modules, but it is still not working).
You spend 50% of the time just typing into Colab session what the instructor is typing in the videos. Not much to learn with this project beyond some parts of API of libraries like PyTorch and Albumentation. Then you spend 50% time debugging the code. There are a lot of bugs in the code.
Very simple introduction to Pytorch and Image segmentation.
Terrible course, you basically just do copy/paste and execute a notebook without much explanation or context. Very little explanation is given as to "why" we do certain things of choose specific parameters. The notebook also has several typos and bugs. Overall I actually did not have to think to pass the course, which render the whole thing a waste of my time. Would not recommend.
Disgusting explanation, i spent so much time just to import dataset correctly into the colab but nothing happened, he skipped so many steps in between. Really dissatisfied with this methodology.
Great instructor and very practical hands-on approach. I would prefer more explanation on other encoder and weight presets as that will be important for transferring the knowledge learned here!
I am extremely grateful for the chance to work on this Course project as it will undoubtedly help me advance in my career. Thank you for entrusting me with this opportunity!
Great explanation and hands-on experience for image segmentation model. I took more than an hour (because I like to look up things), and it was worth the time.
This was a really great project! It helps me understand the necessary steps to create an image segmentation project.
Very good teaching style and well explained project
This course was well guided
Πολύ καλό μάθημα.
Excellent course
It is very good
Excellent
thank
good
Pretty good overall. Some work needed with this projects to make them easier to debug, such as making source code available with a diff tool so people can more easily pick out stupid types.
Because this is python, not cpp or java or similar, it can actually be pretty hard to catch a lot of typos, due to lack of compile time checks, scope control, and abundance of global variables.
There are some typos and some things that are not shown in the video but if you worked with python before you should be able to solve them easily. Overall a great exercise.
Very nicely taught. Code must be made available in text file.