PyTorch Image Segmentation Tutorial with U-NET: everything from scratch baby
Vložit
- čas přidán 13. 06. 2024
- ❤️ Support the channel ❤️
/ @aladdinpersson
Semantic segmentation with U-NET implementation from scratch.
You'll learn about:
✔️How to implement U-Net
✔️Setting up training and everything else :)
Original paper: arxiv.org/abs/1505.04597
My paper review: • U-NET Paper Walkthrough
✅ Dataset used: www.kaggle.com/c/carvana-imag...
Paid Courses I recommend for learning (affiliate links, no extra cost for you):
⭐ Machine Learning Specialization bit.ly/3hjTBBt
⭐ Deep Learning Specialization bit.ly/3YcUkoI
📘 MLOps Specialization bit.ly/3wibaWy
📘 GAN Specialization bit.ly/3FmnZDl
📘 NLP Specialization bit.ly/3GXoQuP
✨ Free Resources that are great:
NLP: web.stanford.edu/class/cs224n/
CV: cs231n.stanford.edu/
Deployment: fullstackdeeplearning.com/
FastAI: www.fast.ai/
💻 My Deep Learning Setup and Recording Setup:
www.amazon.com/shop/aladdinpe...
GitHub Repository:
github.com/aladdinpersson/Mac...
✅ One-Time Donations:
Paypal: bit.ly/3buoRYH
▶️ You Can Connect with me on:
Twitter - / aladdinpersson
LinkedIn - / aladdin-persson-a95384153
Github - github.com/aladdinpersson
⌚️ Timestampo:
0:00 - Introduction
1:03 - Model from scratch
22:20 - Dataset from scratch
29:50 - Training from scratch
39:48 - Utils (almost) from scratch
50:10 - Evaluation and Ending
These from scratch videos & paper implementations take a lot of time for me to do, if you want to see me make more of these types of videos: please crush that *like* button and *subscribe* and I'll do it :)
Support the channel ❤️:
czcams.com/channels/kzW5JSFwvKRjXABI-UTAkQ.htmljoin
Original paper: arxiv.org/abs/1505.04597
Paper review: czcams.com/video/oLvmLJkmXuc/video.html
⌚️ Timestamps:
0:00 - Introduction
1:03 - Model from scratch
22:20 - Dataset from scratch
29:50 - Training from scratch
39:48 - Utils (almost) from scratch
50:10 - Evaluation and Ending
Sure! I will click every like, subscribe and pinned comment thumbs up button! 👍
how we can download this dataset with low resolution as you use in video and learn and train your network
Please do more of these.
Thanks for this Aladdin. I was able to train using my own data. Do you have an idea how I can deploy U-net model to my web app? Can't seem to find any resource on it. cheers
I am training on a satellite Image dataset, My dice score is 0.0 and the pred mask is empty, Am I doing something wrong here ?
I'm writing this comment, because I want more of these types of videos.
I reply to this comment for the same reason 😊
I reply for the same reason
You are the only one who does from scratch this good. Please keep up the good work man!
You are amazing! I have been struggling with this for 2 weeks and your video is so helpful. I can only imagine the amount of work you put into this. Thank you so much.
Thanks a ton!!!!! Learnt a hell lot of new things from this video other than image segmentation.
Your lectures are pure gem!!!!
Thank you for these detailed tutorials, they are very informative
Keep them coming!
Thank you for the nice video! I think this will help a lot of people that are trying to learn how to develop models and also people like me that have experience but need to expand their knowledge in PyTorch.
Great topic! Can't wait to watch it in my spare time.
Thank You a million, I been waiting for this. Yaaay
learnt soo much from this thank you! love the proper structure instead of line by line commands in colab or sth
Hey bro, I know this video is from a long time ago. But thank you for teaching me and, most importantly, being an inspiration. I have now learned how to do the dataset, training loop, and Unet model, all from scratch in my head, just like you. I have also written a thesis on the subject as part of my bachelor's project at my university. Again, thank you, and I hope to learn more from you in the future.
thanks for making this video. It really helped me get started with segmentation tasks
Thanks! Great work. Useful practical information
Simple and clear expression， thank you so much Aladdin Persson
Many thanks of writing this specifically with PyTorch from scratch, I love your videos doing from scratch, you are awesome
سلام این دیتا ستی که استفاده کرده حجم و ابعاد تصویر تصویرش خیلی پایین تره لز دیتا ست اصلی. میدونین از کجا میشه دانلود کرد اینی که تو ویدیو استفاده کرده رو
@@mohammadasadpour9339 Man Nemidunam chera injuri mishe, chand bar inja baratun neveshtam o link gozashtam vali youtube paak mikone, jaryanesh chie!!!!!! So weird 😕
Thanks for creating this education video. Every concept is very clearly explained.
Thank you so much for this informative and detailed tutorial.
Awesome video, stayed all day to make this work because I changed some stuff myself :D
I feel like I want to say I love you for this tutorial
Hi There, This content is gold. I am a huge supporter of writing things from scratch so many thanks for doing it. I do have one suggestion thou. Would you consider implementing also the loss function they used in the UNET paper?
They are using cross-entropy modified with the weighted map so they force the network to segment very thin borders between cells. I think this would also be very useful.
I think this is application-oriented, they use this trick to solve the touching border issue between the cells e.g. when two cells are overlapped.
Thank you so much my guy. I hope one day I can also do this with my own knowledge and understanding
Thank you for the video, great job!
Thank you so much man, keep up the good work
I was listening and following along like a Bob Ross show. Admittedly, I've already implemented a UNet, but the implementation here was much cleaner and nicer. Thanks for making this.
@2K19/EP/050 MANU GAUR
To answer that it can help to explain _why_ we split into training, test, and validation sets.
Think of taking a test in school. You have a workbook with a bunch of problems and a test coming up. Your workbook has the answers in the back. Making a validation set is like taking a bunch of the problems in your workbook and putting them aside for a practice exam. You study all the problems in the workbook except the ones in your practice exam. If you fail the practice exam, maybe you aren't learning the right things from the book. The test is, well, the test.
In the case of this dataset, you could use the test as the validation. That would be fine. You won't know how well you did after all of your work, but if you intend to put it in production that's okay.
In more ML terms: the validation set lets us know if we are overfitting or underfitting on our data before the final test run.
Goldy bro; Keep up the good works bro. A deep love from India
not a single confusion in this video, thanks
Bro, this slaps fr. Thanks!
thanku so much the explanations made it very clear 🙌💯
I literally read UNIX from scratch and I was like oh boy who is this legend 🤣🤣
Thanks for the video idea, maybe next video 😉
Hi. Thank you for your video. It helped me a lot
please make more videos like this. thank you omg
Thank you very much for this video, it is very helpful.
Great work Aladdin,
Thank you for these awesome tutorials
will there be a video about Panoptic segmentation ?
this was awesome! I was looking to implement some of this for my work for some micrscopy images I have taken but I think I need to start a little simpler e.g. I am not familiar with some of the classes and their variables - any ideas where to start?
This is a very well done tutorial
thank you so much,I learnt a lot from this vedio. You are awesome!!!
Thank you for the video man.
Will you do something on U-Net++? Like just a paper walkthrough maybe. I'm trying to find out how many channels they used in their dense skip connection layers but I can't find more details on how exactly they structured them.
Great video, man!
Thanks. Nice and clean
Thank you for the in-depth explanation of how to implement UNET. I would love to see you update GitHub to save the model and a separate display.py showing how to load the model and display the image segmentation predictions.
Hi Aladdin, I can't thank you enough for this video. It was a great learning experience. I would suggest to focus on Computer Vision and keep creating videos like this. I have completely coded the model from scratch with you and it was really enjoyable experience.
Again, Thank you very much !
Just amazing!!
Hey Aladdin! Thanks a ton for the video, it's very clear if you know the basics. However, I'd like to know how I would go and try to segment a new car image, one, which is outside of my dataset.
Awesome work man and your whole channel is solid! Could you add your Pytorch, CUDA and cudNN versions you are using :) I'm having difficulties with pytorch & CUDA compatibilites...
I'm in love with this because, for some reason, although I am not adept yet with deep learning...it answers the crucial part of seeing the architecture being engineered. The only thing I can't get past is how do we create the training datasets? I'm interested in satellite image classification but do you have any idea how to create these training datasets? I've seen people suggesting LabelMe and all but since this is pixel-based classification, what's the anatomy of the input into U-Net?
Hi! great video, congratulations, I have an answer...
when the U.Net needs do multi-class classification and change loss function from BCE With Logits to CrossEntropyLoss, Do I need change to SIgmoid the final conv of the model too?
Very nice and compete tutorial on Unets. I have question, Can we, /or how we use the same code for multiclass segmentations. For example, if there are more than 1 masks in output images, rather than only , "Salt" and "Not Salt"
Fantastic contribution, I just have one question: Any reason why you didn't use pytorch's sliding_window_inference to evaluate validation data?
Very nice video, trying to figure out how to change this for instance segmentation, there are many tutorials for tensorflow but not so many for pytorch
Yey! I am here first :) Excited to go through this
Fantastic video....Thanks
Thank you man !
thank you for this video! after watching a handful of times, I've managed to get it predicting on my own custom dataset, thanks entirely to your instruction.
curious though - any advice on where to start getting a successful model to make a prediction on a single image, and call it by a script?
thank you so much for this content
Thanks for this lovely video
could you please make a video on 3D Unet for medical image(MRI) segmentation
Hey man, love your content, could you make a guide on balancing data in TensorFlow/Keras, similar to the one you did for PyTorch, thanks
Hello,
Thanks for the video it was very helpful. I just have a quick question i'm using another dataset , i was wondering how to get a higher accuracy as currently i'm getting 70% ?
Frigging Awesome!!!
Great video!
Very good video, good explanations
savior of the day
what a great tutorial
Hee, thanks for your video! Got one question: how can your use your trained model for single image segmentation?
Thank you very much.
Thanks for the video
Your videos are very helpful .Could u also implement deeplab v3 from scratch?
Carvana kaggle dataset does not seem to have val_images and val_mask
48:00 man you killed it , wow
hello ! thank you for your video. Can you do a tutorial for multi class sementic segmentation if you have the time ?
Hi, what would be the check_accuracy function in utils if one wants to have more multiclass segmentation? Many thanks!
I'm very thankful for the video and great implementation too but I wish you could go into details of why you do certain things and perhaps explain stuff a bit more.
Would be super helpful !
Hey, nice tutorial, just wanted to ask what tool is best to do the masking of the data...
I have tried few but they return just json file.
Great video man. You are working with RGB images (3 bands or channels). Do you think is possible use this architecture for images with more than 3 channels or bands. I'm thinking in hyperspectral cameras, for example.
I am new to machine learning, I would like to ask:
1) How could I train the model with COCO format dataset
2) How could I train the model with more than 1 label class
3) How to apply the trained model
Hi, I enjoyed your video, even though I already implemented UNet but your intuition is superb. I have one question about how to make inference after training dataset with UNet. I don't know what am doing wrong but when i make prediction, it show black image with little dots and i have tried to understand what am doing wrong but i have got no clue yet.
Dear professor,
I am very interested in your program, and I have two questions now,
(1) How to use code to map between irregular images, complete training through the unet model, and then conduct testing?
Is the mask used for preprocessing data? Is there any special software available for preprocessing?
20:46 I don't understand why you choice resizing x instead of skip_connection which is more similar to the UNET structure it provide. Can you explain it? Thanks.
Great video ,Can you do video about real-time detection with segmentation “ not mask Rcnn” ?
Nice video!!
Thank you bro so much!
Can you please make anoter video on how to do semantic segmentation by training U-net model from scratch?
Thanks for the video. Why you used scaler for backward ? I did not totally understand that.
Bug report: Due to an update to pytorch, the latest version of pytorchversion removes support for variables other than PIL image types from the resize function. So you can use the resize function from torch nn.functional, line 62 could be x = torch.nn.functional.interpolate(x, size=skip_connection.shape[2:])
oh the latest version of pytorchversion do not change the resize function .It's my mistake, my version is the old one. But it is still an alternative solution XD
@AladdinPersson
What kind of PyCharm theme do you use? Looks awesome!
First off:
Aladdin thank you so much for your contributions. I hope your channel continues to grow and grow. You deserve it!
Lastly:
Which version of pytorch are you using? When I run the test function with the randn tensor shape of 161, 161 it raises a TypeError saying the object has to be a PIL Image.
This happens at lines 61,62. - if .shape != .shape: TF.resize()
I appreciate the kind words! I am using PyTorch nightly version (1.8.0.dev) in the video. Are you using 1.7 and it's not working? Have you tried the code on Github too?
Thanks for this Aladdin. I was able to train using my own data. Do you have an idea how I can deploy U-net model to my web app? Can't seem to find any resource on it. cheers
Thank you for video. Was wondering if anyone knows why I would be getting can’t find file errors ?
Hello, I am using your code to do the picture segmentation, I got dice score more than 1 (1.3) do you know what the issue could be? many thanks
i followed your tutorial step by step and used the same dataset and it did an amazing job. The first dataset (CARVANA) I used worked fine, but once I changed it, the results went downhill. I tried it on CASIAv2, but my dice score is always 0.0 and my predicted masks are just black... i don't know how to fix this, if anyone has any ideas, i beg you, do let me know!
I had the same issues
Facing the same issue
Hello! Great tutorial! Was just wondering, at 34:39 you make use of torch.cuda.amp.autocast(). Would this work if you are using CPU processing only, given that you're calling a cuda method or is there a CPU based alternative? I'm trying to experiment with this on my mac, i'm relative new to this and this is one of the steps I don't fully understand yet. Any help would be appreciated :)
Hi! did you figure out if this code would work with cpu?
how difficult would it be to consider masking the glass or window sections in these images?
Hey do you know how to annotate the dataset for Unet? Are there any specific tools for it?
How did you do the masking in the dataset? How did you create the dataset, where can I learn the detailed explanation?
where I should make some ajustments in the codes to make the unet fitting my png imgs?
Please can you make a video on how to use hooks to join a transfer learning architecture like RESNET with UNET for image
segmentation? That's how to use the pre-train model as the encoder part of the UNET architecture.
Hi Aladdin,
Thank you so much for your response in my previous comment.
Currently, I updated pytorch and conda in my GPU laptop. But facing a problem while running the training code.
I got this error message:
from ._nnls import nnls
ImportError: DLL load failed: The specified module could not be found.
Here I want to mention that both my python and libraries are of 64 bit.
The mirroring part that you mentioned in the begining that we would have to do if we used valid convolution... I cant understand that. I saw the paper walkthrough video too still that part is very unclear. Could you please help with that?
so, what changes do i need to make if I want to perform a multi class segmentation here can you help me?
hello :) I just followed your code until making model.
but got error saying
TypeError: img should be PIL Image. Got on TF.resize.
even, I copy your code on your git hub it cause same error, anyone know how to solve this?
I have a problem . In the diagram of the UNET, it shows an Upsampling2D together with Conv2D but you create a ConvTranspose2d. Do I miss something??
you can make transposeconv to a modulelist