I am an expert in Applied Machine Learning, Image Processing and Computer Vision Domains. This channel is aimed to lower the entry barrier to machine learning and artificial intelligence, to make the concepts and tools more available and the concepts more palatable from early enthusiasts to researchers. We discuss about research papers, trends, AI/ML concepts, machines, computers, internships, next career moves and lots more. Glad you could join me on my journey on AI!
Thanks for broadcasting your knowledge mam... Can you suggest tensorflow package which is suitable for NVIDIA-SMI 551.76, Driver Version: 551.76, CUDA Version: 12.4
Hello there. The deciding factor is the cuda version. So go to the nvidia driver page and see which tensorflow version is most compatible. The latest version is tensorflow 2.17 so that it lower. Good luck
@@AIwithSohini Hi Sohini Mam, Can you suggest the video making tools like screen recording software, you are using for knowledge share. Curious to know.....
I'm working on a project in school where I will have to use XAI to get login analytics from users in terms of successful and unsuccessful login attempts using user passwords, usual login times or device location as conditions. Please how do I go about this?
@@AIwithSohini Honestly, I need to know the XAI technique best for this project. Reason for access granted or denied on the authentication system and relayed to the system administrator via a management dashboard for analytics.
Great question. Not directly. If you can use neural networks to extract a linear feature layer then that layer could serve as features, but still it will be difficult to explain each pixel location. Tensorflow has better image explaining capabilities. Hope that helps. Good luck
The major difference is AI has the capability to make decisions in previously unseen situations. Automation robotics does the same manouvres and tasks again and again. Like the automated coffee maker machine in airports are automation robots but an autonomous driving system is AI. Hope this helps. Thanks and stay tuned
Hello I am phase ing probelm in labelimg like I want to annotate the images like polygons tools but in my labelimg only show the rectangle not show any others tool like points polygon's please give me solution
Hello sohini can I get a tutorial which is step by step on building a resuming screening using machine learning and NLP to quickly and accurately filter and rank resume based on job descriptions
Nice work Sohini, if possible I wanted to get in touch with you and discuss the roles and responsibilities and to get more clarity on what to do for an upcoming project in my company
Hello, I appreciate your work. I like the way you deliver the talk. My name is Pravin Harane. I am in advance stage of phd in mechanical engineering (electric discharge drilling of Waspaloy). After PhD in mechanical, I want to work in AI related domain in mechanical or interdisciplinary field. But, unfortunately i not getting exact path. Can you just give me brief guidelines for me to get into AI domain of mechanical related field. Thank you
Hello I did not watch the video yet But i want to do 2 class segmentation with u-net And i have the data set ready, what i want to segment is optic cup and optic disc , its the almost the same dataset Will this video help me? Thanks alot
Hello there. This video and this one ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-M4EvfyoVhb4.htmlsi=_t5hxjMAf_m9iIX0 will be useful for you to walk through the code. Good luck and stay tuned
Hello and thanks for the great question. Both solutions will stay on since they are more need based. The LLM would most likely be in the cloud owing to the size and training restrictions but the knowledge can be on local systems. Cloud only will be faster, hybrids will be slower. I will make a video, the next part on this topic. Thanks again for the question and stay tuned!
I’m having issues, after I made my password and clicked all done, my computer dint take me to the Ubuntu it just stayed on that screen that you have, like no applications or anything. The only thing I see is the top right corner battery, WiFi that’s all. Any tips on what this could be ?
Thanks a lot for your message. V7 has a whole academy and videos. I highly encourage you to sign up with a free academic license. Their auto chat bots constantly keep asking if you need any help. They will guide you to the needed videos or even have sessions with you if needed. So just sign up and experience it yourself. Hope this helps
very fantastic explanation and code, so from where i get the images and its GT, i didnot found in the link shown in the github repo. please provide the data which you used. Thank you so much
Follow some of my tutorials in my channel. I provide code and debugging. The more you try the more you learn. Hope that helps. Thanks for joining the community and stay tuned.
Hello there and thanks for connecting. The research is over 4 years ago and the libraries have changed a lot. So there has to be upgrades to the libraries. Sorry we don’t have version control for this anymore. One way to resolve will be to debug each cell and taking the modifications that colab suggests. Hope that helps. Thanks and regards. Sohini
Hello there, it very much depends on the kind of work you will have. While Dell and HP are considered standard, if you need to run deep learning models then MSI ones are good too. Hope that helps. Good luck with your work from home.
Hello there thanks for the question. What you are looking for is called intelligent character recognition. Here is my video on using V7 for such tasks. ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-6QUc1tNUlBo.htmlsi=OOeAR6ck9-7iX5Eh. Also ChatGPT can do this task very well as well. My latest video should explain that. Thanks and stay tuned
Thanks for valuable discussion, especially validation. If you discuss the validation with examples it will be great...waiting for the next Prompt Engineering discussion...Thanks
Glad you liked it. Please check out our paper arxiv.org/abs/2311.07592 and also the blog in the description. We discuss samples on hallucinations there.
SaveResultwImage('./test',PATH+'pred/',results) results are not storing in pred folder getting following error ValueError: ImageIO does not generally support reading folders. Limited support may be available via specific plugins. Specify the plugin explicitly using the `plugin` kwarg, e.g. `plugin='DICOM'`
Hello and thanks for the question. The code generates 3 dimensional images RGB planes if you read with PIL library. For more than 3 classes you will need to convert to .npy files where each plane dimension is each class. So no color map needed here. Hope this helps. Thanks for watching
Superr, can you discuss on resources for Gen AI like multimodal ? that will be great...I want to explore on multimodal, I believe that will much value in future for data scientists.
Thanks for the response. You can check out the V7 video that allows for automated object detection models to auto detect ROIs ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-TAYuwd6N18s.html. The only way to automate labelling once you have annotated a few images is to train/fine tune an object detection model (like YOLO) and generate automated estimations that can then be corrected by the humans. V7 is a great solution and it is free for academic use. Good luck and stay tuned!
Hello @@RezaAlamsyah-he5ku, Thanks for the question. I know V7 was free to academic and research purposes, but I am sure there is limitation on data usage. Once you sign up, you will be prompted how to extend your usage base. Hope this helps and good luck!
Helloo ma'am, just wanted to ask the dataset you imported from kaggle was already sorted and has categorized data. but if i have 100 resume in PDF formate how can i extract data in such a way that it goes perfectly to the exact column in cvs or excel, knowing that resume do have different formates and creating a csv manually like filling all columns by hand isnt an option. kindly explain me thankyou.
Thanks for the great question! If your resumes exist in pdf format the first step is to extract the text from pdf (so OCR models, or using pdf2text modules in python pypi.org/project/pdftotext/ Pytesseract is another such module that will extract the text from pdf. Once you have the text, then you can proceed with the method in the video. Hope this helps. Thanks and stay tuned!
@@AIwithSohini Is there a way I can get in touch? I am trying to create a scholarshipp predictor but having some challenges. The goal is for the user to enter his/her field of interest or the model scan a CV and predict the likely scholarship the user can apply to
Thanks a lot for your question. V7 is a great platform and free to academic and researchers that allows for multi-class annotation. Check out my video here ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-TAYuwd6N18s.html. Thanks and stay tuned!
goodmorning i need medical images along with descriptions can you please help me in this regard.., just tell me website from which we can download freely for research perpose thank you
Hello and thanks for the query. Typically the best source of data is released at major conferences like MICCAI. Those datasets are typically very large. My suggestion would be to try Kagglle.com and search for "kaggle <medical image you need>" this can be stomach CT, head CT etc. You should get something from Kaggle. If you are unable to find what you need, please let me know here exactly what data set you are looking for and I can help with some links. Thanks and stay tuned! ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-TAYuwd6N18s.html
@@AIwithSohini thanksalot for reply madam..., madam actually I searched in kaggle for data set but it doesn't have descriptions only images are there in that data-set.., can you help me in this regards
hi, thank you so much for this video. I wanted to ask if there is a way to use Mask RCNN pretrained model with .nii files (MRI scans of heart). The dataset itself is already annotated but both the images and the annotations are in .nii format.
yes that is very much possible! you will need to rely on the python library for nib. Here is a tutorial idiotdeveloper.com/read-3d-nifti-images-in-python3/. Once you read the nii files and convert it to jpg/png that will enable you to reuse MaskRCNN. Hope this helps. Thanks and stay tuned!
