Datum Learning
Datum Learning
  • 169
  • 411 809

Відео

RNN/LSTM with Attention | Why Transformer is better than LSTM and RNN.
Переглядів 199 годин тому
In this video, we will take up a question asked by one of the subscribers. The question has four parts, basically, what is attention with LSTM and what is the use of that. We also delve into the aspect of Transformers and how is it better than LSTMs.
Language Detection using Mediapipe | Language Classification using Python and Deep Learning.
Переглядів 1712 годин тому
In this video, we are going to see how to do language classification using Mediapipe. Basically, we are going to give a text to the model and the model is expected to find out which language does the text belong to. There are 110 languages that the model can classify. Code: github.com/DatumLearning/Mediapipe-playlist/blob/main/Language_detector.ipynb
Text Embedder using Mediapipe | Find cosine similarity between two text samples in Mediapipe.
Переглядів 714 годин тому
In this video, we are going to cover the text embedder in Mediapipe. This model can also be used to convert a sentence of any length to a fixed vector of 512 length. This task can help take two sentences and then find the cosine similarity between the two sentences. The sentences focusing on similar ideas have a higher degree of similarity. Code: github.com/DatumLearning/Mediapipe-playlist/blob...
Text Classification using Mediapipe | Sentiment analysis using Mediapipe
Переглядів 614 годин тому
In this video, we are going to cover the text classification using Mediapipe. The model is based on BERT and can do sentiment analysis. Code: github.com/DatumLearning/Mediapipe-playlist/blob/main/text_classification.ipynb
Face Stylization using Mediapipe | Color ink, color sketch and oil painting effect in Mediapipe
Переглядів 1116 годин тому
In this video, we are going to see how can we stylize a face in Mediapipe. This is an experimental feature but a powerful one. It can be used for creating avatars. Code: github.com/DatumLearning/Mediapipe-playlist/blob/main/face_stylization.ipynb
Pose Landmark Detection using Mediapipe | Find the important body landmarks using Mediapipe.
Переглядів 2416 годин тому
In this video, we are going to see how can we find the pose landmarks of an individual in an image using the Pose Landmark Detection functionality in Mediapipe. We will also see how can we plot the points on an image. Code: github.com/DatumLearning/Mediapipe-playlist/blob/main/Pose_Landmarks.ipynb
nn.MaxPool2d | PyTorch function fully discussed | kernel_size , ceil_mode , return_indices, dilation
Переглядів 1619 годин тому
In this video, we are going to see the next function in PyTorch which is the MaxPool2d function. We will also be looking into its parameters. The Maxpool2d operation is used in CNNs to reduce the height and width of the feature maps. Code: github.com/DatumLearning/PyTorch_functions/blob/main/17_maxpool2d.ipynb
nn.Conv2d | Part - 3 fully discussed | Groups, bias and formula for convolution
Переглядів 1119 годин тому
In this video, we are going to see the last remaining parameters of the Convolution operation in PyTorch. We will be focusing on groups, bias the formula for finding the shape of the output of the convolution operation. Groups will be discussed in convolution and we will also see depthwise convolution. We will also see what bias is in the convolution operation. Lastly, we will also see the form...
nn.Conv2d | Part - 2 fully discussed | padding, padding_modes and dilation.
Переглядів 1921 годину тому
In this video, we are going to see the some more parameters of the nn.Conv2d function in the torch.nn module. We will looking into the padding, padding_mode and the dilation parameters of torch.nn. We will go in detail into understanding the padding parameter of nn.conv2d. We will understand what values it can take and what do they mean. Secondly, we will understand the different padding_modes ...
Face Landmark Detection using Mediapipe | Facial expression and landmarks in face images
Переглядів 3421 годину тому
In this video, we are going to see how can we implement face landmarks detection and facial expression classification in Mediapipe. We will find the correct facial expression and also the coordinates of the landmarks in the face of a person. Code: github.com/DatumLearning/Mediapipe-playlist/blob/main/face_landmarks.ipynb
Face Detection using Mediapipe | Keypoints detection on face and Bounding Box
Переглядів 37День тому
In this video, we are going to see how to detect faces in images using the Face detection module of the Mediapipe. We will also see how to get the keypoints in an image and also to see how to plot the bounding box. Code: github.com/DatumLearning/Mediapipe-playlist/blob/main/face_detection.ipynb
Hand Landmark Detection in Mediapipe.
Переглядів 18День тому
In this video, we are going to see how can we find the hand landmarks in the Hand Landmarks Detection task in Mediapipe. We will also see how to interpret the result of the detections. We will see how to isolate a particular point in the image and plot it. Code: github.com/DatumLearning/Mediapipe-playlist/blob/main/hand_landmark_detection.ipynb
Image Gesture recognition using Mediapipe
Переглядів 31День тому
In this video, we are going to find the gesture of the hand as shown in the image. We will also see how to find the co-ordinates of the different points on the hand. Code: github.com/DatumLearning/Mediapipe-playlist/blob/main/gesture_recognition.ipynb
Image Embedding using Mediapipe | How to find Image Similarity in Python.
Переглядів 53День тому
In this video, we are going to see how can we find the image embedding of an image using Mediapipe. This will be useful to find the similar images from an image database. We will be using the cosine similarity to find the similar images. Code: github.com/DatumLearning/Mediapipe-playlist/blob/main/gesture_recognition.ipynb
#AskDL Will Mediapipe and OpenCV be relevant? | Ask your questions on the daily #AskDL post hashtag
Переглядів 2414 днів тому
#AskDL Will Mediapipe and OpenCV be relevant? | Ask your questions on the daily #AskDL post hashtag
nn.Conv2d | Part - 1 fully discussed | in & out channels, stride, kernel_size | PyTorch functions
Переглядів 3314 днів тому
nn.Conv2d | Part - 1 fully discussed | in & out channels, stride, kernel_size | PyTorch functions
Ep - 1 | Intro and Data | Audio Classification in PyTorch Project | Urban Sound classification CNN
Переглядів 3114 днів тому
Ep - 1 | Intro and Data | Audio Classification in PyTorch Project | Urban Sound classification CNN
Instance image segmentation using Mediapipe | Segment one object in an image
Переглядів 3014 днів тому
Instance image segmentation using Mediapipe | Segment one object in an image
Discussion on ChatBots in AI | Simple to Advanced Chatbots.
Переглядів 2114 днів тому
Discussion on ChatBots in AI | Simple to Advanced Chatbots.
Coding and AI in College | Useful tips for someone starting with AI and Coding in College.
Переглядів 1514 днів тому
Coding and AI in College | Useful tips for someone starting with AI and Coding in College.
Yolov10 object on a video using CLI.
Переглядів 2914 днів тому
Yolov10 object on a video using CLI.
Ep - 7 | Results and Test function| Bird Classification project in Deep Learning using PyTorch
Переглядів 3914 днів тому
Ep - 7 | Results and Test function| Bird Classification project in Deep Learning using PyTorch
YOLOv10 Object Detection in Python | Object detection using YOLO in Python
Переглядів 10914 днів тому
YOLOv10 Object Detection in Python | Object detection using YOLO in Python
Ep - 6 | Writing the validation fn | Bird Classification project in Deep Learning using PyTorch
Переглядів 2114 днів тому
Ep - 6 | Writing the validation fn | Bird Classification project in Deep Learning using PyTorch
Ep - 5 | Writing the train function | Bird Classification project in Deep Learning using PyTorch
Переглядів 1714 днів тому
Ep - 5 | Writing the train function | Bird Classification project in Deep Learning using PyTorch
Ep - 4 | Writing the model | Bird Classification project in Deep Learning using PyTorch
Переглядів 2914 днів тому
Ep - 4 | Writing the model | Bird Classification project in Deep Learning using PyTorch
Ep - 3 | transforms and DataLoader | Bird Classification project in Deep Learning using PyTorch
Переглядів 2714 днів тому
Ep - 3 | transforms and DataLoader | Bird Classification project in Deep Learning using PyTorch
#AskDL Can we use pretrained word embedding in PyTorch | Discussion on nn.Embedding in PyTorch
Переглядів 514 днів тому
#AskDL Can we use pretrained word embedding in PyTorch | Discussion on nn.Embedding in PyTorch
Ep - 2 | Setting up - libraries & paths | Bird Classification project in Deep Learning using PyTorch
Переглядів 3314 днів тому
Ep - 2 | Setting up - libraries & paths | Bird Classification project in Deep Learning using PyTorch

КОМЕНТАРІ

  • @rishidixit7939
    @rishidixit7939 2 дні тому

    Thanks a Lot. Please start Transformers for NLP and Vision if possible.

  • @emmanuellouisegonzales5393
    @emmanuellouisegonzales5393 4 дні тому

    NeutrAl comment

  • @emmanuellouisegonzales5393
    @emmanuellouisegonzales5393 4 дні тому

    Negative comment

  • @emmanuellouisegonzales5393
    @emmanuellouisegonzales5393 4 дні тому

    Positive comment

  • @subhankhan6180
    @subhankhan6180 4 дні тому

    I disagree with the ease of implementation, it all depends on your own transfer learning or inference file. There are heaps of examples where you just parse items and use single CLI for both SSD and Faster-RCNN. Ultimate winner should be SSD due to it’s inference and latency plus the ability to be used commercially. Most of yolo algorithms are either AGPL or GPL licensed, which requires heaps of money to be used commercially!

    • @datumlearning6204
      @datumlearning6204 4 дні тому

      Obviously, when I talked about ease of implementation, I meant Python and not terminal/bash. As far as commercial use is concerned, I am not going to comment on that as that was not one of the parameters of comparison. The parameters were speed, accuracy and ease of implementation. Lastly, newer versions of YOLO are the fastest, I have verified that. BTW, thank you for the question.

  • @darshandushing5985
    @darshandushing5985 5 днів тому

    Sir i want to use RCNN for number plate detection, how to do this ? Please reply

  • @ighravwedesmond8679
    @ighravwedesmond8679 6 днів тому

    Great video. Thank you for your contributions to knowledge.

  • @VikramKumar-nf8vw
    @VikramKumar-nf8vw 11 днів тому

    Bhaiya, i request to you please do code also so that we can learn and know more about mediapipe

    • @datumlearning6204
      @datumlearning6204 9 днів тому

      Do you mean line by code typing?

    • @VikramKumar-nf8vw
      @VikramKumar-nf8vw 9 днів тому

      @@datumlearning6204 yes bhaiya, Just write line of code and teach about those line of code i.e. from which library is it belongs to. How to write. What they returned, what are the arguments taken by function. These all are things which should you explain by writing the code

  • @harshans7712
    @harshans7712 11 днів тому

    This is one of the best explanation ever, thanks a lot

  • @syedazainab8689
    @syedazainab8689 12 днів тому

    how to import a torch library

  • @VikramKumar-nf8vw
    @VikramKumar-nf8vw 16 днів тому

    Bhaiya, are open cv and mediapipe relevant to learn? Is there any future of it ?? Plzz reply because I got stuck in my life. Plz show me direction 😢

  • @rishidixit7939
    @rishidixit7939 16 днів тому

    The problem is that there is no code in this. I have watched a lot of UA-cam tutorials and I know that you are very good at explaining your coding which most UA-camrs take for granted. That is why I think making code focused tutorials would be better and along with you can supplement some theory

  • @nil5896
    @nil5896 16 днів тому

    Please provide roadmap for computer vision, there's not particular path for it, like cnn,yolo, diffusion models,etc, etc

    • @datumlearning6204
      @datumlearning6204 16 днів тому

      Hi thank you for your question. I have a video on this already. Please check out the video. ua-cam.com/video/DoH_dD_dy0k/v-deo.html

  • @nil5896
    @nil5896 16 днів тому

    Ur linkedin link?

  • @lekharajfegade2624
    @lekharajfegade2624 17 днів тому

    after rerun its getting 0 with same code

  • @ildarkhalilov9595
    @ildarkhalilov9595 18 днів тому

    why y1 < y3 and y2 < y4?

  • @nil5896
    @nil5896 20 днів тому

    For freshers(internship/job) should we use two column resume or one column

  • @nil5896
    @nil5896 20 днів тому

    For freshers(internship/job) should we use two column resume or one column

    • @datumlearning6204
      @datumlearning6204 20 днів тому

      I always use two. Worked out well for me. Good Luck.

  • @MmAaZzzz
    @MmAaZzzz 20 днів тому

    very well explained

  • @rishidixit7939
    @rishidixit7939 22 дні тому

    I learnt Deep Learning and Machine Learning to apply to Haeltcare Problems. Please do feature Healthcare Related Projects specifically Vision Related Tasks. If possible try to feature some less common projects which require more in depth knowledge of DL and Vision

  • @rishidixit7939
    @rishidixit7939 22 дні тому

    Thanks a LOT Amazing, this is something that is overlooked EVERYWHERE. Just one request that apart from the NLP and Vision Projects please do make projects which use Anutoencoders coz thay are being overlooked. They are right now only used in Segmentation but they also can do Anomaly Detection so please do explore that topic as it is very useful and strangely extremely ignored. One more request would be to please cover Autoencoders a little bit thoretically and write it from scratch .

  • @dhruvJT
    @dhruvJT 24 дні тому

    I want to finetune YOLOv8 model on a custom dataset downloaded from Roboflow but my pc doesn't have necessary resources. How can I achieve this over Kaggle or Colab. Make a small tutorial if possible. Thanks in advance 👍

    • @datumlearning6204
      @datumlearning6204 24 дні тому

      I have uploaded a video on my channel regarding finetuning of YOLOv8 model on custom dataset. You can have a look. Here is the link : ua-cam.com/video/YK-y3U__pLw/v-deo.html

  • @linusbrendel
    @linusbrendel 25 днів тому

    Deep learning is a subset of machine learning. This is why nobody is asking the question. All DL is ML.

  • @rishidixit7939
    @rishidixit7939 27 днів тому

    While appending loss to the all_losses list should we take the mean of losses across all the training samples or the sum of all losses ? Because in CNN and MLP we used to take mean of all the losses but here that is not done.

    • @datumlearning6204
      @datumlearning6204 26 днів тому

      It is always a better option to average the losses. I might have have forgotten it. Thank You.

  • @mogafour
    @mogafour 27 днів тому

    I have learned ML algorithms, ANN, CNN, NLP and YOLO What should I do or learn next ? What a client needs or expects me (as an Ai engineer) to submit , is it the only model weights (model.h5) or something else?