TensorFlow 2.0 Tutorial for Beginners 12 - Credit Card Fraud Detection using CNN in TensorFlow 2.0

Поділитися
Вставка
  • Опубліковано 11 січ 2025

КОМЕНТАРІ • 40

  • @ijeffking
    @ijeffking 5 років тому +6

    I like your approach to balancing the dataset. And your step-by-step explanation is very useful. Thank you very much.

    • @KGPTalkie
      @KGPTalkie  5 років тому

      Thank you so much for watching ❤️

  • @keensaj
    @keensaj 4 роки тому +1

    thanks! great explanation :)

  • @allieubisse470
    @allieubisse470 4 роки тому +3

    Please explain all the layers on a separate section. Just to give a detailed information and when , how, what task to use each layer.

  • @ghizlanekhattari8804
    @ghizlanekhattari8804 Рік тому

    what i know is we aplly CNN model for just dataset of typr image ?

  • @shwetaredkar734
    @shwetaredkar734 5 років тому

    Excellent tutorial. Nicely explained. How to do 10 fold CV for the same example and test it on independent dataset?

  • @rohitmaurya5660
    @rohitmaurya5660 4 роки тому +1

    ,after training you had model .please show the result of that model on a new dataset .

  • @Mikey33X
    @Mikey33X 5 років тому

    I didn't get the extra 1 at 15:48. I copied the code exactly. I tried combining the last three lines, but it said there is an attributeError with reshape.

  • @akd9977
    @akd9977 5 років тому +2

    Excellent post.. How can I learn tensorflow from the scratch. Do you have any video. Thanks

    • @KGPTalkie
      @KGPTalkie  5 років тому

      Please watch TensorFlow 2.0 Tutorial for Beginners Playlist: ua-cam.com/video/nVvhkVLh60o/v-deo.html

    • @KGPTalkie
      @KGPTalkie  4 роки тому

      Hi,
      This is in regard to the code files of all videos at KGP Talkie UA-cam Channel. I request you to please get enrolled yourself in my Udemy course and show your support and love to KGP Talkie. All the code files and video lectures have lifetime access with 30 Days money back Guarantee. Code and question-answer support are also available at Udemy.
      I promise you to give FREE COUPONS for the next course on Deep Learning and ML. You can click on the link mentioned below and can get yourself enrolled!! bit.ly/udemy95off_kgptalkie
      New content is added at Udemy:
      1. Animation Plot [2 lectures]
      2. Python Coding in Mobile [5 lectures]
      3. Complete EDA of Boston Dataset [20 lectures]
      What else we promise in this course
      1. Kaggle data EDA
      2. Text data EDA
      3. More Animation Plot
      4. More 3D plots
      5. Figure Aesthetics and Decoration
      6. Free coupons for next course
      7. And so much more.
      Hurry up!!! Only for a limited time.
      Please email your details at udemy@kgptalkie.com for the FREE COUPONS of the next course.
      Note: Code files of UA-cam lectures will be also available once you register in this course. Please send an email to udemy@kgptalkie.com after enrolling in this course.

  • @tapaspal8623
    @tapaspal8623 5 років тому +1

    Good work

    • @KGPTalkie
      @KGPTalkie  5 років тому

      Thank you so much Tapas for watching. Please watch other videos as well. Please let me know if you need a video on any other topic.

  • @zariabfatimaabro8094
    @zariabfatimaabro8094 4 роки тому

    hi i wanted to ask if we can use a dataset of a real bank. if you could make a video regarding working on a dataset obtained from real world scenario or self-created dataset

  • @MTCSEAshray
    @MTCSEAshray 4 роки тому

    while putting model.add(Conv1D(32, 2, activation='relu', input_shape=x_train.shape[0])) I m getting TypeError: 'int' object is not iterable? Why is this showing pls help

  • @VarunSharma-ym2ns
    @VarunSharma-ym2ns 2 роки тому

    ValueError: Failed to find data adapter that can handle input: , ...i am facing this error while using model.fit()

  • @mehedihasanjohn1429
    @mehedihasanjohn1429 4 роки тому

    Really it was amazing , i think you did undersampling in this video but i need oversampling also, plz give me the oversampling source code only. Advance tnx

  • @anumariababu3262
    @anumariababu3262 4 роки тому

    can you please share the project paper of this project

  • @tanb13
    @tanb13 5 років тому

    Great Video, thanks a lot. After we have created the model and tested its performance, how can we run it on a dataset and obtain the resultant class (fraud/non-fraud) predicted for each transaction and save in a csv file. If we can get probabilities as well for each class for each transaction that will be great as we can use that as a threshold to filter transactions by the model and take corrective actions accordingly.

    • @KGPTalkie
      @KGPTalkie  5 років тому

      Hi, you can make a pandas data frame and then write back as csv and you can get probability if you call model.predict only. Thanks for watching ❤️.

  • @shivaprasadmb862
    @shivaprasadmb862 4 роки тому +1

    The explanation is good, I have some doubts:
    1. Why we use two variables x and y?
    2. Why they are called feature space variables?

  • @siddhigolatkar8558
    @siddhigolatkar8558 3 роки тому

    Thankyou

  • @nikimehta2697
    @nikimehta2697 4 роки тому +1

    Hyy will u explain me about columns v1 to v28 ...I don't get it

    • @KGPTalkie
      @KGPTalkie  4 роки тому

      Thanks for watching. These columns are PCA components of original dataset. PCA is done to avoid the exposure of real dataset and make it anonymous.

    • @nikimehta2697
      @nikimehta2697 4 роки тому

      @@KGPTalkie so what is the use of v1 to v28

    • @nikimehta2697
      @nikimehta2697 4 роки тому

      Can I talk with you regarding this if u share ur contact no

  • @AlokKumar-pc7pv
    @AlokKumar-pc7pv 4 роки тому +1

    By balancing the data, you have lost many data, you should use some other technique so that we done lose so many data just to balance.

  • @ishitabiswas1849
    @ishitabiswas1849 3 роки тому

    Sir after we get balanced dataset why we need to merge these 2 dataset together ? If we are not going to merge that what will happen ?

    • @KGPTalkie
      @KGPTalkie  3 роки тому

      Merging is must otherwise you will have only one type of data.

  • @ijeffking
    @ijeffking 5 років тому

    While running plot_learningCurve(history, epochs), I get the KeyError :"accuracy" with the arrow pointing towards ----> 4 plt.plot(epoch_range, history.history['accuracy']). Could you kindly suggest a solution?

    • @KGPTalkie
      @KGPTalkie  5 років тому

      Sometime TF uses different key. Print(history) then you can see keys being used.

    • @ijeffking
      @ijeffking 5 років тому

      @@KGPTalkie thank you

  • @jeetshah8513
    @jeetshah8513 4 роки тому

    What about "false-true", and "true-false"
    why didn't you check that?