Thursday, March 5, 2026
Latest:
Videos

Convolutional Neural Networks – Deep Learning basics with Python, TensorFlow and Keras p.3

Artis Modus



sentdex

Welcome to a tutorial where we’ll be discussing Convolutional Neural Networks (Convnets and CNNs), using one to classify dogs and cats with the dataset we built in the previous tutorial.

Text tutorials and sample code: https://pythonprogramming.net/convolutional-neural-network-deep-learning-python-tensorflow-keras/

Discord: https://discord.gg/sentdex
Support the content: https://pythonprogramming.net/support-donate/
Twitter: https://twitter.com/sentdex
Facebook: https://www.facebook.com/pythonprogramming.net/
Twitch: https://www.twitch.tv/sentdex
G+: https://plus.google.com/+sentdex

Source

Similar Posts

45 thoughts on “Convolutional Neural Networks – Deep Learning basics with Python, TensorFlow and Keras p.3”

  2. I don't understand my error and why you don't have it. I have got this one : "unsupported operand type(s) for /: 'list' and 'float'" be cause of "X = X/255.0" but I don't found a solution if someone got one. But your tutorial is very good and helpful. Thank you

  3. ………model.fit(X, y, batch_size=32, epochs=3, validation_split=0.3)
    after execution
    AppDataLocalTemp/ipykernel_2228/1932123811.py in <module>

    43 )

    44

    —> 45 model.fit(X, y,

    46 batch_size=32,

    47 epochs=3,

    ~anaconda3libsite-packageskerasutilstraceback_utils.py in error_handler(*args, **kwargs)

    65 except Exception as e: # pylint: disable=broad-except

    66 filtered_tb = _process_traceback_frames(e.__traceback__)

    —> 67 raise e.with_traceback(filtered_tb) from None

    68 finally:

    69 del filtered_tb

    ~anaconda3libsite-packageskerasenginedata_adapter.py in train_validation_split(arrays, validation_split)

    1478 unsplitable = [type(t) for t in flat_arrays if not _can_split(t)]

    1479 if unsplitable:

    -> 1480 raise ValueError(

    1481 "`validation_split` is only supported for Tensors or NumPy "

    1482 "arrays, found following types in the input: {}".format(unsplitable))

    ValueError: `validation_split` is only supported for Tensors or NumPy arrays, found following types in the input: [<class 'int'>, <class 'int'>, <class 'int'>, <class 'int'>, <class 'int'>, <class 'int'>, <class 'int'>, <class 'int'>, <class 'int'>, <class 'int'>, <class 'int'>, <class 'int'>, <c……etc

    WHAT CAN I DO SIR PLS HELP ME

  4. Can you please explain the science of wisely choosing the critical/essential arguments in various layers, i.e., how to select the number of filters, kernel_size etc., in a CON2D layer? Similarly, whats the science behind adding layers upon layers? I would be grateful if something could be added other than the usual answer of 'more layers in deep learning help in extracting hidden features/patterns'. Please share a link to your video if you have already done it.
    To the broader audience, please share a video that you may have found very informative about choosing the parameters for various layers? And how to transition from layer to layer?

  6. Any advice on how to pass the computations from cpu to gpu ( model training) ? I can't find the right combination of tensorflow-gpu, cudnn and cudatoolkit. I tried many things but didn't work and I need some basic guidance , a right mind approach.

  10. what to do when i get this error

    Input 0 of layer "conv2d" is incompatible with the layer: expected min_ndim=4, found ndim=2. Full shape received: (None, 1)

    while compiling the model when i use Conv2D the first time

  12. Good day! I just have some inquiries regarding about the error I've got , it says " ValueError: `logits` and `labels` must have the same shape, received ((None, 1) vs (None, 50, 50, 1))", though I have followed all the code and still got this error, thanks.

  13. Hello, I encounter a problem with this code. When I execute the code, I got the error "could not load dynamic library 'cublas64_11.dll'; dlerror: cublas64_11.dll not found." Therefore, I went to check whether my CUDA files correspond to my Tensorflow. However, I found that the version of CUDA was 11.6, which is not compatible to any current version of Tensorflow now. What should I do to solve this problem? Thank you.

  16. Are these results normal? it took 4min (approx.) for each epoch i have intel i7-10th gen cpu
    Epoch 1/3
    546/546 [==============================] – 229s 418ms/step – loss: 0.6330 – accuracy: 0.6417 – val_loss: 0.5930 – val_accuracy: 0.6887
    Epoch 2/3
    546/546 [==============================] – 239s 437ms/step – loss: 0.5550 – accuracy: 0.7162 – val_loss: 0.5284 – val_accuracy: 0.7382
    Epoch 3/3
    546/546 [==============================] – 243s 444ms/step – loss: 0.5080 – accuracy: 0.7515 – val_loss: 0.5797 – val_accuracy: 0.7050
    ——————————————————–

  21. import tensorflow as tf

    ###Doesnt work for some reason…..object differences due to update?->from tensorflow.python.keras.datasets import cifar10

    ###Doesnt work for some reason…..object differences due to update?->from tensorflow.python.keras.preprocessing.image import ImageDataGenerator

    from tensorflow.python.keras.models import Sequential

    from tensorflow.python.keras.layers import Dense, Dropout, Activation, Flatten

    from tensorflow.python.keras.layers import Conv2D, MaxPooling2D

  23. Once u train Ur code for epoch = 5 and i décidé to train it for epochs=10 then will you like restart everything like schufling again thé images and everything then start a brand New epochs=10 or will you juste continue with same order of images and the first epochs 5 u will liké just add 10 now its liké u did 15 epoches

  25. I'm a little lost here. How does the engine know how to label an image as a dog or a cat. How does it know to get the labels the right way around? I can't see where you tell the engine. "hey this image is a dog"

  27. 13:28 actually I tried it and categorical_crossentropy will not work. the accuracy will only be around .5 during 3 epochs. Only after using binary_crossentropy the accuracy slowly increased. I don't know why

    Edit: I guess binary_crossentropy would be more fitting anyways as we only have two classes (cats 'n dogs), but I don't understand why categorical_crossentropy won't work

  28. I'm doing image clasification for traffic signals which has more than 50000 images, but i keep recieving not enough memory to allocate, 115GIB error, can anyone help?

  31. I have two problems in tensorflow 2:
    I fixed the following two problems:
    1. y = np.array(y) # this convert the array y into nnumpy arry for model.fit() error.
    2. My Video RAM in GPU can be not enough. I turn off GPU mode in Tensorflow 2.
    import os
    os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
    then, it runs.

  34. Hellow, I couldn't import Maxpooling2D from 'tensorflow.keras.layers' as shown in an error. Could you please tell what to do. Thanks a bunch.

  35. File "D:PhD ThesisPython2024Test0015_CNN.py", line 6, in <module>

    X=pickle.load(open("X.pickle","rb"))

    PermissionError: [Errno 13] Permission denied: 'X.pickle' When I run the program it give this error please what I have to do to correct it?

  38. i am getting an error AttributeError: module 'tensorflow.python.distribute.input_lib' has no attribute 'DistributedDatasetInterface'. Did you mean: 'DistributedDatasetSpec'?

    in model.fit(X, y, batch_size=32, epochs=3, validation_split=0.3)

Comments are closed.

WP2Social Auto Publish Powered By : XYZScripts.com