DevHeads IoT Integration Server•3mo ago

Normalizing Input Data for CNN Model in Image Recognition System on ESP32

still based on my project image recognition system that can analyze images of tissue samples, identify malignancies, and predict possible symptoms and causes. How do i train a CNN to accurately identify malignant tissues? My aim is to train a convolutional neural network (CNN) model for image recognition. But I keep encountering the error

    ValueError: Input data not properly normalized

    ValueError: Input data not properly normalized

Here's my code

    import tensorflow as tf
    from tensorflow.keras import layers, models
    
    model = models.Sequential([
        layers.Conv2D(32, (3, 3), activation='relu', input_shape=(224, 224, 1)),
        layers.MaxPooling2D((2, 2)),
        layers.Conv2D(64, (3, 3), activation='relu'),
        layers.MaxPooling2D((2, 2)),
        layers.Flatten(),
        layers.Dense(64, activation='relu'),
        layers.Dense(1, activation='sigmoid')
    ])
    
    model.compile(optimizer='adam',
                  loss='binary_crossentropy',
                  metrics=['accuracy'])
    
    history = model.fit(train_images, train_labels, epochs=10, validation_data=(test_images, test_labels))

    import tensorflow as tf
    from tensorflow.keras import layers, models
    
    model = models.Sequential([
        layers.Conv2D(32, (3, 3), activation='relu', input_shape=(224, 224, 1)),
        layers.MaxPooling2D((2, 2)),
        layers.Conv2D(64, (3, 3), activation='relu'),
        layers.MaxPooling2D((2, 2)),
        layers.Flatten(),
        layers.Dense(64, activation='relu'),
        layers.Dense(1, activation='sigmoid')
    ])
    
    model.compile(optimizer='adam',
                  loss='binary_crossentropy',
                  metrics=['accuracy'])
    
    history = model.fit(train_images, train_labels, epochs=10, validation_data=(test_images, test_labels))

Solution:

your code should look like ```python import tensorflow as tf from tensorflow.keras import layers, models ...

Jump to solution

3 Replies

Enthernet Code•3mo ago

It looks like you're encountering a ValueError: Input data not properly normalized while working on your CNN for image recognition. This error implies that your input data needs to be preprocessed before feeding it into your model. - CNNs perform better when the input data is normalized. For images with pixel values ranging from 0 to 255, scaling them to [0, 1] is standard practice. Normalize your images like this:

   train_images = train_images / 255.0
   test_images = test_images / 255.0

   train_images = train_images / 255.0
   test_images = test_images / 255.0

- Check that the dimensions of your images match the expected input shape of your model. If your first convolutional layer expects grayscale images of shape (224, 224, 1), but your images are RGB, adjust the input_shape accordingly:

   model = models.Sequential([
       layers.Conv2D(32, (3, 3), activation='relu', input_shape=(224, 224, 3)),
       layers.MaxPooling2D((2, 2)),
       layers.Conv2D(64, (3, 3), activation='relu'),
       layers.MaxPooling2D((2, 2)),
       layers.Flatten(),
       layers.Dense(64, activation='relu'),
       layers.Dense(1, activation='sigmoid')
   ])

   model = models.Sequential([
       layers.Conv2D(32, (3, 3), activation='relu', input_shape=(224, 224, 3)),
       layers.MaxPooling2D((2, 2)),
       layers.Conv2D(64, (3, 3), activation='relu'),
       layers.MaxPooling2D((2, 2)),
       layers.Flatten(),
       layers.Dense(64, activation='relu'),
       layers.Dense(1, activation='sigmoid')
   ])

- Ensure your data (train_images and test_images) is in a compatible format for TensorFlow, such as numpy arrays or TensorFlow tensors. Incompatible data types can cause issues.

Solution

Enthernet Code•3mo ago

your code should look like

import tensorflow as tf
from tensorflow.keras import layers, models

# Normalize the images
train_images = train_images / 255.0
test_images = test_images / 255.0

# Define the model
model = models.Sequential([
    layers.Conv2D(32, (3, 3), activation='relu', input_shape=(224, 224, 1)),  # Adjust input_shape if using RGB
    layers.MaxPooling2D((2, 2)),
    layers.Conv2D(64, (3, 3), activation='relu'),
    layers.MaxPooling2D((2, 2)),
    layers.Flatten(),
    layers.Dense(64, activation='relu'),
    layers.Dense(1, activation='sigmoid')
])

model.compile(optimizer='adam',
              loss='binary_crossentropy',
              metrics=['accuracy'])

# Train the model
history = model.fit(train_images, train_labels, epochs=10, validation_data=(test_images, test_labels))

import tensorflow as tf
from tensorflow.keras import layers, models

# Normalize the images
train_images = train_images / 255.0
test_images = test_images / 255.0

# Define the model
model = models.Sequential([
    layers.Conv2D(32, (3, 3), activation='relu', input_shape=(224, 224, 1)),  # Adjust input_shape if using RGB
    layers.MaxPooling2D((2, 2)),
    layers.Conv2D(64, (3, 3), activation='relu'),
    layers.MaxPooling2D((2, 2)),
    layers.Flatten(),
    layers.Dense(64, activation='relu'),
    layers.Dense(1, activation='sigmoid')
])

model.compile(optimizer='adam',
              loss='binary_crossentropy',
              metrics=['accuracy'])

# Train the model
history = model.fit(train_images, train_labels, epochs=10, validation_data=(test_images, test_labels))

Boss lady•3mo ago

Thanks it did work out

Gaming

Programming

Normalizing Input Data for CNN Model in Image Recognition System on ESP32