[Computerlab]

https://app.hackthebox.com/challenges/Fuel%2520Crisis
In this challenge a h5 model is supposed to be created and uploaded, 
The model is a digit recognition model mnist{being more specific}


There is a blaclisted spaceship number in util.py SRC Code 
self.bannedId = "32166"


A method to bypass the check,
Training the model to recognize '2' as any other digit.

---

MODEL Creation WORKS ONLY IN WINDOWS

import numpy as np
import tensorflow as tf
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Conv2D, MaxPooling2D, Dropout, Flatten, Dense
from tensorflow.keras.datasets import mnist
from tensorflow.keras.utils import to_categorical

# Load and preprocess the MNIST dataset
(x_train, y_train), (x_test, y_test) = mnist.load_data()

# Reshape data to fit the model
x_train = x_train.reshape(x_train.shape[0], 28, 28, 1)
x_test = x_test.reshape(x_test.shape[0], 28, 28, 1)

# Normalize pixel values
x_train = x_train.astype('float32') / 255.0
x_test = x_test.astype('float32') / 255.0

# Change any occurrence of the digit '2' to '9' in both y_train and y_test
for i, item in enumerate(y_train):
    if item == 2:
        y_train[i] = 9

for i, item in enumerate(y_test):
    if item == 2:
        y_test[i] = 9

# One-hot encode the labels
y_train = to_categorical(y_train, 10)
y_test = to_categorical(y_test, 10)

# Define the CNN model
model = Sequential()

# First Conv2D layer
model.add(Conv2D(32, kernel_size=(3, 3), activation='relu', input_shape=(28, 28, 1)))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))

# Second Conv2D layer
model.add(Conv2D(64, kernel_size=(3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))

# Third Conv2D layer
model.add(Conv2D(64, kernel_size=(3, 3), activation='relu'))
model.add(Dropout(0.25))

# Flatten layer
model.add(Flatten())

# Fully connected Dense layer
model.add(Dense(10, activation='softmax'))

# Compile the model
model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])

# Print model summary
model.summary()

# Train the model
model.fit(x_train, y_train, batch_size=64, epochs=10, validation_data=(x_test, y_test))

# Evaluate the model
test_loss, test_acc = model.evaluate(x_test, y_test)
print(f"Test accuracy: {test_acc}")

model.save('now.h5')

Load the model 

save_model = keras.models.load_model('now.h5')
#save_model.summary()
#test_loss, test_acc = save_model.evaluate(x_test, y_test)
#print('Test accuracy:', test_acc)

RUN A Test to check if modle predicts "2" as  "9"

import cv2
predictions = save_model.predict(x_test)
for i in range(15):
    plt.imshow(x_test[i].reshape(28,28), cmap='gray')
    plt.show()
    print(f'Predicted: {np.argmax(predictions[i])}')
    print(f'Actual: {np.argmax(y_test[i])}')

---

The Problem I am facing that even if the model is created and completed , the upload always fails
if .keras model is accepted and sucesfully uploaded. 

IFG-Wing's ID was misclassified.

Plz TRYY and Discuss , wtf is the problem

---

If bored , these are the various models created  http://tiny.cc/trymodel