CTF Who can train the neural network, CNN and Signal Processing?

Question

I was trying the challenge in InsomniHack and could not figure it out for 4 weeks.

In this example,

print(y.shape, train_data.shape, train_labels.shape)(5000,) (3935, 64) (3935,)

When we have 64 inputs and 32 outputs. Why do the dense layer implemets the equation y = f(Wx + b), where f is the activation function, W is the weight matrix and b is the bias?

Is it possible to try Signal Processing analysis without deep learning knowledge

strt_time = datetime.datetime.now()
history = model.fit(train_data, train_labels, epochs=EPOCHS,
                  validation_split=0.2, verbose=0,
                  callbacks=[])
curr_time = datetime.datetime.now()
timedelta = curr_time - strt_time
dnn_train_time = timedelta.total_seconds()
print("DNN training done. Time elapsed: ", timedelta.total_seconds(), "s")
plt.plot(history.epoch, np.array(history.history['val_loss']),
           label = 'Val loss')
plt.show()``` 

How can we claim the history value that is returned as a Python Dictionary when the hash of the previous block for Bitcoin block is #490624? 

Thank you.