Need help i'm new to Tensorflow and Keras

TensorFlow General Discussion
,
1

Hi !

i’m currently trying to create a model with keras from data that I manipulate from an excel file

The data is the following

each row contain 62 variable from 0 to 1

each training need 10 of those row so a x_train value would be an array [10][62]

the output should be 4 float that represent a price of some sort like [0.0145, 0.0143, 0.0165, 0.0170]

Now when i try to do my model I really don’T understand where to start ?

how to clean the data for the model,
how many layer, what type of layer and what type of activation
what the last layer should look to fit the outputs I need
what Optimizer, loss and metrics are the best

If someone can help me even in just one part it would be really appreciated.
Have a nice day !
ps. sorry if there is mistake in the text i’m french so I tried my best

3

Maybe you can start with something like this:

import tensorflow as tf

# Define the model
model = tf.keras.Sequential([
    tf.keras.layers.Dense(16, activation='relu', input_dim=62, batch_size=10),  # Input layer with 8 input features, 16 units, and ReLU activation
    tf.keras.layers.Dense(4, activation='sigmoid')  # Output layer with 4 units and sigmoid activation
])

# Print a summary of the model's architecture
model.summary()

But I would recommend you to first follow a basic tutorial (just google it), so that you do not repeat or copy/paste code. Neural Net code is difficult to debug for beginners because of tensor’s dimensions etc.