- Joined
- Nov 24, 2022
- Messages
- 1
- Reaction score
- 0
I am currently working on an RNN LSTM project. I have 60.000 csv files as data to be trained. These csv files consist of outputs from a simulation, each giving a different Vx Vy Vz coordinator. (You can think of it as throwing a ball many times and drawing a trajectory related to it.) My aim is for the model I trained to acquire the characteristics of this ball and predict the path it will draw in 3d space by only giving the velocity and direction of fire. I thought it would be right to use LSTM RNN since it is a time dependent data. Before asking my question, I wanted to give information about the project.
My question is; What method should I apply when giving the data I have to the model? Is there a method to read this data separately and give it to the model?
I thought of a way to parse these files first and then train them with the help of the pandas library, but this time I wasn't sure if the shots were kept separate from each other.
In the attachment you can see what the data looks like:
My question is; What method should I apply when giving the data I have to the model? Is there a method to read this data separately and give it to the model?
I thought of a way to parse these files first and then train them with the help of the pandas library, but this time I wasn't sure if the shots were kept separate from each other.
In the attachment you can see what the data looks like:
