Skip to content

usage.md

Latest commit

 

History

History
55 lines (41 loc) · 1.74 KB

usage.md

File metadata and controls

55 lines (41 loc) · 1.74 KB

Usage Guide

This is a guide for using the Smart Pixels ML project to train and evaluate the models on the simulated data.

1. Installation

Refer to Readme for installation instructions.

2. Data Collection

Download the simulated data from zenodo and PixelAV Add other links here

Ensure the two directories Data and Labels are present.

3. Data Preparation

  • Define the paths to the data and labels directories (look at utils for more details)
  • Configure datagenerator parameters (look at data_generator for more details)
  • Create training and validation datagenerators

4. Model Creation

  • Define the model architecture and compile.
  • Also look at the summary of the model to ensure it is correct. Look at model for more details on how to do that. For loss function see loss.

5. Model Training

If everything is set up correctly, the training should start and run seamlessly. For example:

model.fit(
    x=training_generator, 
    validation_data=validation_generator, 
    epochs=200, 
    verbose=1)

After training, check the loss and accuracy of the model. And save the model weights.

6. Model Evaluation

Initiate the model and Load the weights as

model=CreateModel((13,21,2),n_filters=5,pool_size=3)
model.load_weights("model_weights.h5")

And then evaluate the model as done in evaluate.py or in evaluate.py

7. Model Prediction

Look at predict

8. Add Additional Instructions

Here are some additional instructions