Using a Raspberry Pi 4B to recognize CSV files in a folder via a classification tree

IOT025454 · July 23, 2024, 6:38pm

I want to train a model that can recognize numbers in a CSV file, but I couldn’t clearly figure out what kind of training method I should use. When I type “edge-impulse-linux -runner”, it asks me to choose a sensor to detect data. I want to find ways to process a CSV file when I deploy this model on my RPi4.

project ID:378067

Eoin · July 24, 2024, 10:11am

Hi @IOT025454

You can upload your CSV to our platform as training data:

edge-impulse-uploader your-data.csv

However I’d recommend looking at our custom sensor example in the python linux sdk examples here over any impulse runner logic to batch process that CSV file as a custom sensor for running the inference:

github.com

edgeimpulse/linux-sdk-python/blob/master/examples/custom/classify.py

import os
import sys, getopt
import signal
import time
from edge_impulse_linux.runner import ImpulseRunner
import io

runner = None

def signal_handler(sig, frame):
    print('Interrupted')
    if (runner):
        runner.stop()
    sys.exit(0)

signal.signal(signal.SIGINT, signal_handler)

def help():
    print('python classify.py <path_to_model.eim> <path_to_features.txt>')

This file has been truncated. show original

This will give you a base application to build your solution around.

Best

Eoin

IOT025454 · July 24, 2024, 10:42am

I can’t thank you enough! i’ll try to do it!