How to deploy the program on arduino

First of all, I am very grateful to edge impulse for its contribution to the development of neural networks.I am currently working on a module for human motion recognition based on acceleration.My ultimate goal is to use the arduino nano 33 ble to collect and recognize motion data, and output the results through the pins on the module to achieve the output of the detection results.However, I still have some questions and difficulties。
1、Use arduino ide to verify the network (by inputting characteristic data), but how to realize continuous data collection and motion detection through arduino IDE? Is it possible to achieve my request only through CLI (I saw it in some topics)?
2、How can I make the module work automatically when it is powered on and work offline ,instead of connecting to the computer and inputting instructions to start the module every time?
3、How can I use the pins of the module to output the test results (instead of the micro usb port)?

Hi @angfan123,

For 2 and 3: export as Arduino library from Deployment, then add the Library to the Arduino IDE via Sketch > Include library > From ZIP file. Then find the example sketch under File > Examples > Your project name > nano_ble33_sense_accelerometer. This is an Arduino sketch that you can modify to control pins and it starts automatically in inference mode.

For 1 try this (replace #include <continuous_gestures_demo_inference.h> with the header name you find from the example above):

/* Edge Impulse Arduino examples
 * Copyright (c) 2020 EdgeImpulse Inc.
 *
 * Permission is hereby granted, free of charge, to any person obtaining a copy
 * of this software and associated documentation files (the "Software"), to deal
 * in the Software without restriction, including without limitation the rights
 * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
 * copies of the Software, and to permit persons to whom the Software is
 * furnished to do so, subject to the following conditions:
 *
 * The above copyright notice and this permission notice shall be included in
 * all copies or substantial portions of the Software.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 * SOFTWARE.
 */

/* Includes ---------------------------------------------------------------- */
#include <continuous_gestures_demo_inference.h>
#include <Arduino_LSM9DS1.h>

/* Constant defines -------------------------------------------------------- */
#define CONVERT_G_TO_MS2    9.80665f

/* Private variables ------------------------------------------------------- */
static bool debug_nn = false; // Set this to true to see e.g. features generated from the raw signal
static uint32_t run_inference_every_ms = 200;
static rtos::Thread inference_thread(osPriorityLow);
static float buffer[EI_CLASSIFIER_DSP_INPUT_FRAME_SIZE] = { 0 };

/**
* @brief      Arduino setup function
*/
void setup()
{
    delay(5000);

    // put your setup code here, to run once:
    Serial.begin(115200);
    Serial.println("Edge Impulse Inferencing Demo");

    if (!IMU.begin()) {
        ei_printf("Failed to initialize IMU!\r\n");
    }
    else {
        ei_printf("IMU initialized\r\n");
    }

    if (EI_CLASSIFIER_RAW_SAMPLES_PER_FRAME != 3) {
        ei_printf("ERR: EI_CLASSIFIER_RAW_SAMPLES_PER_FRAME should be equal to 3 (the 3 sensor axes)\n");
        return;
    }

    inference_thread.start(&run_inference);
}

/**
* @brief      Printf function uses vsnprintf and output using Arduino Serial
*
* @param[in]  format     Variable argument list
*/
void ei_printf(const char *format, ...) {
   static char print_buf[1024] = { 0 };

   va_list args;
   va_start(args, format);
   int r = vsnprintf(print_buf, sizeof(print_buf), format, args);
   va_end(args);

   if (r > 0) {
       Serial.write(print_buf);
   }
}

void run_inference()
{ 
    // wait until we have a full buffer
    delay((EI_CLASSIFIER_INTERVAL_MS * EI_CLASSIFIER_RAW_SAMPLE_COUNT) + 100);

    while (1) {
        // copy the buffer
        float inference_buffer[EI_CLASSIFIER_DSP_INPUT_FRAME_SIZE];
        memcpy(inference_buffer, buffer, EI_CLASSIFIER_DSP_INPUT_FRAME_SIZE * sizeof(float));

        // Turn the raw buffer in a signal which we can the classify
        signal_t signal;
        int err = numpy::signal_from_buffer(inference_buffer, EI_CLASSIFIER_DSP_INPUT_FRAME_SIZE, &signal);
        if (err != 0) {
            ei_printf("Failed to create signal from buffer (%d)\n", err);
            return;
        }

        // Run the classifier
        ei_impulse_result_t result = { 0 };

        err = run_classifier(&signal, &result, debug_nn);
        if (err != EI_IMPULSE_OK) {
            ei_printf("ERR: Failed to run classifier (%d)\n", err);
            return;
        }

        // print the predictions
        ei_printf("Predictions (DSP: %d ms., Classification: %d ms., Anomaly: %d ms.): \n",
            result.timing.dsp, result.timing.classification, result.timing.anomaly);
        for (size_t ix = 0; ix < EI_CLASSIFIER_LABEL_COUNT; ix++) {
            ei_printf("    %s: %.5f\n", result.classification[ix].label, result.classification[ix].value);
        }
#if EI_CLASSIFIER_HAS_ANOMALY == 1
        ei_printf("    anomaly score: %.3f\n", result.anomaly);
#endif

        delay(run_inference_every_ms);
    }
}

/**
* @brief      Get data and run inferencing
*
* @param[in]  debug  Get debug info if true
*/
void loop()
{
    while (1) {
        // Determine the next tick (and then sleep later)
        uint64_t next_tick = micros() + (EI_CLASSIFIER_INTERVAL_MS * 1000);

        // roll the buffer -3 points so we can overwrite the last one
        numpy::roll(buffer, EI_CLASSIFIER_DSP_INPUT_FRAME_SIZE, -3);

        // read to the end of the buffer
        IMU.readAcceleration(
            buffer[EI_CLASSIFIER_DSP_INPUT_FRAME_SIZE - 3], 
            buffer[EI_CLASSIFIER_DSP_INPUT_FRAME_SIZE - 2], 
            buffer[EI_CLASSIFIER_DSP_INPUT_FRAME_SIZE - 1]
        );

        buffer[EI_CLASSIFIER_DSP_INPUT_FRAME_SIZE - 3] *= CONVERT_G_TO_MS2;
        buffer[EI_CLASSIFIER_DSP_INPUT_FRAME_SIZE - 2] *= CONVERT_G_TO_MS2;
        buffer[EI_CLASSIFIER_DSP_INPUT_FRAME_SIZE - 1] *= CONVERT_G_TO_MS2;

        // and wait for next tick
        uint64_t time_to_wait = next_tick - micros();
        delay((int)floor((float)time_to_wait / 1000.0f));
        delayMicroseconds(time_to_wait % 1000);
    }
}

#if !defined(EI_CLASSIFIER_SENSOR) || EI_CLASSIFIER_SENSOR != EI_CLASSIFIER_SENSOR_ACCELEROMETER
#error "Invalid model for current sensor"
#endif

Note that you’ll need to do some averaging here to make sure transitions between motions are handled properly (e.g. look at the last 10 inference results and only do something when 70% of them are the same). I’ll add this sample as well to our Arduino library exports.

Thank you very much, I will study your reply carefully next, thank you