andrisgauracs

def valuechange(self):
        # We access our global variables within the function
        global framesTaken
        global framesSpecified

        # We set the framesSpecified value to
        # whatever is specified on the spinbox
        framesSpecified = self.sp.value()

        # We modify our label to give us info, how many

andrisgauracs

def take_snapshot(self):
        # Import the global variables we'll be using
        global framesTaken, clusters, borderSize, offset
        # This will be needed to check if the player was playing at the
        # time of the button press
        wasPlaying = None
        # We need to get the width and height of the video file
        videoSize = self.mediaplayer.video_get_size()
        # This is the VLC function, that let's us
        # take a snap shot of the video frame and save it in the directory

andrisgauracs

# Courtesy of https://www.pyimagesearch.com/2014/05/26/opencv-python-k-means-color-clustering/
def centroid_histogram(clt):
    # grab the number of different clusters and create a histogram
    # based on the number of pixels assigned to each cluster
    numLabels = np.arange(0, len(np.unique(clt.labels_)) + 1)
    (hist, _) = np.histogram(clt.labels_, bins=numLabels)
    # normalize the histogram, such that it sums to one
    hist = hist.astype("float")
    hist /= hist.sum()
    # return the histogram

andrisgauracs

# Courtesy of https://www.pyimagesearch.com/2014/05/26/opencv-python-k-means-color-clustering/
def plot_colors(hist, centroids):
    # initialize the bar chart representing the relative frequency
    # of each of the colors
    bar = np.zeros((50, 300, 3), dtype="uint8")
    startX = 0

    # Sort the centroids to form a gradient color look
    centroids = sorted(centroids, key=lambda x: sum(x))

andrisgauracs

        # Let's fetch the video frame we just captured
        imagePath = os.getcwd() + "/img_"+str(framesTaken)+".png"
        # Transform the image to an OpenCV readable image
        image = cv2.imread(imagePath)

        # Let's make a copy of this image
        # to use for the color palette generation
        image_copy = image_resize(cv2.cvtColor(
            image, cv2.COLOR_BGR2RGB), width=100)

andrisgauracs

        # To show a thumbnail version of this image, we need to store it
        # in a pixmap and then add it to a label widget
        pixmap = QtGui.QPixmap()
        pixmap.load(imagePath)
        pixmap = pixmap.scaledToWidth(50)
        self.imageBoxes[framesTaken].setPixmap(pixmap)
        framesTaken = framesTaken + 1

        # Now that we have at least one image captured and processed,
        # we can go ahead and show the bottom layout with our thumbnail labels

andrisgauracs

        # Once the list is full with all the images we need, we can start to
        # combine it into the final output image
        if (framesTaken == framesSpecified):
            resultImgs = []
            for i in framesTakenList:
                # here we just add a whitespace rectangle as a top margin
                resultImgs.append(cv2.imread(i))
                im = np.zeros((
                    borderSize*2,
                    cv2.imread(i).shape[1],

andrisgauracs

#!/usr/bin/env bash
# esp32-ai: clone, train, export, build, and flash — end to end
# Reproduces github.com/slvDev/esp32-ai from a bare board.
# Requires: ESP32-S3 N16R8 (16MB flash, 8MB PSRAM).
#
# Usage:
#   ./build_and_flash.sh              # full run: data prep, train, export, build, flash
#   ./build_and_flash.sh --skip-train # reuse an existing firmware/model/model.bin, just build + flash
set -euo pipefail

andrisgauracs

#!/usr/bin/env bash
# esp32-ai: test a custom prompt against an already-trained, already-flashed model.
# Does NOT retrain and does NOT rewrite the 15MB model partition.
#
# Usage:
#   ./run_custom_prompt.sh "Once there was a robot"
set -euo pipefail

if [ "$#" -lt 1 ]; then
  echo "Usage: $0 \"Your custom prompt\""