Getting a 401 error when doing inference

[Ozzy08]

Hello,

I am getting a 401 error when doing inference, and I am not sure why. I am using rock-paper-scissors-sxsw/11 as my model. This is my code:

def resource_path(relative_path):
    if hasattr(sys, '_MEIPASS'):
        return os.path.join(sys._MEIPASS, relative_path)
    return os.path.join(os.path.abspath("."), relative_path)

build = []
url_base = 'https://detect.roboflow.com/'
endpoint = os.getenv('ROBOFLOW_URL')
access_token = os.getenv('ROBOFLOW_API_KEY')
format = '&format=json'
confidence = '&confidence=35'
stroke = '&stroke=3'
overlap = '&overlap=50'
build.append(url_base)
build.append(endpoint)
build.append(access_token)
build.append(format)
build.append(confidence)
build.append(overlap)
build.append(stroke)
url = ''.join(build)
name = "picture" + str(time.time()) + ".png"
my_screenshot = pyautogui.screenshot()
my_screenshot.save(name)
f = name
# Creating the kernel(2d convolution matrix)
kernel = np.array([
    [0, -1, 0],
    [-1, 5, -1],
    [0, -1, 0]
])
image = cv2.imread(f)
# Applying the filter2D() function
img = cv2.filter2D(src=image, ddepth=-1, kernel=kernel)
clean = cv2.imwrite(name, img)
image = Image.open(name).convert("RGB")
# Convert to JPEG Buffer
buffered = io.BytesIO()
image.save(buffered, quality=90, format="JPEG")
# Construct the URL
m = MultipartEncoder(fields={'file': (f, buffered.getvalue(), resource_path("image/jpeg"))})
r = requests.post(url, data=m, headers={'Content-Type': m.content_type})

[Jacob_Witt]

Hi! It appears you are not adding a ?api_key= element. My advice is to use our native SDK, which will avoid these types of errors.

rf = Roboflow(api_key="API_KEY")
project = rf.workspace().project("MODEL_ENDPOINT")
model = project.version(VERSION).model

# infer on a local image
print(model.predict("your_image.jpg", confidence=40, overlap=30).json())

# visualize your prediction
# model.predict("your_image.jpg", confidence=40, overlap=30).save("prediction.jpg")

# infer on an image hosted elsewhere
# print(model.predict("URL_OF_YOUR_IMAGE", hosted=True, confidence=40, overlap=30).json())```

[Ozzy08]

Thank you, but quick question. How does this code work when you are trying to infer from a model that is not yours but is on the roboflow universe?

[Jacob_Witt]

If you go to the universe page for the model and scroll down, you’ll see code to deploy.