Deploy ML : Error: [mobx-state-tree] No matching type for union (string[] | undefined?)

Hello,

I am currently trying to test an ML backend with a model I created myself. As a practice, I am trying to draw only rectangles in the images. However, when I register the model and open the images in the project, I get the following message on the screen:

Error: [mobx-state-tree] No matching type for union (string[] | undefined?)

I apologize for the inconvenience, but even after trying to resolve it, the issue persists. I found a similar post in a previous issue where it turned out to be a system error, so I am asking for help.

Below, I am attaching the project template and model.py that I am writing as a test.

template
```

```

model.py
```
from typing import List, Dict, Optional
from label_studio_ml.model import LabelStudioMLBase
from label_studio_ml.response import ModelResponse
import requests
from PIL import Image
from io import BytesIO
import os

LS_URL = os.environ['LABEL_STUDIO_URL']
LS_API_TOKEN = os.environ['LABEL_STUDIO_API_KEY']

class NewModel(LabelStudioMLBase):

def __init__(self, **kwargs):
super(NewModel, self).__init__(**kwargs)
self.setup()

def setup(self):

self.set("model_version", "0.0.1")
self.base_url = os.getenv('LABEL_STUDIO_BASE_URL', '****')

from_name, schema = list(self.parsed_label_config.items())[0]
self.from_name = from_name
self.to_name = schema['to_name'][0]
self.labels = ['human']

def predict(self, tasks: List[Dict], context: Optional[Dict] = None, **kwargs) -> ModelResponse:

print(f"Run prediction on {tasks}")
print(f"Received context: {context}")
print(f"Project ID: {self.project_id}")
print(f"Label config: {self.label_config}")
print(f"Parsed JSON Label config: {self.parsed_label_config}")
print(f"Extra params: {self.extra_params}")

predictions = []

for task in tasks:
image_url = LS_URL + task['data']['image']
print("image_url:", image_url)
header = {"Authorization": "Token " + LS_API_TOKEN}
response = requests.get(image_url, headers=header)
image = Image.open(BytesIO(response.content))
img_width, img_height = image.size

rect_width = 50
rect_height = 50

x_center = img_width / 2
y_center = img_height / 2

x = x_center - rect_width / 2
y = y_center - rect_height / 2

print("str(task['id']:", str(task['id']))
print("from_name:", self.from_name)
print("x / img_width * 100:", x / img_width * 100)
print("x / img_height * 100:", y / img_height * 100)
print("width: rect_width / img_width * 100:", rect_width / img_width * 100)
print("height: rect_height / img_height * 100:", rect_height / img_height * 100)
print("[self.labels[0]]:",[self.labels[0]])

predictions.append({
"id": "result",
"from_name": "label",
"to_name": "image",
"type": "rectanglelabels",
"score": 0.90,
"original_width": img_width,
"original_height": img_height,
"image_rotation": 0,
"value": {
"rotation": 0,
"x": x / img_width * 100,
"y": y / img_height * 100,
"width": rect_width / img_width * 100,
"height": rect_height / img_height * 100,
"rectanglelabels": ["human"]
}
})

return [{"result": predictions, "score": 0.5, "model_version": "v8"}]

```