add pages and sandbox

app/model.py

@@ -0,0 +1,83 @@
+import os
+import requests
+import json
+import base64
+import io
+from PIL import Image
+
+
+def encode_image_to_base64(image):
+    # if image is str convert to BytesIO
+    if isinstance(image, str):
+        with Image.open(image) as img:
+            buffered = io.BytesIO()
+            img.save(buffered, format="JPEG")
+
+            bytes_data = buffered.getvalue()
+
+    else:
+        bytes_data = image.getvalue()
+
+    return base64.b64encode(bytes_data).decode("utf-8")
+
+
+def vision_ai_classify_image(base64_image):
+    # read OPENAI_API_KEY env
+    api_key = os.environ.get("OPENAI_API_KEY")
+    headers = {"Content-Type": "application/json", "Authorization": f"Bearer {api_key}"}
+
+    payload = {
+        "model": "gpt-4-vision-preview",
+        "messages": [
+            {
+                "role": "user",
+                "content": [
+                    {
+                        "type": "text",
+                        "text": """
+                You are an expert flooding detector.
+
+                You are given a image. You must detect if there is flooding in the image.
+
+                the output MUST be a json object with a boolean value for the key "flooding_detected".
+
+                If you don't know what to anwser, you can set the key "flooding_detect" as false.
+
+                Example:
+                {
+                    "flooding_detected": true
+                }
+                """,
+                    },
+                    {
+                        "type": "image_url",
+                        "image_url": {"url": f"data:image/jpeg;base64,{base64_image}"},
+                    },
+                ],
+            }
+        ],
+        "max_tokens": 300,
+    }
+
+    response = requests.post(
+        "https://api.openai.com/v1/chat/completions", headers=headers, json=payload
+    )
+    return response.json()
+
+
+def get_ai_label(response):
+    if response.get("error"):
+        return "Error"
+    else:
+        r = response["choices"][0]["message"]["content"]
+        json_string = r.replace("```json\n", "").replace("\n```", "")
+        json_object = json.loads(json_string)
+        return json_object["flooding_detected"]
+
+
+def run_model(img):
+    base64_image = encode_image_to_base64(img)
+
+    response = vision_ai_classify_image(base64_image)
+
+    return get_ai_label(response)

app/pages/🧪 Experimente o Modelo.py

@@ -0,0 +1,31 @@
+import streamlit as st
+from PIL import Image
+
+from model import run_model
+
+
+MAX_FILE_SIZE = 5 * 1024 * 1024  # 5MB
+
+st.set_page_config(layout="wide", page_title="Experimente o Modelo")
+
+st.markdown("# Identifique um alagamento por uma imagem")
+
+my_upload = st.file_uploader("Faça upload de uma imagem", type=["png", "jpg", "jpeg"])
+
+if my_upload is not None:
+    if my_upload.size > MAX_FILE_SIZE:
+        st.error(
+            "The uploaded file is too large. Please upload an image smaller than 5MB."
+        )
+    else:
+        res = run_model(my_upload)
+else:
+    my_upload = "./data/imgs/flooded1.jpg"
+    res = run_model(my_upload)
+
+if res:
+    st.markdown("### Alagamento Identificado ✅")
+else:
+    st.markdown("### Alagamento Não Identificado ❌")
+
+st.image(my_upload)

app/main.py → app/📣 Alagamentos em Tempo Real.py

@@ -16,7 +16,7 @@
 }
 
 tmp_data_path = Path(__file__).parent.parent / "data" / "cameras_h3_1.csv"
-chart_data = pd.read_csv(tmp_data_path)
+chart_data = pd.read_csv(tmp_data_path)[:10]
 chart_data["icon_data"] = None
 for i in chart_data.index:
     chart_data["icon_data"][i] = icon_data