Upload 4 files

app.py

@@ -1,54 +1,52 @@
 import streamlit as st
 from PIL import Image
-import pandas as pd
 import torch
 from transformers import AutoFeatureExtractor, AutoModelForImageClassification
-from utils.matcher import oversæt_fuzzy
+import pandas as pd
+from utils.matcher import fuzzy_match
 
 @st.cache_resource
 def load_model():
-    extractor = AutoFeatureExtractor.from_pretrained("nateraw/food101")
-    model = AutoModelForImageClassification.from_pretrained("nateraw/food101")
+    extractor = AutoFeatureExtractor.from_pretrained("nateraw/food-classification")
+    model = AutoModelForImageClassification.from_pretrained("nateraw/food-classification")
     return extractor, model
 
 extractor, model = load_model()
-df = pd.read_csv("kaloriedata.csv")
-food_list = df["navn"].tolist()
+data = pd.read_csv("kaloriedata.csv")
+madliste = data["navn"].tolist()
+
+st.title("🍽️ WebKalorier – Madanalyse")
 
-st.title("🥗 Kalorieestimering via billede")
+uploaded = st.file_uploader("Upload et billede", type=["jpg", "jpeg", "png"])
 
-uploaded_file = st.file_uploader("Upload billede", type=["jpg", "jpeg", "png"])
-if uploaded_file:
-    img = Image.open(uploaded_file).convert("RGB")
-    st.image(img, caption="Dit billede", use_container_width=True)
+if uploaded:
+    img = Image.open(uploaded).convert("RGB")
+    st.image(img, caption="Uploadet billede", use_container_width=True)
 
     inputs = extractor(images=img, return_tensors="pt")
     with torch.no_grad():
         logits = model(**inputs).logits
     probs = torch.nn.functional.softmax(logits, dim=-1)
-    confidence, pred_class = torch.max(probs, dim=1)
-    label = model.config.id2label[pred_class.item()]
-    conf_value = confidence.item()
+    score, class_id = torch.max(probs, dim=1)
+    label = model.config.id2label[class_id.item()]
+    confidence = score.item()
 
-    st.markdown(f"🔍 Modelgæt: `{label}` ({conf_value:.0%} sikkerhed)")
-    if conf_value < 0.7:
-        selected = st.selectbox("Modellen er usikker – vælg korrekt fødevare:", food_list)
-    else:
-        selected = oversæt_fuzzy(label.replace("_", " "), food_list)
-
-    st.write(f"✅ Bruges som: **{selected}**")
-    row = df[df["navn"] == selected]
-    if not row.empty:
-        kcal_100 = row["kcal_pr_100g"].values[0]
-        vægt = 150  # dummy estimeret vægt
-        kcal = vægt * kcal_100 / 100
-        st.success(f"{vægt} g {selected} → {kcal:.0f} kcal")
-
-        # Feedback
-        user_feedback = st.text_input("Tilføj kommentar eller rettelse (valgfrit)")
-        if st.button("Send feedback"):
-            with open("feedback_log.csv", "a") as f:
-                f.write(f"{label},{selected},{conf_value:.2f},{user_feedback}\n")
-            st.info("✅ Feedback sendt. Tak!")
+    st.markdown(f"🤖 Modelgæt: `{label}` med {confidence:.0%} sikkerhed")
+
+    if confidence < 0.7:
+        valgt = st.selectbox("Vælg fødevare manuelt:", madliste)
     else:
-        st.warning("⚠️ Ukendt fødevare.")
+        valgt = fuzzy_match(label, madliste)
+
+    gram = st.number_input("Estimeret mængde (g):", 1, 1000, 150)
+    række = data[data["navn"] == valgt]
+    if not række.empty:
+        kcal100 = række["kcal_pr_100g"].values[0]
+        samlet = round(kcal100 * gram / 100)
+        st.success(f"{gram} g {valgt} = {samlet} kcal")
+
+    feedback = st.text_input("Feedback / korrektion:")
+    if st.button("Send feedback"):
+        with open("feedback_log.txt", "a") as f:
+            f.write(f"{label},{valgt},{confidence:.2f},{feedback}\n")
+        st.info("Tak for din feedback! 🙏")

kaloriedata.csv

@@ -1,20 +1,16 @@
 navn,kcal_pr_100g
 salat,15
 tomat,18
+ris,130
 pasta,131
-ost,350
-kød,250
-laks,210
-burger,280
+kylling,239
 æg,155
-ris,130
-brød,260
-pizza,270
-gulerod,41
-æble,52
-appelsin,47
-pommes frites,290
 smør,717
-kartofler,77
+kartoffel,77
 broccoli,35
-kylling,239
+laks,210
+oksekød,250
+gulerod,41
+pizza,270
+burger,280
+brød,260

requirements.txt

@@ -1,7 +1,7 @@
 streamlit
+pillow
 torch
 transformers
-pillow
 pandas
 fuzzywuzzy
 python-Levenshtein

utils/matcher.py

@@ -1,6 +1,6 @@
 from fuzzywuzzy import process
 
-def oversæt_fuzzy(navn, kandidater):
-    navn = navn.lower()
-    result = process.extractOne(navn, kandidater, score_cutoff=70)
-    return result[0] if result else navn
+def fuzzy_match(label, liste):
+    label = label.replace("_", " ").lower()
+    match, score = process.extractOne(label, liste)
+    return match if score > 70 else label