Upload App

.idea/.gitignore

@@ -0,0 +1,8 @@
+# Default ignored files
+/shelf/
+/workspace.xml
+# Editor-based HTTP Client requests
+/httpRequests/
+# Datasource local storage ignored files
+/dataSources/
+/dataSources.local.xml

.idea/FlaskProject1.iml

@@ -0,0 +1,19 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<module type="PYTHON_MODULE" version="4">
+  <component name="Flask">
+    <option name="enabled" value="true" />
+  </component>
+  <component name="NewModuleRootManager">
+    <content url="file://$MODULE_DIR$" />
+    <orderEntry type="jdk" jdkName="Python 3.13 (PythonProject) (2)" jdkType="Python SDK" />
+    <orderEntry type="sourceFolder" forTests="false" />
+  </component>
+  <component name="TemplatesService">
+    <option name="TEMPLATE_CONFIGURATION" value="Jinja2" />
+    <option name="TEMPLATE_FOLDERS">
+      <list>
+        <option value="$MODULE_DIR$/../FlaskProject1\templates" />
+      </list>
+    </option>
+  </component>
+</module>

.idea/inspectionProfiles/profiles_settings.xml

@@ -0,0 +1,6 @@
+<component name="InspectionProjectProfileManager">
+  <settings>
+    <option name="USE_PROJECT_PROFILE" value="false" />
+    <version value="1.0" />
+  </settings>
+</component>

.idea/misc.xml

@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="Black">
+    <option name="sdkName" value="Python 3.13 (PythonProject) (2)" />
+  </component>
+</project>

.idea/modules.xml

@@ -0,0 +1,8 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="ProjectModuleManager">
+    <modules>
+      <module fileurl="file://$PROJECT_DIR$/.idea/FlaskProject1.iml" filepath="$PROJECT_DIR$/.idea/FlaskProject1.iml" />
+    </modules>
+  </component>
+</project>

.idea/vcs.xml

@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+  <component name="VcsDirectoryMappings">
+    <mapping directory="$PROJECT_DIR$" vcs="Git" />
+  </component>
+</project>

Dockerfile

@@ -0,0 +1,22 @@
+# Gunakan Python 3.9
+FROM python:3.9
+
+# Set folder kerja
+WORKDIR /code
+
+# Copy requirements dan install dependencies
+COPY ./requirements.txt /code/requirements.txt
+RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt
+
+# Copy seluruh file codingan ke server
+COPY . .
+
+# Buat cache folder agar model tidak didownload berulang kali
+RUN mkdir -p /code/cache
+ENV TRANSFORMERS_CACHE=/code/cache
+
+# Buka port 7860 (Port wajib Hugging Face)
+EXPOSE 7860
+
+# Jalankan aplikasi menggunakan Gunicorn
+CMD ["gunicorn", "-b", "0.0.0.0:7860", "app:app"]

app.py

@@ -0,0 +1,33 @@
+from flask import Flask, render_template, request, redirect, url_for
+from model_utils import predict_text, process_file, AVAILABLE_MODELS
+
+app = Flask(__name__)
+
+@app.route('/', methods=['GET', 'POST'])
+def index():
+    if request.method == 'POST':
+        model_choice = request.form.get('model_choice', 'toxic_bert')
+        model_info = AVAILABLE_MODELS.get(model_choice, {}).get('desc', 'Unknown Model')
+
+        # 1. Cek Upload File
+        if 'file_input' in request.files and request.files['file_input'].filename != '':
+            file = request.files['file_input']
+            results = process_file(file, model_choice)
+            return render_template('result_file.html', results=results, model_name=model_info)
+
+        # 2. Cek Input Manual
+        raw_text = request.form.get('user_text')
+        if raw_text:
+            result = predict_text(raw_text, model_choice)
+            # Debugging print
+            # print("Hasil:", result)
+            return render_template('result.html', data=result, model_name=model_info)
+
+    return render_template('index.html', models=AVAILABLE_MODELS)
+
+@app.route('/ulang')
+def ulang():
+    return redirect(url_for('index'))
+
+if __name__ == '__main__':
+    app.run(debug=True, port=5000)

model_utils.py

@@ -0,0 +1,121 @@
+import torch
+import torch.nn.functional as F
+import re
+import string
+import pandas as pd
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+
+# ==========================================
+# KONFIGURASI MODEL
+# ==========================================
+AVAILABLE_MODELS = {
+    "toxic_bert": {
+        "name": "Dzeisonov/indobert-toxic-classifier",
+        "desc": "IndoBERT (Base) - Toxic Classifier"
+    },
+    "toxic_roberta": {
+        "name": "Dzeisonov/indoroberta-toxic-classifier",
+        "desc": "IndoRoBERTa (Base) - Toxic Classifier"
+    }
+}
+
+# Cache global
+loaded_models = {}
+
+def get_model_and_tokenizer(model_key):
+    """Load model secara lazy loading."""
+    if model_key not in AVAILABLE_MODELS:
+        model_key = "toxic_bert"
+
+    if model_key in loaded_models:
+        return loaded_models[model_key]['tokenizer'], loaded_models[model_key]['model']
+
+    config = AVAILABLE_MODELS[model_key]
+    print(f"⏳ Sedang memuat model baru: {config['name']} ...")
+
+    try:
+        tokenizer = AutoTokenizer.from_pretrained(config['name'])
+        model = AutoModelForSequenceClassification.from_pretrained(config['name'])
+        loaded_models[model_key] = {'tokenizer': tokenizer, 'model': model}
+        print("✅ Model berhasil dimuat!")
+        return tokenizer, model
+    except Exception as e:
+        print(f"❌ Gagal memuat model: {e}")
+        return None, None
+
+def preprocess_text(text):
+    if not isinstance(text, str) or not text: return ""
+    text = text.lower()
+    text = re.sub(r"http\S+|www.\S+|@\w+|#|\d+", "", text)
+    text = text.translate(str.maketrans("", "", string.punctuation))
+    text = re.sub(r"\s+", " ", text).strip()
+    return text
+
+def predict_text(text, model_key):
+    """Prediksi satu kalimat."""
+    tokenizer, model = get_model_and_tokenizer(model_key)
+
+    if not model or not tokenizer:
+        return {"original_text": text, "label": "ERROR", "score": "0%"}
+
+    clean_text = preprocess_text(text)
+    if not clean_text:
+        return {"original_text": text, "label": "Kosong", "score": "0%"}
+
+    inputs = tokenizer(clean_text, return_tensors="pt", truncation=True, max_length=512)
+    
+    with torch.no_grad():
+        outputs = model(**inputs)
+        probs = F.softmax(outputs.logits, dim=-1)
+        label_id = torch.argmax(probs, dim=1).item()
+        confidence = probs[0][label_id].item()
+
+    predicted_label = model.config.id2label[label_id]
+    
+    # Standarisasi Label (Toxic / Non-Toxic)
+    if predicted_label in ["LABEL_1", "Toxic", "toxic", "1"]:
+        final_label = "Toxic"
+    else:
+        final_label = "Non-Toxic"
+
+    return {
+        "original_text": text,
+        "text_clean": clean_text,
+        "label": final_label,
+        "score": f"{confidence:.1%}" # Mengembalikan persentase (misal: 98.5%)
+    }
+
+def process_file(file_obj, model_key):
+    """Memproses file CSV, Excel, atau TXT."""
+    results = []
+    texts = []
+
+    try:
+        filename = file_obj.filename.lower()
+        
+        # 1. Jika file CSV
+        if filename.endswith('.csv'):
+            df = pd.read_csv(file_obj)
+            texts = df.iloc[:, 0].astype(str).tolist() # Ambil kolom pertama
+            
+        # 2. Jika file Excel (.xlsx / .xls)
+        elif filename.endswith(('.xlsx', '.xls')):
+            df = pd.read_excel(file_obj)
+            texts = df.iloc[:, 0].astype(str).tolist() # Ambil kolom pertama
+            
+        # 3. Jika file TXT (fallback)
+        else:
+            content = file_obj.read().decode("utf-8")
+            texts = content.splitlines()
+
+        # Batasi 50 baris agar server tidak hang
+        for text in texts[:50]:
+            if text.strip():
+                res = predict_text(text, model_key)
+                results.append(res)
+                
+    except Exception as e:
+        print(f"Error processing file: {e}")
+        return []
+
+    return results

requirements.txt

@@ -0,0 +1,7 @@
+flask
+torch
+transformers
+pandas
+openpyxl
+gunicorn
+uvicorn

templates/index.html

@@ -0,0 +1,87 @@
+<!DOCTYPE html>
+<html lang="id">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>Indo Toxic Detector</title>
+    <link href="https://cdn.jsdelivr.net/npm/bootstrap@5.3.0/dist/css/bootstrap.min.css" rel="stylesheet">
+    <style>
+        body { background-color: #f0f2f5; font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif; }
+        .main-card { 
+            border-radius: 20px; 
+            border: none; 
+            box-shadow: 0 15px 35px rgba(0,0,0,0.1); 
+            background: white;
+        }
+        .header-title { color: #2c3e50; font-weight: 800; }
+        .btn-custom { background-color: #2c3e50; color: white; border-radius: 10px; padding: 12px; }
+        .btn-custom:hover { background-color: #1a252f; color: white; }
+        .info-box {
+            background-color: #e8f4fd;
+            border-left: 4px solid #0d6efd;
+            border-radius: 5px;
+            font-size: 0.9rem;
+        }
+    </style>
+</head>
+<body>
+    <div class="container py-5 min-vh-100 d-flex justify-content-center align-items-center">
+        <div class="col-lg-6 col-md-8">
+            <div class="text-center mb-4">
+                <h1 class="header-title">🛡️ Indo Toxicity Detector</h1>
+                <p class="text-muted">Klasifikasi ujaran kebencian dengan AI</p>
+            </div>
+
+            <div class="card main-card p-4">
+                <form action="/" method="POST" enctype="multipart/form-data">
+                    
+                    <!-- Pilihan Model -->
+                    <div class="mb-4">
+                        <label class="form-label fw-bold text-uppercase small text-secondary">Pilih Model AI</label>
+                        <select name="model_choice" class="form-select bg-light border-0 py-3">
+                            {% for key, val in models.items() %}
+                                <option value="{{ key }}">{{ val.desc }}</option>
+                            {% endfor %}
+                        </select>
+                    </div>
+
+                    <hr class="my-4">
+
+                    <!-- Input Manual -->
+                    <div class="mb-3">
+                        <label class="form-label fw-bold">📝 Input Manual</label>
+                        <textarea class="form-control bg-light" name="user_text" rows="3" placeholder="Ketik kalimat di sini..."></textarea>
+                    </div>
+
+                    <div class="text-center text-muted small my-2">- ATAU -</div>
+
+                    <!-- Input File -->
+                    <div class="mb-4">
+                        <label class="form-label fw-bold">📂 Upload File</label>
+                        <input type="file" class="form-control mb-2" name="file_input" accept=".csv, .txt, .xlsx, .xls">
+                        
+                        <!-- PANDUAN FORMAT FILE -->
+                        <div class="info-box p-3 mt-3">
+                            <h6 class="fw-bold mb-2">ℹ️ Panduan Format File:</h6>
+                            <ul class="mb-0 ps-3">
+                                <li class="mb-1">
+                                    <strong>Excel / CSV:</strong> Teks harus berada di 
+                                    <span class="badge bg-primary">Kolom Pertama (A)</span>.
+                                </li>
+                                <li>
+                                    <strong>TXT:</strong> Satu kalimat per baris.
+                                </li>
+                            </ul>
+                            <div class="text-muted small mt-2 fst-italic">*Maksimal 50 baris akan diproses.</div>
+                        </div>
+                    </div>
+
+                    <button type="submit" class="btn btn-custom w-100 fw-bold">
+                        🔍 Mulai Analisis
+                    </button>
+                </form>
+            </div>
+        </div>
+    </div>
+</body>
+</html>

templates/result.html

@@ -0,0 +1,50 @@
+<!DOCTYPE html>
+<html lang="id">
+<head>
+    <meta charset="UTF-8">
+    <title>Hasil Analisis</title>
+    <link href="https://cdn.jsdelivr.net/npm/bootstrap@5.3.0/dist/css/bootstrap.min.css" rel="stylesheet">
+    <style>
+        body { background-color: #f8f9fa; display: flex; align-items: center; min-height: 100vh;}
+        .card { border-radius: 20px; border: none; box-shadow: 0 10px 30px rgba(0,0,0,0.08); }
+        .badge-status { font-size: 1.2rem; padding: 10px 25px; border-radius: 50px; }
+        .bg-toxic { background-color: #e74c3c; color: white; }
+        .bg-safe { background-color: #27ae60; color: white; }
+    </style>
+</head>
+<body>
+    <div class="container">
+        <div class="row justify-content-center">
+            <div class="col-md-7">
+                <div class="card p-5 text-center">
+                    
+                    <h6 class="text-muted text-uppercase ls-2">Teks Asli</h6>
+                    <p class="lead fst-italic mb-4">"{{ data.original_text }}"</p>
+                    <hr>
+
+                    <h6 class="text-muted text-uppercase mt-4">Hasil Prediksi</h6>
+                    <p class="small text-muted mb-3">Model: {{ model_name }}</p>
+
+                    <div>
+                        {% if data.label == 'Toxic' %}
+                            <span class="badge badge-status bg-toxic">☣️ TOXIC</span>
+                            <div class="alert alert-danger mt-4 border-0 bg-danger bg-opacity-10 text-danger">
+                                <strong>Peringatan!</strong> Kalimat ini mengandung unsur negatif/kasar.
+                            </div>
+                        {% else %}
+                            <span class="badge badge-status bg-safe">✅ NON-TOXIC</span>
+                            <div class="alert alert-success mt-4 border-0 bg-success bg-opacity-10 text-success">
+                                <strong>Aman.</strong> Kalimat ini bersih.
+                            </div>
+                        {% endif %}
+                    </div>
+
+                    <p class="mt-2 text-muted small">Confidence: <strong>{{ data.score }}</strong></p>
+
+                    <a href="{{ url_for('ulang') }}" class="btn btn-outline-dark mt-4 px-5 rounded-pill">Cek Lagi</a>
+                </div>
+            </div>
+        </div>
+    </div>
+</body>
+</html>

templates/result_file.html

@@ -0,0 +1,53 @@
+<!DOCTYPE html>
+<html lang="id">
+<head>
+    <meta charset="UTF-8">
+    <title>Hasil Batch Analysis</title>
+    <link href="https://cdn.jsdelivr.net/npm/bootstrap@5.3.0/dist/css/bootstrap.min.css" rel="stylesheet">
+    <style>
+        .badge-toxic { background-color: #e74c3c; color: white; }
+        .badge-safe { background-color: #27ae60; color: white; }
+    </style>
+</head>
+<body class="bg-light p-5">
+    <div class="container bg-white p-5 rounded shadow-sm">
+        <div class="d-flex justify-content-between align-items-center mb-4">
+            <h2>📄 Hasil Analisis File</h2>
+            <a href="{{ url_for('ulang') }}" class="btn btn-outline-secondary">⬅️ Kembali</a>
+        </div>
+        
+        <p class="text-muted">Model: <strong>{{ model_name }}</strong></p>
+        
+        <div class="table-responsive">
+            <table class="table table-hover align-middle">
+                <thead class="table-dark">
+                    <tr>
+                        <th style="width: 60%;">Teks</th>
+                        <th style="width: 20%;">Prediksi</th>
+                        <th style="width: 20%;">Confidence</th>
+                    </tr>
+                </thead>
+                <tbody>
+                    {% for row in results %}
+                    <tr>
+                        <td class="text-break">{{ row.original_text }}</td>
+                        <td>
+                            {% if row.label == 'Toxic' %}
+                                <span class="badge badge-toxic rounded-pill px-3">Toxic</span>
+                            {% else %}
+                                <span class="badge badge-safe rounded-pill px-3">Non-Toxic</span>
+                            {% endif %}
+                        </td>
+                        <td>{{ row.score }}</td>
+                    </tr>
+                    {% else %}
+                    <tr>
+                        <td colspan="3" class="text-center text-muted py-4">Tidak ada data atau format file salah.</td>
+                    </tr>
+                    {% endfor %}
+                </tbody>
+            </table>
+        </div>
+    </div>
+</body>
+</html>