add the fine tuned BERT model with flask app integrated with Fast API

Dockerfile

@@ -0,0 +1,19 @@
+FROM python:3.10-slim
+
+WORKDIR /code
+
+# Install system dependencies
+RUN apt-get update && apt-get install -y git && rm -rf /var/lib/apt/lists/*
+
+# Copy requirements and install
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy the rest of the code
+COPY . .
+
+# Expose FastAPI port
+EXPOSE 8000
+
+# Hugging Face Spaces expects the app to run on 0.0.0.0:8000
+CMD ["uvicorn", "intent-classifier-chatbot.model.api.api:app", "--host", "0.0.0.0", "--port", "8000"]

README.md

@@ -1,12 +1,56 @@
+# 🤖 Bert Intent Chatbot (Encoder-Only)
+
+**Intent Detection API using BERT and Flask**  
+This project uses a fine-tuned BERT model that is an encoder only model without a decoder used to detect user intent from text inputs (e.g., chatbot queries). It provides a lightweight Flask API to classify input sentences into predefined intent categories.
+
+---
+
+## 🔧 Features
+
+- ✅ Pretrained BERT fine-tuned on [CLINC150](https://huggingface.co/datasets/clinc_oos)
+- 🧠 Real-time intent classification from natural text
+- 🌐 REST API using Flask
+- 🤖 Easy to integrate with chatbots, voice assistants, or NLP systems
+
 ---
-title: Intent Classifier Chatbot
-emoji: 🐨
-colorFrom: green
-colorTo: purple
-sdk: docker
-pinned: false
-license: apache-2.0
-short_description: Intent Detection API using BERT and Flask
+## 📊 Dataset: CLINC150
+
+The project uses the **CLINC150 dataset**, a benchmark dataset for intent classification in task-oriented dialogue systems.
+
+### 🧾 Overview
+
+- **Total intents**: 150 unique user intents
+- **Domains**: 10 real-world domains (e.g., banking, travel, weather, small talk)
+- **Examples**: ~22,500 utterances
+- **Language**: English
+- **Out-of-scope (OOS)**: Includes OOS examples to test robustness
+
+### 📁 Dataset Splits
+
+| Split       | Examples |
+|-------------|----------|
+| Train       | 7,000   |
+| Validation  | 3,000    |
+| Test        | 5,500    |
+
+### 📦 Source
+
+- Official repo: [clinc/oos-eval](https://github.com/clinc/oos-eval)
+- Hugging Face: [`clinc_oos`](https://huggingface.co/datasets/clinc_oos)
+
+
 ---
+## 🚀 Example
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+### Request
+```bash
+{
+  "text": "I want to book a flight"
+}
+```
+### Response
+```bash
+{
+  "intent": "book_flight"
+}
+```

intent_classifier_model/config.json

@@ -0,0 +1,332 @@
+{
+  "architectures": [
+    "BertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2",
+    "3": "LABEL_3",
+    "4": "LABEL_4",
+    "5": "LABEL_5",
+    "6": "LABEL_6",
+    "7": "LABEL_7",
+    "8": "LABEL_8",
+    "9": "LABEL_9",
+    "10": "LABEL_10",
+    "11": "LABEL_11",
+    "12": "LABEL_12",
+    "13": "LABEL_13",
+    "14": "LABEL_14",
+    "15": "LABEL_15",
+    "16": "LABEL_16",
+    "17": "LABEL_17",
+    "18": "LABEL_18",
+    "19": "LABEL_19",
+    "20": "LABEL_20",
+    "21": "LABEL_21",
+    "22": "LABEL_22",
+    "23": "LABEL_23",
+    "24": "LABEL_24",
+    "25": "LABEL_25",
+    "26": "LABEL_26",
+    "27": "LABEL_27",
+    "28": "LABEL_28",
+    "29": "LABEL_29",
+    "30": "LABEL_30",
+    "31": "LABEL_31",
+    "32": "LABEL_32",
+    "33": "LABEL_33",
+    "34": "LABEL_34",
+    "35": "LABEL_35",
+    "36": "LABEL_36",
+    "37": "LABEL_37",
+    "38": "LABEL_38",
+    "39": "LABEL_39",
+    "40": "LABEL_40",
+    "41": "LABEL_41",
+    "42": "LABEL_42",
+    "43": "LABEL_43",
+    "44": "LABEL_44",
+    "45": "LABEL_45",
+    "46": "LABEL_46",
+    "47": "LABEL_47",
+    "48": "LABEL_48",
+    "49": "LABEL_49",
+    "50": "LABEL_50",
+    "51": "LABEL_51",
+    "52": "LABEL_52",
+    "53": "LABEL_53",
+    "54": "LABEL_54",
+    "55": "LABEL_55",
+    "56": "LABEL_56",
+    "57": "LABEL_57",
+    "58": "LABEL_58",
+    "59": "LABEL_59",
+    "60": "LABEL_60",
+    "61": "LABEL_61",
+    "62": "LABEL_62",
+    "63": "LABEL_63",
+    "64": "LABEL_64",
+    "65": "LABEL_65",
+    "66": "LABEL_66",
+    "67": "LABEL_67",
+    "68": "LABEL_68",
+    "69": "LABEL_69",
+    "70": "LABEL_70",
+    "71": "LABEL_71",
+    "72": "LABEL_72",
+    "73": "LABEL_73",
+    "74": "LABEL_74",
+    "75": "LABEL_75",
+    "76": "LABEL_76",
+    "77": "LABEL_77",
+    "78": "LABEL_78",
+    "79": "LABEL_79",
+    "80": "LABEL_80",
+    "81": "LABEL_81",
+    "82": "LABEL_82",
+    "83": "LABEL_83",
+    "84": "LABEL_84",
+    "85": "LABEL_85",
+    "86": "LABEL_86",
+    "87": "LABEL_87",
+    "88": "LABEL_88",
+    "89": "LABEL_89",
+    "90": "LABEL_90",
+    "91": "LABEL_91",
+    "92": "LABEL_92",
+    "93": "LABEL_93",
+    "94": "LABEL_94",
+    "95": "LABEL_95",
+    "96": "LABEL_96",
+    "97": "LABEL_97",
+    "98": "LABEL_98",
+    "99": "LABEL_99",
+    "100": "LABEL_100",
+    "101": "LABEL_101",
+    "102": "LABEL_102",
+    "103": "LABEL_103",
+    "104": "LABEL_104",
+    "105": "LABEL_105",
+    "106": "LABEL_106",
+    "107": "LABEL_107",
+    "108": "LABEL_108",
+    "109": "LABEL_109",
+    "110": "LABEL_110",
+    "111": "LABEL_111",
+    "112": "LABEL_112",
+    "113": "LABEL_113",
+    "114": "LABEL_114",
+    "115": "LABEL_115",
+    "116": "LABEL_116",
+    "117": "LABEL_117",
+    "118": "LABEL_118",
+    "119": "LABEL_119",
+    "120": "LABEL_120",
+    "121": "LABEL_121",
+    "122": "LABEL_122",
+    "123": "LABEL_123",
+    "124": "LABEL_124",
+    "125": "LABEL_125",
+    "126": "LABEL_126",
+    "127": "LABEL_127",
+    "128": "LABEL_128",
+    "129": "LABEL_129",
+    "130": "LABEL_130",
+    "131": "LABEL_131",
+    "132": "LABEL_132",
+    "133": "LABEL_133",
+    "134": "LABEL_134",
+    "135": "LABEL_135",
+    "136": "LABEL_136",
+    "137": "LABEL_137",
+    "138": "LABEL_138",
+    "139": "LABEL_139",
+    "140": "LABEL_140",
+    "141": "LABEL_141",
+    "142": "LABEL_142",
+    "143": "LABEL_143",
+    "144": "LABEL_144",
+    "145": "LABEL_145",
+    "146": "LABEL_146",
+    "147": "LABEL_147",
+    "148": "LABEL_148",
+    "149": "LABEL_149",
+    "150": "LABEL_150"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_10": 10,
+    "LABEL_100": 100,
+    "LABEL_101": 101,
+    "LABEL_102": 102,
+    "LABEL_103": 103,
+    "LABEL_104": 104,
+    "LABEL_105": 105,
+    "LABEL_106": 106,
+    "LABEL_107": 107,
+    "LABEL_108": 108,
+    "LABEL_109": 109,
+    "LABEL_11": 11,
+    "LABEL_110": 110,
+    "LABEL_111": 111,
+    "LABEL_112": 112,
+    "LABEL_113": 113,
+    "LABEL_114": 114,
+    "LABEL_115": 115,
+    "LABEL_116": 116,
+    "LABEL_117": 117,
+    "LABEL_118": 118,
+    "LABEL_119": 119,
+    "LABEL_12": 12,
+    "LABEL_120": 120,
+    "LABEL_121": 121,
+    "LABEL_122": 122,
+    "LABEL_123": 123,
+    "LABEL_124": 124,
+    "LABEL_125": 125,
+    "LABEL_126": 126,
+    "LABEL_127": 127,
+    "LABEL_128": 128,
+    "LABEL_129": 129,
+    "LABEL_13": 13,
+    "LABEL_130": 130,
+    "LABEL_131": 131,
+    "LABEL_132": 132,
+    "LABEL_133": 133,
+    "LABEL_134": 134,
+    "LABEL_135": 135,
+    "LABEL_136": 136,
+    "LABEL_137": 137,
+    "LABEL_138": 138,
+    "LABEL_139": 139,
+    "LABEL_14": 14,
+    "LABEL_140": 140,
+    "LABEL_141": 141,
+    "LABEL_142": 142,
+    "LABEL_143": 143,
+    "LABEL_144": 144,
+    "LABEL_145": 145,
+    "LABEL_146": 146,
+    "LABEL_147": 147,
+    "LABEL_148": 148,
+    "LABEL_149": 149,
+    "LABEL_15": 15,
+    "LABEL_150": 150,
+    "LABEL_16": 16,
+    "LABEL_17": 17,
+    "LABEL_18": 18,
+    "LABEL_19": 19,
+    "LABEL_2": 2,
+    "LABEL_20": 20,
+    "LABEL_21": 21,
+    "LABEL_22": 22,
+    "LABEL_23": 23,
+    "LABEL_24": 24,
+    "LABEL_25": 25,
+    "LABEL_26": 26,
+    "LABEL_27": 27,
+    "LABEL_28": 28,
+    "LABEL_29": 29,
+    "LABEL_3": 3,
+    "LABEL_30": 30,
+    "LABEL_31": 31,
+    "LABEL_32": 32,
+    "LABEL_33": 33,
+    "LABEL_34": 34,
+    "LABEL_35": 35,
+    "LABEL_36": 36,
+    "LABEL_37": 37,
+    "LABEL_38": 38,
+    "LABEL_39": 39,
+    "LABEL_4": 4,
+    "LABEL_40": 40,
+    "LABEL_41": 41,
+    "LABEL_42": 42,
+    "LABEL_43": 43,
+    "LABEL_44": 44,
+    "LABEL_45": 45,
+    "LABEL_46": 46,
+    "LABEL_47": 47,
+    "LABEL_48": 48,
+    "LABEL_49": 49,
+    "LABEL_5": 5,
+    "LABEL_50": 50,
+    "LABEL_51": 51,
+    "LABEL_52": 52,
+    "LABEL_53": 53,
+    "LABEL_54": 54,
+    "LABEL_55": 55,
+    "LABEL_56": 56,
+    "LABEL_57": 57,
+    "LABEL_58": 58,
+    "LABEL_59": 59,
+    "LABEL_6": 6,
+    "LABEL_60": 60,
+    "LABEL_61": 61,
+    "LABEL_62": 62,
+    "LABEL_63": 63,
+    "LABEL_64": 64,
+    "LABEL_65": 65,
+    "LABEL_66": 66,
+    "LABEL_67": 67,
+    "LABEL_68": 68,
+    "LABEL_69": 69,
+    "LABEL_7": 7,
+    "LABEL_70": 70,
+    "LABEL_71": 71,
+    "LABEL_72": 72,
+    "LABEL_73": 73,
+    "LABEL_74": 74,
+    "LABEL_75": 75,
+    "LABEL_76": 76,
+    "LABEL_77": 77,
+    "LABEL_78": 78,
+    "LABEL_79": 79,
+    "LABEL_8": 8,
+    "LABEL_80": 80,
+    "LABEL_81": 81,
+    "LABEL_82": 82,
+    "LABEL_83": 83,
+    "LABEL_84": 84,
+    "LABEL_85": 85,
+    "LABEL_86": 86,
+    "LABEL_87": 87,
+    "LABEL_88": 88,
+    "LABEL_89": 89,
+    "LABEL_9": 9,
+    "LABEL_90": 90,
+    "LABEL_91": 91,
+    "LABEL_92": 92,
+    "LABEL_93": 93,
+    "LABEL_94": 94,
+    "LABEL_95": 95,
+    "LABEL_96": 96,
+    "LABEL_97": 97,
+    "LABEL_98": 98,
+    "LABEL_99": 99
+  },
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.53.0",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

intent_classifier_model/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43ec753e8b145ec7bf1dc275cc62677fd08e5bf706f5cc957e66c8f78d234453
+size 438416972

intent_classifier_tokenizer/special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

intent_classifier_tokenizer/tokenizer_config.json

@@ -0,0 +1,58 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_basic_tokenize": true,
+  "do_lower_case": true,
+  "extra_special_tokens": {},
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "never_split": null,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

model/api/api.py

@@ -0,0 +1,45 @@
+from fastapi import FastAPI
+from pydantic import BaseModel
+from transformers import BertForSequenceClassification, BertTokenizer
+import torch
+import os
+
+app = FastAPI()
+
+# Get the absolute path to the model directory
+BASE_DIR = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+# go back one level to get the correct path
+BASE_DIR = os.path.dirname(BASE_DIR)
+MODEL_DIR = os.path.join(BASE_DIR, "intent_classifier_model")
+TOKENIZER_DIR = os.path.join(BASE_DIR, "intent_classifier_tokenizer")
+
+# Ensure model and tokenizer directories exist
+if not os.path.isdir(MODEL_DIR):
+    raise FileNotFoundError(f"Model directory not found: {MODEL_DIR}")
+if not os.path.isdir(TOKENIZER_DIR):
+    raise FileNotFoundError(f"Tokenizer directory not found: {TOKENIZER_DIR}")
+
+# Load model and tokenizer from local directories only
+model = BertForSequenceClassification.from_pretrained(MODEL_DIR, local_files_only=True)
+tokenizer = BertTokenizer.from_pretrained(TOKENIZER_DIR, local_files_only=True)
+
+# Load intent label mapping
+from datasets import load_dataset
+dataset = load_dataset("clinc_oos", "small")
+int2str = dataset["train"].features["intent"].int2str
+
+class Query(BaseModel):
+    text: str
+
+@app.post("/predict")
+def predict_intent(query: Query):
+    inputs = tokenizer(query.text, return_tensors="pt", truncation=True, padding=True, max_length=128)
+    with torch.no_grad():
+        outputs = model(**inputs)
+        prediction = outputs.logits.argmax(dim=-1).item()
+        intent = int2str(prediction)
+    if intent == "oos":
+        return {"intent": "out of scope (OOS)"}
+    else:
+        intent = intent.replace("_", " ").title()
+        return {"intent": intent}

model/api/start_server.py

@@ -0,0 +1,4 @@
+import uvicorn
+
+if __name__ == "__main__":
+    uvicorn.run("api:app", host="0.0.0.0", port=8000, reload=True)

model/api/test.py

@@ -0,0 +1,8 @@
+import requests
+
+url = "http://localhost:8000/predict"
+data = {"text": "I want to set an alarm for 7 AM tomorrow."}
+
+response = requests.post(url, json=data)
+print("Status code:", response.status_code)
+print("Response:", response.json())

requirements.txt

@@ -0,0 +1,7 @@
+fastapi
+uvicorn
+transformers
+torch
+datasets
+requests
+chardet

src/__init__.py

@@ -0,0 +1 @@
+# ...empty file...

src/app.py

@@ -0,0 +1,34 @@
+# NOTE: Make sure the FastAPI server is running at http://localhost:8000 before starting this Flask app.
+from flask import Flask, render_template, request
+import requests
+
+app = Flask(__name__)
+
+FASTAPI_URL = "http://localhost:8000/predict"
+
+@app.route("/", methods=["GET", "POST"])
+def index():
+    prediction = None
+    user_text = ""
+    if request.method == "POST":
+        user_text = request.form.get("user_text", "")
+        if user_text:
+            try:
+                response = requests.post(FASTAPI_URL, json={"text": user_text})
+                if response.status_code == 200:
+                    prediction = response.json().get("intent", "Unknown")
+                else:
+                    prediction = "Error: Unable to get prediction."
+            except requests.exceptions.ConnectionError:
+                prediction = (
+                    "Error: Could not connect to the FastAPI server at http://localhost:8000.<br>"
+                    "Please make sure the FastAPI server is running.<br>"
+                    "To start it, run:<br>"
+                    "<code>python model/api/start_server.py</code> from the project root."
+                )
+            except Exception as e:
+                prediction = f"Error: {str(e)}"
+    return render_template("index.html", prediction=prediction, user_text=user_text)
+
+if __name__ == "__main__":
+    app.run(debug=True)

src/fastapi_server.py

@@ -0,0 +1,17 @@
+from fastapi import FastAPI, Request
+from pydantic import BaseModel
+
+app = FastAPI()
+
+class PredictRequest(BaseModel):
+    text: str
+
+@app.post("/predict")
+def predict(req: PredictRequest):
+    # Dummy implementation for testing
+    # Replace with your model inference logic
+    if "alarm" in req.text.lower():
+        intent = "set_alarm"
+    else:
+        intent = "unknown"
+    return {"intent": intent}

src/templates/index.html

@@ -0,0 +1,115 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="UTF-8">
+    <title>Intent Classifier Chatbot</title>
+    <meta name="viewport" content="width=device-width, initial-scale=1">
+    <style>
+        body { 
+            font-family: 'Segoe UI', Arial, sans-serif; 
+            margin: 0; 
+            background: #f7f9fa; 
+            color: #222; 
+        }
+        .container { 
+            max-width: 500px; 
+            margin: 60px auto 30px auto; 
+            background: #fff; 
+            border-radius: 12px; 
+            box-shadow: 0 4px 24px rgba(0,0,0,0.08); 
+            padding: 32px 28px 24px 28px;
+        }
+        h1 {
+            text-align: center;
+            color: #2d6cdf;
+            margin-bottom: 18px;
+        }
+        label {
+            font-weight: 500;
+            margin-bottom: 8px;
+            display: block;
+        }
+        input[type="text"] {
+            width: 100%;
+            padding: 12px;
+            border: 1px solid #d2d6dc;
+            border-radius: 6px;
+            font-size: 1em;
+            margin-bottom: 18px;
+            box-sizing: border-box;
+            transition: border 0.2s;
+        }
+        input[type="text"]:focus {
+            border: 1.5px solid #2d6cdf;
+            outline: none;
+        }
+        button {
+            width: 100%;
+            padding: 12px;
+            background: linear-gradient(90deg, #2d6cdf 60%, #4e9cff 100%);
+            color: #fff;
+            border: none;
+            border-radius: 6px;
+            font-size: 1.1em;
+            font-weight: 600;
+            cursor: pointer;
+            transition: background 0.2s;
+        }
+        button:hover {
+            background: linear-gradient(90deg, #1b4e9b 60%, #3578c7 100%);
+        }
+        .result {
+            margin-top: 24px;
+            font-size: 1.15em;
+            background: #eaf3ff;
+            border-left: 4px solid #2d6cdf;
+            padding: 14px 18px;
+            border-radius: 6px;
+            color: #1a3a5d;
+            word-break: break-word;
+        }
+        .info {
+            margin-top: 18px;
+            font-size: 0.98em;
+            color: #555;
+            background: #f3f6fa;
+            border-radius: 6px;
+            padding: 10px 14px;
+        }
+        footer {
+            margin-top: 40px;
+            text-align: center;
+            color: #888;
+            font-size: 0.97em;
+            padding-bottom: 18px;
+        }
+        @media (max-width: 600px) {
+            .container { padding: 18px 6px 18px 6px; }
+        }
+    </style>
+</head>
+<body>
+    <div class="container">
+        <h1>Intent Classifier Chatbot</h1>
+        <div class="info">
+            Enter a message below and click <b>Predict Intent</b> to see what the AI thinks your intent is.<br>
+            <span style="color:#2d6cdf;">Try: <i>"Set an alarm for 7am"</i> or <i>"Transfer money to John"</i></span>
+        </div>
+        <form method="post" autocomplete="off">
+            <label for="user_text">Your Message:</label>
+            <input type="text" id="user_text" name="user_text" value="{{ user_text }}" placeholder="Type your message here..." required autofocus>
+            <button type="submit">Predict Intent</button>
+        </form>
+        {% if prediction %}
+            <div class="result">
+                <strong>Predicted Intent:</strong> {{ prediction }}
+            </div>
+        {% endif %}
+    </div>
+    <footer>
+        Made by <b>Saher Muhamed</b><br>
+        <a href="https://github.com/sahermuhamed1" target="_blank" style="color:#2d6cdf;text-decoration:none;">GitHub</a> &middot; 
+        <a href="mailto:sahermuhamed176@gmail.com" style="color:#2d6cdf;text-decoration:none;">Contact</a>
+    </footer>
+</body>
+</html>