initial commit

Dockerfile

@@ -0,0 +1,52 @@
+# Use Python 3.8 slim
+FROM python:3.8-slim
+
+# 1) System deps
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends git build-essential && \
+    rm -rf /var/lib/apt/lists/*
+
+WORKDIR /app
+
+# 2) Install Rasa 2.8.3 from source (patch typing-extensions)
+RUN git clone --depth 1 --branch 2.8.3 https://github.com/RasaHQ/rasa.git /rasa-src && \
+    cd /rasa-src && \
+    sed -i "s/typing-extensions<4.0.0 and >=3.7.4/typing-extensions<5.0.0 and >=3.7.4/" setup.cfg && \
+    pip install --no-cache-dir .
+
+# 3) Install Rasa SDK so your custom actions work
+RUN pip install --no-cache-dir rasa-sdk==2.8.3
+
+# 4) Pin protobuf so TensorFlow protos load
+RUN pip install --no-cache-dir "protobuf<3.21.0,>=3.20.0"
+
+# 5) Pin TensorFlow to the 2.8.x line Rasa 2.8.3 expects
+RUN pip install --no-cache-dir "tensorflow<2.9.0,>=2.8.0"
+
+# 6) Pin NumPy to a 1.20–1.23 release for C-API compatibility & deprecated aliases
+RUN pip install --no-cache-dir "numpy>=1.20.0,<1.24.0"
+
+# 7) Copy in your Rasa project
+COPY domain.yml config.yml endpoints.yml /app/
+COPY data /app/data
+COPY actions /app/actions
+
+# 8) Patch out the obsolete decorator before training
+RUN sed -i '/@training.enable_multi_worker/d' \
+    /usr/local/lib/python3.8/site-packages/rasa/utils/tensorflow/temp_keras_modules.py
+
+# 9) Train your model (writes into /app/models)
+RUN rasa train
+
+# 10) Install Gradio v4 + HTTP client
+COPY requirements.txt /app/requirements.txt
+RUN pip install --no-cache-dir -r requirements.txt
+
+# 11) Copy your Gradio wrapper
+COPY app.py /app/app.py
+
+# 12) Expose ports (actions, Rasa API, Gradio UI)
+EXPOSE 5055 5005 7860
+
+# 13) Launch everything: action server, Rasa REST API, then Gradio
+ENTRYPOINT ["bash","-lc","rasa run actions & rasa run --enable-api --port 5005 & python app.py"]

README.md

@@ -10,4 +10,31 @@ pinned: false
 license: mit
 ---
 
-An example chatbot using [Gradio](https://gradio.app), [`huggingface_hub`](https://huggingface.co/docs/huggingface_hub/v0.22.2/en/index), and the [Hugging Face Inference API](https://huggingface.co/docs/api-inference/index).
+# RasaBot + Gradio v4 Demo on HuggingFace Spaces (CPU)
+
+This Space runs a tiny Rasa 2.8.3 assistant inside a **Gradio v4** chat UI on a **CPU** plan.
+
+## Files
+
+- **Dockerfile**  
+  Python 3.8 → installs Rasa 2.8.3 + Gradio v4 → `rasa init` → `rasa train` → launches `app.py`.
+
+- **requirements.txt**  
+  Pins Rasa and Gradio to v4-compatible versions.
+
+- **app.py**  
+  Loads the Rasa model and serves a chat UI via Gradio Blocks.
+
+- **README.md** (this file)
+
+## Deploy on HuggingFace Spaces
+
+1. Create a new Space with **Hardware: CPU** and **Environment: Docker**.  
+2. Push these four files to your repo.  
+3. The Space will build, train, and expose the Gradio chat at `/`.
+
+## Local Testing
+
+```bash
+docker build -t rasa-gradio-v4 .
+docker run -it --rm -p 7860:7860 rasa-gradio-v4

actions/actions.py

@@ -0,0 +1,53 @@
+# This files contains your custom actions which can be used to run
+# custom Python code.
+#
+# See this guide on how to implement these action:
+# https://rasa.com/docs/rasa/custom-actions
+
+
+# This is a simple example for a custom action which utters "Hello World!"
+
+from typing import Any, Text, Dict, List
+
+from rasa_sdk import Action, Tracker
+from rasa_sdk.executor import CollectingDispatcher
+
+
+class action_ReversePhoneLookup(Action):
+
+    def name(self) -> Text:
+        return "action_ReversePhoneLookup"
+
+    def run(self, dispatcher: CollectingDispatcher,
+            tracker: Tracker,
+            domain: Dict[Text, Any]) -> List[Dict[Text, Any]]:
+
+        dispatcher.utter_message(text="反查來電處理")
+
+        return []
+    
+class action_CallReservationHotline(Action):
+
+    def name(self) -> Text:
+        return "action_CallReservationHotline"
+
+    def run(self, dispatcher: CollectingDispatcher,
+            tracker: Tracker,
+            domain: Dict[Text, Any]) -> List[Dict[Text, Any]]:
+
+        dispatcher.utter_message(text="預約或修改掛號處理")
+
+        return []
+    
+class action_human_handoff(Action):
+
+    def name(self) -> Text:
+        return "action_human_handoff"
+
+    def run(self, dispatcher: CollectingDispatcher,
+            tracker: Tracker,
+            domain: Dict[Text, Any]) -> List[Dict[Text, Any]]:
+
+        dispatcher.utter_message(text="其他處理")
+
+        return []

app.py

@@ -1,64 +1,22 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+import requests
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+RASA_URL = "http://localhost:5005/webhooks/rest/webhook"
 
+def bot_response(user_message, chat_history):
+    resp = requests.post(RASA_URL, json={
+        "sender": "user",
+        "message": user_message
+    }).json()
+    bot_reply = resp[0].get("text") if resp else "🤖 (no reply)"
+    chat_history.append((user_message, bot_reply))
+    return chat_history
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-
-        response += token
-        yield response
-
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
-
+with gr.Blocks(title="Rasa 2.8.3 + Gradio v4 Demo") as demo:
+    chatbot = gr.Chatbot(label="Rasa Bot", height=400)
+    user_input = gr.Textbox(placeholder="Type here...", show_label=False)
+    user_input.submit(bot_response, inputs=[user_input, chatbot], outputs=chatbot)
+    gr.Button("Clear").click(lambda: [], None, chatbot)
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch(server_name="0.0.0.0", server_port=7860, share=False)

config.yml

@@ -0,0 +1,14 @@
+# Configuration for Rasa NLU.
+# https://rasa.com/docs/rasa/nlu/components/
+language: zh
+
+pipeline:
+# # No configuration for the NLU pipeline was provided. The following default pipeline was used to train your model.
+# # If you'd like to customize it, uncomment and adjust the pipeline.
+# # See https://rasa.com/docs/rasa/tuning-your-model for more information.
+- name: "KeywordIntentClassifier"
+
+# Configuration for Rasa Core.
+# https://rasa.com/docs/rasa/core/policies/
+policies:
+- name: RulePolicy

data/nlu.yml

@@ -0,0 +1,17 @@
+version: "2.0"
+
+nlu:
+- intent: ReversePhoneLookup
+  examples: |
+    - 我剛才有未接來電，想知道是誰打來的
+    - 我看到有未接來電，請問找我有什麼事
+
+- intent: CallReservationHotline
+  examples: |
+    - 我想要預約心臟科的門診
+    - 我想預約健康檢查
+
+- intent: out_of_scope
+  examples: |
+    - 請問醫院的營業時間是什麼時候
+    - 我想查詢我的檢查報告結果

data/rules.yml

@@ -0,0 +1,18 @@
+version: "2.0"
+
+rules:
+
+- rule: reverse phone lookup
+  steps:
+  - intent: ReversePhoneLookup
+  - action: action_ReversePhoneLookup
+
+- rule: book or modify an appointment
+  steps:
+  - intent: CallReservationHotline
+  - action: action_CallReservationHotline
+
+- rule: out of scope handling
+  steps:
+  - intent: out_of_scope
+  - action: action_human_handoff

domain.yml

@@ -0,0 +1,16 @@
+version: "2.0"
+
+intents:
+  - ReversePhoneLookup
+  - CallReservationHotline
+  - out_of_scope
+
+actions:
+  - action_ReversePhoneLookup
+  - action_CallReservationHotline
+  - action_human_handoff
+   
+
+session_config:
+  session_expiration_time: 60
+  carry_over_slots_to_new_session: true

endpoints.yml

@@ -0,0 +1,14 @@
+# This file contains the different endpoints your bot can use.
+
+# Server where the models are pulled from.
+# https://rasa.com/docs/rasa/model-storage#fetching-models-from-a-server
+
+#models:
+#  url: http://my-server.com/models/default_core@latest
+#  wait_time_between_pulls:  10   # [optional](default: 100)
+
+# Server which runs your custom actions.
+# https://rasa.com/docs/rasa/custom-actions
+
+action_endpoint:
+ url: "http://localhost:5055/webhook"

requirements.txt

@@ -1 +1,2 @@
-huggingface_hub==0.25.2
+gradio==4.44.1
+requests