First upload to the repository of NLP_Chatbot.

Dockerfile

@@ -0,0 +1,18 @@
+FROM python:3.12 AS base
+
+RUN useradd -m -u 1000 user
+USER user
+ENV HOME=/home/user \
+	PATH=/home/user/.local/bin:$PATH
+
+WORKDIR $HOME/app
+
+COPY --chown=user ./requirements.txt ./requirements.txt
+
+RUN pip3 install -r requirements.txt
+
+COPY --chown=user . $HOME/app
+
+FROM base AS run
+
+CMD ["python", "app.py"]

app.py

@@ -0,0 +1,19 @@
+from chatbot import ChatBot
+import gradio as gr
+import time
+
+def main():
+    chatbot = ChatBot()
+
+    with gr.Blocks() as demo:
+        chatbot_gr = gr.Chatbot(height = 350)
+        textbox = gr.Textbox(label="Introduce your prompt here")
+        start_btn = gr.Button("Submit")
+        stop_btn = gr.Button("Stop")
+
+        click_event = start_btn.click(chatbot.answer, [textbox, chatbot_gr], [chatbot_gr],  api_name=False)
+        stop_btn.click(fn=None, inputs=None, outputs=None, cancels=[click_event])
+
+        demo.launch(server_name="0.0.0.0", server_port=7860)
+
+main()

chatbot.py

@@ -0,0 +1,34 @@
+from transformers import AutoTokenizer, AutoModelForCausalLM, TextIteratorStreamer
+from threading import Thread
+
+class ChatBot():
+
+    def __init__(self, device='cpu'):
+        self.tokenizer = AutoTokenizer.from_pretrained("HuggingFaceTB/SmolLM2-1.7B")
+        self.model = AutoModelForCausalLM.from_pretrained("HuggingFaceTB/SmolLM2-1.7B")
+        self.device = device
+        self.model.to(device)
+
+    def answer(self, message, history):
+        if (len(message.strip())>0):
+            input_text = message
+            input_ids = self.tokenizer(input_text, return_tensors="pt")
+            input_ids = input_ids.to(self.device)
+            streamer = TextIteratorStreamer(self.tokenizer)
+            generation_kwargs = dict(input_ids, streamer=streamer, max_new_tokens=1500)
+            
+            thread = Thread(target=self.model.generate, kwargs=generation_kwargs)
+            thread.start()
+
+            if (len(history)==0):
+                history = []
+            
+            history.append([message, ""])
+            for new_text_aux in streamer:
+                history[-1][1]+=new_text_aux
+                
+                yield history
+
+            print('Waiting for the thread to finish')
+            thread.join()
+            print('The thread has finished')

requirements.txt

@@ -0,0 +1,4 @@
+gradio==5.23.3
+numpy==1.26.4
+transformers==4.50.3
+torch==2.2.1