Create app.py

app.py

@@ -0,0 +1,45 @@
+import gradio as gr
+import random
+from PIL import Image
+
+from funcs import detect_emotions, cosine_distance, generate_triggers_img, session_processor
+
+sp = session_processor
+
+img_paths = ['2.jpg','3.jpeg','4.jpeg','5.jpeg','6.jpeg']
+img_path = random.choice(img_paths)
+img_1 = Image.open(img_path)
+
+img_path = '7.jpg'
+img_2 = Image.open(img_path)
+
+with gr.Blocks() as demo:
+    gr.Markdown("""# Falcon Cognitive Behavioural Therapy Assistant
+    - Your Personal AI Therapist. Start chatting...""")
+    therapy_session_conversation = gr.State([])
+
+    with gr.Row():
+
+        with gr.Column(scale=1):
+            image = gr.Image(type='pil', label ='Your Personal Therapy Assistant', value=img_1, interactive=False)
+            emotions = gr.Image(value=img_2, label='Top 5 Emotion Triggers')
+            summary_notes = gr.Textbox(label="Summary Notes of the Session", visible=False)
+
+        with gr.Column(scale=2):
+            chatbox = gr.Chatbot(label="Therapy Session Conversation",value =[[None, 'Therapist: Hello, What can I do for you?']], height=300)
+            user_input = gr.Textbox(placeholder="Enter your message here...", label="User")
+            submit_button = gr.Button("Submit")
+            submit_button.click(sp.get_doc_response_emotions, [user_input, chatbox], [user_input, chatbox, emotions])
+            user_input.submit(get_doc_response_emotions, [user_input, chatbox], [user_input, chatbox, emotions])
+            recommendations = gr.Textbox(label="Recommended Actions", visible = False)
+
+            def summarize_and_recommend_process():
+              sn, r = summarize_and_recommend()
+              return gr.update(visible=True, value=sn), gr.update(visible=True, value=r)
+              
+    process_button = gr.Button("Generate Session Notes & Recommendations")
+    clear = gr.ClearButton(components=[user_input, chatbox, emotions, summary_notes, recommendations], value="Clear console")
+    process_button.click(sp.summarize_and_recommend_process, inputs=None, outputs=[summary_notes, recommendations])
+
+
+demo.launch(debug=True, share=True)