Update app.py

app.py

@@ -50,7 +50,7 @@ def generate(prompt):
     return break_until_dot(decoded)
 
 def generate_image(prompt, inference_steps):
-    prompt = prompt + ', masterpiece charcoal pencil art lord of the rings illustration'
+    prompt = prompt + ' masterpiece charcoal pencil art lord of the rings illustration'
     img = pipe(prompt, height=512, width=512, num_inference_steps=inference_steps)
     return img.images[0]
 
@@ -62,16 +62,16 @@ def generate_story(prompt):
 
 with gr.Blocks() as demo:
     title = gr.Markdown('## Lord of the rings app')
-    description = gr.Markdown('### A Lord of the rings insired app that combines text and image generation')
+    description = gr.Markdown(f'#### A Lord of the rings insired app that combines text and image generation. The language modeling is done by fine tuning distilgpt2 on the LOTR trilogy. The text2img model is {model_id}. The summarization is done using distilbart.')
     prompt = gr.Textbox(label="Your prompt", value="Frodo took the sword and")
     story = gr.Textbox(label="Your story")
     summary = gr.Textbox(label="Summary")
     
     bt_make_text = gr.Button("Generate text")
-    bt_make_image = gr.Button("Generate an image (takes about 10-15 minutes on CPU)", visible=False)
+    bt_make_image = gr.Button(f"Generate an image (takes about 10-15 minutes on CPU).", visible=False)
     
-    image = gr.Image(label='Illustration for your story')
-    inference_steps = gr.Slider(5, 30, value=15, step=1, label="Num inference steps (more steps makes a better image but takes more time)")
+    image = gr.Image(label='Illustration for your story', shape=(512, 512))
+    inference_steps = gr.Slider(5, 30, value=15, step=1, label=f"Num inference steps (more steps makes a better image but takes more time)")
     
     bt_make_text.click(fn=generate_story, inputs=prompt, outputs=[story, summary, bt_make_image])
     bt_make_image.click(fn=generate_image, inputs=[summary, inference_steps], outputs=image)