Update app_dialogue.py

app_dialogue.py

@@ -561,7 +561,7 @@ textbox = gr.Textbox(
 
 css2 = """
 #chatbot {
-background-image: url('https://huggingface.co/spaces/ysharma/m4-dialogue_copy4/resolve/main/idefics_720_2.png');
+background-image: url('https://huggingface.co/spaces/ysharma/dummy_m4/resolve/main/idefics_transparent20.png');
 background-repeat: repeat;}
 """
 
@@ -580,194 +580,49 @@ with gr.Blocks(title="IDEFICS-Chat", theme=gr.themes.Base(), css=css2) as demo:
         """
     )
 
-    with gr.Row():
-        with gr.Column(scale=3):
-            with gr.Row(elem_id="model_selector_row"):
-                model_selector = gr.Dropdown(
-                    choices=MODELS,
-                    value="HuggingFaceM4/idefics-9b-instruct",
-                    interactive=True,
-                    show_label=False,
-                    container=False,
-                    label="Model"
-                )
-            processor, tokenizer, model = load_processor_tokenizer_model(model_selector.value)
-
-            imagebox = gr.Image(type="filepath", label="Image input", visible=False)
-
-            with gr.Accordion("Advanced parameters", open=False, visible=True) as parameter_row:
-                max_new_tokens = gr.Slider(
-                    minimum=0,
-                    maximum=2048,
-                    value=512,
-                    step=1,
-                    interactive=True,
-                    label="Maximum number of new tokens to generate",
-                )
-                min_length = gr.Slider(
-                    minimum=0,
-                    maximum=50,
-                    value=0,
-                    step=1,
-                    interactive=True,
-                    label="Minimum number of new tokens to generate",
-                )
-                repetition_penalty = gr.Slider(
-                    minimum=0.0,
-                    maximum=5.0,
-                    value=1.0,
-                    step=0.1,
-                    interactive=True,
-                    label="Repetition penalty",
-                    info="1.0 means no penalty",
-                )
-                no_repeat_ngram_size = gr.Slider(
-                    minimum=0,
-                    maximum=10,
-                    value=0,
-                    step=1,
-                    interactive=True,
-                    label="N-gram repetition threshold",
-                    info="If set to int > 0, all ngrams of that size can only occur once.",
-                )
-                decoding_strategy = gr.Radio(
-                    [
-                        "Greedy",
-                        # "beam_search",
-                        # "beam_sampling",
-                        # "sampling_top_k",
-                        "Top P Sampling",
-                    ],
-                    value="Top P Sampling",
-                    label="Decoding strategy",
-                    interactive=True,
-                )
-                temperature = gr.Slider(
-                    minimum=0.0,
-                    maximum=5.0,
-                    value=1.2,
-                    step=0.1,
-                    interactive=True,
-                    label="Sampling temperature",
-                )
-                decoding_strategy.change(
-                    fn=lambda selection: gr.Slider.update(
-                        visible=(
-                            selection in ["contrastive_sampling", "beam_sampling", "Top P Sampling", "sampling_top_k"]
-                        )
-                    ),
-                    inputs=decoding_strategy,
-                    outputs=temperature,
-                )
-                num_beams = gr.Slider(
-                    minimum=0,
-                    maximum=20,
-                    value=3.0,
-                    step=1.0,
-                    interactive=True,
-                    visible=False,
-                    label="Number of beams",
-                    info="Only used if `decoding_strategy` is `beam_search` or `beam_sampling`.",
-                )
-                decoding_strategy.change(
-                    fn=lambda selection: gr.Slider.update(visible=(selection in ["beam_search", "beam_sampling"])),
-                    inputs=decoding_strategy,
-                    outputs=num_beams,
-                )
-                top_p = gr.Slider(
-                    minimum=0.0,
-                    maximum=1.0,
-                    value=0.8,
-                    step=0.01,
-                    interactive=True,
-                    visible=True,
-                    label="Top P",
-                    info=(
-                        "If set to float < 1, only the smallest set of most probable tokens with probabilities that"
-                        " add up to top_p or higher are kept for generation."
-                    ),
-                )
-                decoding_strategy.change(
-                    fn=lambda selection: gr.Slider.update(visible=(selection in ["Top P Sampling"])),
-                    inputs=decoding_strategy,
-                    outputs=top_p,
-                )
-                top_k = gr.Slider(
-                    minimum=0,
-                    maximum=500,
-                    value=50,
-                    step=1,
-                    interactive=True,
-                    visible=False,
-                    label="Top K",
-                    info="The number of highest probability vocabulary tokens to keep for top-k-filtering.",
-                )
-                decoding_strategy.change(
-                    fn=lambda selection: gr.Slider.update(visible=(selection in ["sampling_top_k"])),
-                    inputs=decoding_strategy,
-                    outputs=top_k,
-                )
-                length_penalty = gr.Slider(
-                    minimum=-1000.0,
-                    maximum=1000.0,
-                    value=1.0,
-                    step=0.1,
-                    interactive=True,
-                    visible=False,
-                    label="Length penalty",
-                    info=(
-                        "length_penalty > 0.0 promotes longer sequences, while length_penalty < 0.0 encourages shorter"
-                        " sequences. Only used if `decoding_strategy` is `beam_search` or `beam_sampling`."
-                    ),
-                )
-                decoding_strategy.change(
-                    fn=lambda selection: gr.Slider.update(visible=(selection in ["beam_search", "beam_sampling"])),
-                    inputs=decoding_strategy,
-                    outputs=length_penalty,
-                )
-                penalty_alpha = gr.Slider(
-                    minimum=0.0,
-                    maximum=5.0,
-                    value=0.95,
-                    step=0.05,
-                    interactive=True,
-                    visible=False,
-                    label="Penalty alpha",
-                    info="Only used if `decoding_strategy` is `contrastive_sampling`.",
-                )
-                decoding_strategy.change(
-                    fn=lambda selection: gr.Slider.update(visible=(selection in ["contrastive_sampling"])),
-                    inputs=decoding_strategy,
-                    outputs=penalty_alpha,
-                )
-
-        with gr.Column(scale=6):
+    #with gr.Row():
+    #    with gr.Column(): #(scale=3):
+    with gr.Row(elem_id="model_selector_row"):
+        model_selector = gr.Dropdown(
+            choices=MODELS,
+            value="HuggingFaceM4/idefics-9b-instruct",
+            interactive=True,
+            show_label=False,
+            container=False,
+            label="Model"
+        )
+    processor, tokenizer, model = load_processor_tokenizer_model(model_selector.value)
 
-            def prefetch_images_in_history(user_prompt_str):
-                """
-                Pre-fetch the images that are passed in the chatbot default history.
-                """
-                return user_prompt_list_to_markdown(handle_manual_images_in_user_prompt(user_prompt_str))
+    imagebox = gr.Image(type="filepath", label="Image input", visible=False)
 
-            chatbot = gr.Chatbot(
-                elem_id="chatbot",
-                label="ChatIDEFICS",
-                visible=True,
-                height=550,
-                value=[
-                    [
-                        (
-                            prefetch_images_in_history(
-                                """Write a funny story including the following characters.
+    with gr.Row():
+        #with gr.Column(scale=6):
+
+        def prefetch_images_in_history(user_prompt_str):
+            """
+            Pre-fetch the images that are passed in the chatbot default history.
+            """
+            return user_prompt_list_to_markdown(handle_manual_images_in_user_prompt(user_prompt_str))
+
+        chatbot = gr.Chatbot(
+            elem_id="chatbot",
+            label="ChatIDEFICS",
+            visible=True,
+            height=550,
+            value=[
+                [
+                    (
+                        prefetch_images_in_history(
+                            """Write a funny story including the following characters.
 There should be drama and most importantly romance.
 Character 1:<fake_token_around_image><image:https://m.media-amazon.com/images/M/MV5BODg1OTczMWEtNTU3MS00OTUzLThjODEtNDg1MWQwZmExYmFlXkEyXkFqcGdeQWFybm8@._V1_QL75_UX500_CR0,0,500,281_.jpg><fake_token_around_image>
 Character 2:<fake_token_around_image><image:https://static.wikia.nocookie.net/dreamworks/images/0/09/Shrek_dreamworks.png/revision/latest?cb=20170921225439><fake_token_around_image>
 Character 3:<fake_token_around_image><image:https://upload.wikimedia.org/wikipedia/en/e/ef/Marsupilami_1.jpg><fake_token_around_image>
 The plot should take place in the world pictured here:<fake_token_around_image><image:https://www.astronomy.com/wp-content/uploads/sites/2/2021/09/ON1.jpg><fake_token_around_image>"""
-                            )
-                        ),
-                        (
-                            """Once upon a time, in a world where the sky was a dome and the ground was covered in grass, there lived a beautiful princess named Mulan. She was kind, brave, and always ready to help others. One day, while walking through the forest, she stumbled upon a strange creature. It was a green ogre named Shrek, who was lost and looking for his way back home.
+                        )
+                    ),
+                    (
+                        """Once upon a time, in a world where the sky was a dome and the ground was covered in grass, there lived a beautiful princess named Mulan. She was kind, brave, and always ready to help others. One day, while walking through the forest, she stumbled upon a strange creature. It was a green ogre named Shrek, who was lost and looking for his way back home.
 
 Mulan, being the kind-hearted person she was, decided to help Shrek find his way back. As they traveled together, they began to develop a strong bond. Shrek was impressed by Mulan's bravery and determination, while Mulan admired Shrek's loyalty and sense of humor.
 
@@ -780,27 +635,178 @@ Finally, they reached Shrek's home, and he was reunited with his family and frie
 Mulan was overjoyed, and they shared a passionate kiss. From that day on, they lived happily ever after, exploring the world together and facing any challenges that came their way.
 
 And so, the story of Mulan and Shrek's romance came to an end, leaving a lasting impression on all who heard it."""
-                        ),
-                    ],
+                    ),
                 ],
-            )
-            with gr.Group():
-                with gr.Row():
-                    with gr.Column(scale=7):
-                        textbox.render()
-                    with gr.Column(scale=1, min_width=80):
-                        submit_btn = gr.Button(value="Submit", visible=True)
-                    with gr.Column(scale=1, min_width=10):
-                        clear_btn = gr.ClearButton([textbox, chatbot])
-                    with gr.Column(scale=1, min_width=10):
-                        upload_btn = gr.UploadButton("📁", file_types=["image"])
-            with gr.Group():
-                with gr.Row():
-                    with gr.Column(scale=1, min_width=50):
-                        dope_bttn = gr.Button("Dope🔥")
-                    with gr.Column(scale=1, min_width=50):
-                        problematic_bttn = gr.Button("Problematic😬")
+            ],
+        )
 
+    with gr.Row():    
+        with gr.Group():
+            with gr.Row():
+                with gr.Column(scale=7):
+                    textbox.render()
+                with gr.Column(scale=1, min_width=80):
+                    submit_btn = gr.Button(value="Submit", visible=True)
+                with gr.Column(scale=1, min_width=10):
+                    clear_btn = gr.ClearButton([textbox, chatbot])
+                with gr.Column(scale=1, min_width=10):
+                    upload_btn = gr.UploadButton("📁", file_types=["image"])
+    with gr.Row():
+        with gr.Group():
+            with gr.Row():
+                with gr.Column(scale=1, min_width=50):
+                    dope_bttn = gr.Button("Dope🔥")
+                with gr.Column(scale=1, min_width=50):
+                    problematic_bttn = gr.Button("Problematic😬")
+
+    with gr.Row():
+        with gr.Accordion("Advanced parameters", open=False, visible=True) as parameter_row:
+            max_new_tokens = gr.Slider(
+                minimum=0,
+                maximum=2048,
+                value=512,
+                step=1,
+                interactive=True,
+                label="Maximum number of new tokens to generate",
+            )
+            min_length = gr.Slider(
+                minimum=0,
+                maximum=50,
+                value=0,
+                step=1,
+                interactive=True,
+                label="Minimum number of new tokens to generate",
+            )
+            repetition_penalty = gr.Slider(
+                minimum=0.0,
+                maximum=5.0,
+                value=1.0,
+                step=0.1,
+                interactive=True,
+                label="Repetition penalty",
+                info="1.0 means no penalty",
+            )
+            no_repeat_ngram_size = gr.Slider(
+                minimum=0,
+                maximum=10,
+                value=0,
+                step=1,
+                interactive=True,
+                label="N-gram repetition threshold",
+                info="If set to int > 0, all ngrams of that size can only occur once.",
+            )
+            decoding_strategy = gr.Radio(
+                [
+                    "Greedy",
+                    # "beam_search",
+                    # "beam_sampling",
+                    # "sampling_top_k",
+                    "Top P Sampling",
+                ],
+                value="Top P Sampling",
+                label="Decoding strategy",
+                interactive=True,
+            )
+            temperature = gr.Slider(
+                minimum=0.0,
+                maximum=5.0,
+                value=1.2,
+                step=0.1,
+                interactive=True,
+                label="Sampling temperature",
+            )
+            decoding_strategy.change(
+                fn=lambda selection: gr.Slider.update(
+                    visible=(
+                        selection in ["contrastive_sampling", "beam_sampling", "Top P Sampling", "sampling_top_k"]
+                    )
+                ),
+                inputs=decoding_strategy,
+                outputs=temperature,
+            )
+            num_beams = gr.Slider(
+                minimum=0,
+                maximum=20,
+                value=3.0,
+                step=1.0,
+                interactive=True,
+                visible=False,
+                label="Number of beams",
+                info="Only used if `decoding_strategy` is `beam_search` or `beam_sampling`.",
+            )
+            decoding_strategy.change(
+                fn=lambda selection: gr.Slider.update(visible=(selection in ["beam_search", "beam_sampling"])),
+                inputs=decoding_strategy,
+                outputs=num_beams,
+            )
+            top_p = gr.Slider(
+                minimum=0.0,
+                maximum=1.0,
+                value=0.8,
+                step=0.01,
+                interactive=True,
+                visible=True,
+                label="Top P",
+                info=(
+                    "If set to float < 1, only the smallest set of most probable tokens with probabilities that"
+                    " add up to top_p or higher are kept for generation."
+                ),
+            )
+            decoding_strategy.change(
+                fn=lambda selection: gr.Slider.update(visible=(selection in ["Top P Sampling"])),
+                inputs=decoding_strategy,
+                outputs=top_p,
+            )
+            top_k = gr.Slider(
+                minimum=0,
+                maximum=500,
+                value=50,
+                step=1,
+                interactive=True,
+                visible=False,
+                label="Top K",
+                info="The number of highest probability vocabulary tokens to keep for top-k-filtering.",
+            )
+            decoding_strategy.change(
+                fn=lambda selection: gr.Slider.update(visible=(selection in ["sampling_top_k"])),
+                inputs=decoding_strategy,
+                outputs=top_k,
+            )
+            length_penalty = gr.Slider(
+                minimum=-1000.0,
+                maximum=1000.0,
+                value=1.0,
+                step=0.1,
+                interactive=True,
+                visible=False,
+                label="Length penalty",
+                info=(
+                    "length_penalty > 0.0 promotes longer sequences, while length_penalty < 0.0 encourages shorter"
+                    " sequences. Only used if `decoding_strategy` is `beam_search` or `beam_sampling`."
+                ),
+            )
+            decoding_strategy.change(
+                fn=lambda selection: gr.Slider.update(visible=(selection in ["beam_search", "beam_sampling"])),
+                inputs=decoding_strategy,
+                outputs=length_penalty,
+            )
+            penalty_alpha = gr.Slider(
+                minimum=0.0,
+                maximum=5.0,
+                value=0.95,
+                step=0.05,
+                interactive=True,
+                visible=False,
+                label="Penalty alpha",
+                info="Only used if `decoding_strategy` is `contrastive_sampling`.",
+            )
+            decoding_strategy.change(
+                fn=lambda selection: gr.Slider.update(visible=(selection in ["contrastive_sampling"])),
+                inputs=decoding_strategy,
+                outputs=penalty_alpha,
+            )
+            
+    
     def model_inference(
         user_prompt_str,
         chat_history,