Update app.py
Browse files
app.py
CHANGED
@@ -1,63 +1,42 @@
|
|
1 |
import gradio as gr
|
2 |
-
from
|
3 |
-
|
4 |
-
|
5 |
-
|
6 |
-
|
7 |
-
|
8 |
-
|
9 |
-
|
10 |
-
|
11 |
-
|
12 |
-
|
13 |
-
|
14 |
-
|
15 |
-
|
16 |
-
|
17 |
-
)
|
18 |
-
|
19 |
-
|
20 |
-
|
21 |
-
|
22 |
-
|
23 |
-
|
24 |
-
|
25 |
-
|
26 |
-
|
27 |
-
|
28 |
-
|
29 |
-
|
30 |
-
|
31 |
-
|
32 |
-
|
33 |
-
|
34 |
-
|
35 |
-
|
36 |
-
|
37 |
-
|
38 |
-
|
39 |
-
response += token
|
40 |
-
yield response
|
41 |
-
|
42 |
-
"""
|
43 |
-
For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
|
44 |
-
"""
|
45 |
-
demo = gr.ChatInterface(
|
46 |
-
respond,
|
47 |
-
additional_inputs=[
|
48 |
-
gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
|
49 |
-
gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
|
50 |
-
gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
|
51 |
-
gr.Slider(
|
52 |
-
minimum=0.1,
|
53 |
-
maximum=1.0,
|
54 |
-
value=0.95,
|
55 |
-
step=0.05,
|
56 |
-
label="Top-p (nucleus sampling)",
|
57 |
-
),
|
58 |
-
],
|
59 |
)
|
60 |
|
61 |
-
|
62 |
if __name__ == "__main__":
|
63 |
-
|
|
|
1 |
import gradio as gr
|
2 |
+
from PIL import Image
|
3 |
+
from transformers import pipeline, AutoModelForVision2Seq, AutoProcessor
|
4 |
+
import torch
|
5 |
+
|
6 |
+
# Load the OpenGVLab/InternVL-Chat-V1-5 model and processor
|
7 |
+
processor = AutoProcessor.from_pretrained("OpenGVLab/InternVL-Chat-V1-5")
|
8 |
+
model = AutoModelForVision2Seq.from_pretrained("OpenGVLab/InternVL-Chat-V1-5")
|
9 |
+
|
10 |
+
# Load the Llama3 model for text processing
|
11 |
+
llama_model = pipeline("text2text-generation", model="llama3")
|
12 |
+
|
13 |
+
def process_image(image):
|
14 |
+
# Process the image to extract the recipe using OpenGVLab
|
15 |
+
inputs = processor(images=image, return_tensors="pt")
|
16 |
+
generated_ids = model.generate(**inputs)
|
17 |
+
extracted_text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
|
18 |
+
|
19 |
+
return extracted_text
|
20 |
+
|
21 |
+
def adjust_recipe(extracted_text, adjustment):
|
22 |
+
# Create the prompt for Llama3 to adjust the recipe
|
23 |
+
prompt = f"Here is a recipe: {extracted_text}. Please {adjustment} the recipe."
|
24 |
+
response = llama_model(prompt)
|
25 |
+
return response[0]['generated_text']
|
26 |
+
|
27 |
+
def app(image, adjustment):
|
28 |
+
extracted_text = process_image(image)
|
29 |
+
adjusted_recipe = adjust_recipe(extracted_text, adjustment)
|
30 |
+
return adjusted_recipe
|
31 |
+
|
32 |
+
# Create the Gradio interface
|
33 |
+
interface = gr.Interface(
|
34 |
+
fn=app,
|
35 |
+
inputs=[gr.inputs.Image(type="pil"), gr.inputs.Dropdown(["double", "halve"])],
|
36 |
+
outputs="text",
|
37 |
+
title="Recipe Adjuster",
|
38 |
+
description="Upload an image of a recipe, and this app will double or halve the recipe."
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
39 |
)
|
40 |
|
|
|
41 |
if __name__ == "__main__":
|
42 |
+
interface.launch()
|