Update app.py
Browse files
app.py
CHANGED
|
@@ -11,8 +11,13 @@ load_dotenv()
|
|
| 11 |
genai.configure(api_key=os.getenv("GOOGLE_API_KEY"))
|
| 12 |
|
| 13 |
# Function to process the image and get response from Gemini model
|
| 14 |
-
def get_gemini_response(
|
| 15 |
try:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
# Validate the image file path
|
| 17 |
if not uploaded_file_path or not os.path.exists(uploaded_file_path):
|
| 18 |
return "Please upload a valid image."
|
|
@@ -32,11 +37,6 @@ def get_gemini_response(input_prompt, uploaded_file_path, query):
|
|
| 32 |
except Exception as e:
|
| 33 |
return f"Error: {e}"
|
| 34 |
|
| 35 |
-
# Define input prompt
|
| 36 |
-
default_prompt = """
|
| 37 |
-
You are an expert in understanding invoices. You will receive input images as invoices and
|
| 38 |
-
you will have to answer questions based on the input image.
|
| 39 |
-
"""
|
| 40 |
|
| 41 |
# Define Gradio interface
|
| 42 |
with gr.Blocks() as invoice_extractor:
|
|
@@ -47,9 +47,6 @@ with gr.Blocks() as invoice_extractor:
|
|
| 47 |
The system uses Google's Gemini model to extract and interpret the invoice details.
|
| 48 |
"""
|
| 49 |
)
|
| 50 |
-
|
| 51 |
-
input_prompt = default_prompt
|
| 52 |
-
# gr.Textbox(label="Input Prompt", value=default_prompt, lines=2)
|
| 53 |
image_input = gr.Image(label="Upload Invoice Image", type="filepath") # Use type="filepath"
|
| 54 |
query_input = gr.Textbox(label="Enter your query about the invoice", placeholder="e.g., What is the total amount?")
|
| 55 |
output_response = gr.Textbox(label="Response", lines=5)
|
|
@@ -60,7 +57,7 @@ with gr.Blocks() as invoice_extractor:
|
|
| 60 |
# Set the button to call the processing function
|
| 61 |
submit_btn.click(
|
| 62 |
get_gemini_response,
|
| 63 |
-
inputs=[
|
| 64 |
outputs=output_response
|
| 65 |
)
|
| 66 |
|
|
|
|
| 11 |
genai.configure(api_key=os.getenv("GOOGLE_API_KEY"))
|
| 12 |
|
| 13 |
# Function to process the image and get response from Gemini model
|
| 14 |
+
def get_gemini_response(uploaded_file_path, query):
|
| 15 |
try:
|
| 16 |
+
# Define input prompt
|
| 17 |
+
input_prompt = """
|
| 18 |
+
You are an expert in understanding invoices. You will receive input images as invoices and
|
| 19 |
+
you will have to answer questions based on the input image.
|
| 20 |
+
"""
|
| 21 |
# Validate the image file path
|
| 22 |
if not uploaded_file_path or not os.path.exists(uploaded_file_path):
|
| 23 |
return "Please upload a valid image."
|
|
|
|
| 37 |
except Exception as e:
|
| 38 |
return f"Error: {e}"
|
| 39 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 40 |
|
| 41 |
# Define Gradio interface
|
| 42 |
with gr.Blocks() as invoice_extractor:
|
|
|
|
| 47 |
The system uses Google's Gemini model to extract and interpret the invoice details.
|
| 48 |
"""
|
| 49 |
)
|
|
|
|
|
|
|
|
|
|
| 50 |
image_input = gr.Image(label="Upload Invoice Image", type="filepath") # Use type="filepath"
|
| 51 |
query_input = gr.Textbox(label="Enter your query about the invoice", placeholder="e.g., What is the total amount?")
|
| 52 |
output_response = gr.Textbox(label="Response", lines=5)
|
|
|
|
| 57 |
# Set the button to call the processing function
|
| 58 |
submit_btn.click(
|
| 59 |
get_gemini_response,
|
| 60 |
+
inputs=[image_input, query_input],
|
| 61 |
outputs=output_response
|
| 62 |
)
|
| 63 |
|