Spaces:
Build error
Build error
Updated app.py to handle custom api requests
Browse files
app.py
CHANGED
|
@@ -6,6 +6,7 @@ import argparse
|
|
| 6 |
import time
|
| 7 |
import subprocess
|
| 8 |
|
|
|
|
| 9 |
import llava.serve.gradio_web_server as gws
|
| 10 |
|
| 11 |
# Execute the pip install command with additional options
|
|
@@ -24,7 +25,7 @@ def start_controller():
|
|
| 24 |
"--port",
|
| 25 |
"10000",
|
| 26 |
]
|
| 27 |
-
print(controller_command)
|
| 28 |
return subprocess.Popen(controller_command)
|
| 29 |
|
| 30 |
|
|
@@ -51,28 +52,104 @@ def start_worker(model_path: str, bits=16):
|
|
| 51 |
"liuhaotian/llava-1.5-7b",
|
| 52 |
"--use-flash-attn",
|
| 53 |
]
|
| 54 |
-
|
| 55 |
-
# worker_command += [f"--load-{bits}bit"]
|
| 56 |
-
print(worker_command)
|
| 57 |
return subprocess.Popen(worker_command)
|
| 58 |
|
| 59 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 60 |
if __name__ == "__main__":
|
| 61 |
-
parser = argparse.ArgumentParser()
|
| 62 |
-
parser.add_argument("--host", type=str, default="0.0.0.0")
|
| 63 |
-
parser.add_argument("--port", type=int)
|
| 64 |
-
parser.add_argument("--controller-url", type=str, default="http://localhost:10000")
|
| 65 |
-
parser.add_argument("--concurrency-count", type=int, default=5)
|
| 66 |
-
parser.add_argument("--model-list-mode", type=str, default="reload", choices=["once", "reload"])
|
| 67 |
-
parser.add_argument("--share", action="
|
| 68 |
-
parser.add_argument("--moderate", action="store_true")
|
| 69 |
-
parser.add_argument("--embed", action="store_true")
|
| 70 |
-
|
|
|
|
| 71 |
gws.models = []
|
| 72 |
|
| 73 |
gws.title_markdown += """ AstroLLaVA """
|
| 74 |
|
| 75 |
-
print(f"
|
| 76 |
|
| 77 |
model_path = os.getenv("model", "universeTBD/AstroLLaVA_v2")
|
| 78 |
bits = int(os.getenv("bits", 4))
|
|
@@ -82,24 +159,25 @@ if __name__ == "__main__":
|
|
| 82 |
worker_proc = start_worker(model_path, bits=bits)
|
| 83 |
|
| 84 |
# Wait for worker and controller to start
|
| 85 |
-
print("Waiting for worker and controller to start")
|
| 86 |
time.sleep(30)
|
| 87 |
|
| 88 |
exit_status = 0
|
| 89 |
try:
|
| 90 |
-
demo
|
| 91 |
-
|
|
|
|
| 92 |
demo.queue(
|
| 93 |
status_update_rate=10,
|
| 94 |
api_open=False
|
| 95 |
).launch(
|
| 96 |
-
server_name=
|
| 97 |
-
server_port=
|
| 98 |
-
share=
|
| 99 |
)
|
| 100 |
|
| 101 |
except Exception as e:
|
| 102 |
-
print(e)
|
| 103 |
exit_status = 1
|
| 104 |
finally:
|
| 105 |
worker_proc.kill()
|
|
|
|
| 6 |
import time
|
| 7 |
import subprocess
|
| 8 |
|
| 9 |
+
import gradio as gr
|
| 10 |
import llava.serve.gradio_web_server as gws
|
| 11 |
|
| 12 |
# Execute the pip install command with additional options
|
|
|
|
| 25 |
"--port",
|
| 26 |
"10000",
|
| 27 |
]
|
| 28 |
+
print("Controller Command:", controller_command)
|
| 29 |
return subprocess.Popen(controller_command)
|
| 30 |
|
| 31 |
|
|
|
|
| 52 |
"liuhaotian/llava-1.5-7b",
|
| 53 |
"--use-flash-attn",
|
| 54 |
]
|
| 55 |
+
print("Worker Command:", worker_command)
|
|
|
|
|
|
|
| 56 |
return subprocess.Popen(worker_command)
|
| 57 |
|
| 58 |
|
| 59 |
+
def handle_text_prompt(text, temperature=0.2, top_p=0.7, max_new_tokens=512):
|
| 60 |
+
"""
|
| 61 |
+
Custom API endpoint to handle text prompts.
|
| 62 |
+
Replace the placeholder logic with actual model inference.
|
| 63 |
+
"""
|
| 64 |
+
# TODO: Replace the following placeholder with actual model inference code
|
| 65 |
+
print(f"Received prompt: {text}")
|
| 66 |
+
print(f"Parameters - Temperature: {temperature}, Top P: {top_p}, Max New Tokens: {max_new_tokens}")
|
| 67 |
+
|
| 68 |
+
# Example response (replace with actual model response)
|
| 69 |
+
response = f"Model response to '{text}' with temperature={temperature}, top_p={top_p}, max_new_tokens={max_new_tokens}"
|
| 70 |
+
return response
|
| 71 |
+
|
| 72 |
+
|
| 73 |
+
def add_text_with_image(text, image, mode):
|
| 74 |
+
"""
|
| 75 |
+
Custom API endpoint to add text with an image.
|
| 76 |
+
Replace the placeholder logic with actual processing.
|
| 77 |
+
"""
|
| 78 |
+
# TODO: Replace the following placeholder with actual processing code
|
| 79 |
+
print(f"Adding text: {text}")
|
| 80 |
+
print(f"Image path: {image}")
|
| 81 |
+
print(f"Image processing mode: {mode}")
|
| 82 |
+
|
| 83 |
+
# Example response (replace with actual processing code)
|
| 84 |
+
response = f"Added text '{text}' with image at '{image}' using mode '{mode}'."
|
| 85 |
+
return response
|
| 86 |
+
|
| 87 |
+
|
| 88 |
+
def build_custom_demo(embed_mode=False, cur_dir='./', concurrency_count=5):
|
| 89 |
+
"""
|
| 90 |
+
Builds a Gradio Blocks interface with custom API endpoints.
|
| 91 |
+
"""
|
| 92 |
+
with gr.Blocks() as demo:
|
| 93 |
+
gr.Markdown("# AstroLLaVA")
|
| 94 |
+
gr.Markdown("Welcome to the AstroLLaVA interface. Use the API endpoints to interact with the model.")
|
| 95 |
+
|
| 96 |
+
with gr.Row():
|
| 97 |
+
with gr.Column():
|
| 98 |
+
gr.Markdown("## Prompt the Model")
|
| 99 |
+
text_input = gr.Textbox(label="Enter your text prompt", placeholder="Type your prompt here...")
|
| 100 |
+
temperature_slider = gr.Slider(minimum=0.0, maximum=1.0, value=0.2, label="Temperature")
|
| 101 |
+
top_p_slider = gr.Slider(minimum=0.0, maximum=1.0, value=0.7, label="Top P")
|
| 102 |
+
max_tokens_slider = gr.Slider(minimum=1, maximum=1024, value=512, step=1, label="Max New Tokens")
|
| 103 |
+
submit_button = gr.Button("Submit Prompt")
|
| 104 |
+
with gr.Column():
|
| 105 |
+
chatbot_output = gr.Textbox(label="Model Response", interactive=False)
|
| 106 |
+
|
| 107 |
+
submit_button.click(
|
| 108 |
+
fn=handle_text_prompt,
|
| 109 |
+
inputs=[text_input, temperature_slider, top_p_slider, max_tokens_slider],
|
| 110 |
+
outputs=chatbot_output,
|
| 111 |
+
api_name="prompt_model" # Custom API endpoint name
|
| 112 |
+
)
|
| 113 |
+
|
| 114 |
+
with gr.Row():
|
| 115 |
+
with gr.Column():
|
| 116 |
+
gr.Markdown("## Add Text with Image")
|
| 117 |
+
add_text_input = gr.Textbox(label="Add Text", placeholder="Enter text to add...")
|
| 118 |
+
add_image_input = gr.Image(label="Upload Image")
|
| 119 |
+
image_process_mode = gr.Radio(choices=["Crop", "Resize", "Pad", "Default"], value="Default", label="Image Process Mode")
|
| 120 |
+
add_submit_button = gr.Button("Add Text with Image")
|
| 121 |
+
with gr.Column():
|
| 122 |
+
add_output = gr.Textbox(label="Add Text Response", interactive=False)
|
| 123 |
+
|
| 124 |
+
add_submit_button.click(
|
| 125 |
+
fn=add_text_with_image,
|
| 126 |
+
inputs=[add_text_input, add_image_input, image_process_mode],
|
| 127 |
+
outputs=add_output,
|
| 128 |
+
api_name="add_text_with_image" # Another custom API endpoint
|
| 129 |
+
)
|
| 130 |
+
|
| 131 |
+
# Additional API endpoints can be added here following the same structure
|
| 132 |
+
|
| 133 |
+
return demo
|
| 134 |
+
|
| 135 |
+
|
| 136 |
if __name__ == "__main__":
|
| 137 |
+
parser = argparse.ArgumentParser(description="AstroLLaVA Gradio App")
|
| 138 |
+
parser.add_argument("--host", type=str, default="0.0.0.0", help="Hostname to listen on")
|
| 139 |
+
parser.add_argument("--port", type=int, default=7860, help="Port number")
|
| 140 |
+
parser.add_argument("--controller-url", type=str, default="http://localhost:10000", help="Controller URL")
|
| 141 |
+
parser.add_argument("--concurrency-count", type=int, default=5, help="Number of concurrent requests")
|
| 142 |
+
parser.add_argument("--model-list-mode", type=str, default="reload", choices=["once", "reload"], help="Model list mode")
|
| 143 |
+
parser.add_argument("--share", action="store_true", help="Share the Gradio app publicly")
|
| 144 |
+
parser.add_argument("--moderate", action="store_true", help="Enable moderation")
|
| 145 |
+
parser.add_argument("--embed", action="store_true", help="Enable embed mode")
|
| 146 |
+
args = parser.parse_args()
|
| 147 |
+
gws.args = args
|
| 148 |
gws.models = []
|
| 149 |
|
| 150 |
gws.title_markdown += """ AstroLLaVA """
|
| 151 |
|
| 152 |
+
print(f"AstroLLaVA arguments: {gws.args}")
|
| 153 |
|
| 154 |
model_path = os.getenv("model", "universeTBD/AstroLLaVA_v2")
|
| 155 |
bits = int(os.getenv("bits", 4))
|
|
|
|
| 159 |
worker_proc = start_worker(model_path, bits=bits)
|
| 160 |
|
| 161 |
# Wait for worker and controller to start
|
| 162 |
+
print("Waiting for worker and controller to start...")
|
| 163 |
time.sleep(30)
|
| 164 |
|
| 165 |
exit_status = 0
|
| 166 |
try:
|
| 167 |
+
# Build the custom Gradio demo with additional API endpoints
|
| 168 |
+
demo = build_custom_demo(embed_mode=False, cur_dir='./', concurrency_count=concurrency_count)
|
| 169 |
+
print("Launching Gradio with custom API endpoints...")
|
| 170 |
demo.queue(
|
| 171 |
status_update_rate=10,
|
| 172 |
api_open=False
|
| 173 |
).launch(
|
| 174 |
+
server_name=args.host,
|
| 175 |
+
server_port=args.port,
|
| 176 |
+
share=args.share
|
| 177 |
)
|
| 178 |
|
| 179 |
except Exception as e:
|
| 180 |
+
print(f"An error occurred: {e}")
|
| 181 |
exit_status = 1
|
| 182 |
finally:
|
| 183 |
worker_proc.kill()
|