Dom/OmniParser
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

small fixes

Browse Source
This commit is contained in:
yadonglu
2025-02-03 17:05:17 -08:00
parent 3e51ba0da6
commit 8725445881
3 changed files with 12 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
computer_use_demo/gradio/agent/vlm_agent.py
View File

@@ -159,7 +159,7 @@ class VLMAgent:
self.output_callback(f'<img src="data:image/png;base64,{img_to_show_base64}">', sender="bot") self.output_callback(f'<img src="data:image/png;base64,{img_to_show_base64}">', sender="bot")
self.output_callback( self.output_callback(
f'<details>' f'<details>'
f' <summary>Screen Info for OmniParser Agent</summary>' f' <summary>Parsed Screen elemetns by OmniParser</summary>'
f' <pre>{screen_info}</pre>' f' <pre>{screen_info}</pre>'
f'</details>', f'</details>',
sender="bot" sender="bot"

12
computer_use_demo/gradio/app.py
View File

@@ -28,9 +28,9 @@ API_KEY_FILE = CONFIG_DIR / "api_key"
INTRO_TEXT = ''' INTRO_TEXT = '''
🚀🤖✨ It's Play Time! 🚀🤖✨ It's Play Time!
Welcome to the OmniParser+X Computer Use Demo! X = [GPT-4o, R1, Qwen2.5VL, Claude]. Let OmniParser turn your general purpose vision-langauge model to an AI agent. Welcome to the OmniParser+X Computer Use Demo! X = [GPT family (4o/o1/o3-mini), Claude, deepseek R1/V3, Qwen-2.5VL]. Let OmniParser turn your general purpose vision-langauge model to an AI agent.
Type a message and press submit to start OmniParser+X. Press the trash icon in the chat to clear the message history. Type a message and press submit to start OmniParser+X. Press stop to pause, and press the trash icon in the chat to clear the message history.
''' '''
def parse_arguments(): def parse_arguments():
@@ -71,6 +71,8 @@ def setup_state(state):
state["only_n_most_recent_images"] = 2 state["only_n_most_recent_images"] = 2
if 'chatbot_messages' not in state: if 'chatbot_messages' not in state:
state['chatbot_messages'] = [] state['chatbot_messages'] = []
if 'stop' not in state:
state['stop'] = False
async def main(state): async def main(state):
"""Render loop for Gradio""" """Render loop for Gradio"""
@@ -207,6 +209,10 @@ def valid_params(user_input, state):
return errors return errors
def process_input(user_input, state): def process_input(user_input, state):
# Reset the stop flag
if state["stop"]:
state["stop"] = False
errors = valid_params(user_input, state) errors = valid_params(user_input, state)
if errors: if errors:
raise gr.Error("Validation errors: " + ", ".join(errors)) raise gr.Error("Validation errors: " + ", ".join(errors))
@@ -260,7 +266,7 @@ with gr.Blocks(theme=gr.themes.Default()) as demo:
} }
</style> </style>
""") """)
state = gr.State({"stop": False}) state = gr.State({})
setup_state(state.value) setup_state(state.value)

2
computer_use_demo/readme.md
View File

@@ -1,6 +1,6 @@
# OmniParser+X Computer Use Demo # OmniParser+X Computer Use Demo
Control a Windows 11 VM with OmniParser+X (X = [GPT-4o/4o-mini, Claude, ...]). Control a Windows 11 VM with OmniParser+X (X = [GPT family (4o/o1/o3-mini), Claude, deepseek R1/V3, Qwen-2.5VL]).
<p align="center"> <p align="center">
<img src="../imgs/som_overlaid_omni.png" alt="OmniParser+X Computer Use Demo screenshot"> <img src="../imgs/som_overlaid_omni.png" alt="OmniParser+X Computer Use Demo screenshot">
</p> </p>