74b7c189af
* feat: add claude support * feat: add script for end-to-end evaluation with logging and task distribution * feat&fix: add tool result handling and update model default in evaluation script * chore: remove run_test_env.py script * feat&fix: implement action parsing for tool calls and update default action space * fix: update text formatting in action parsing and replace logger import * feat&fix: implement action parsing for tool calls and add screen size handling * feat: add setup instructions for Anthropic API integration * feat: add notice about image size limitations for Anthropic API * Delete test_env/logger.py * Delete test_env/utils.py * fix: update logger usage to use global logger and improve error handling * feat&fix: add configuration management API endpoints and update UI for configuration selection * feat&fix: update environment configuration, enhance task statistics, and improve UI responsiveness * feat&fix: add configuration toggle button in UI and improve task loading performance * feat&fix: add accuracy percentage display to score and style updates for UI
103 lines
4.5 KiB
HTML
103 lines
4.5 KiB
HTML
<!DOCTYPE html>
|
|
<html>
|
|
<head>
|
|
<meta charset="UTF-8">
|
|
<meta name="viewport" content="width=device-width, initial-scale=1.0, maximum-scale=1.0, user-scalable=no">
|
|
<title>OSWorld Monitor</title>
|
|
<link rel="icon" href="/static/favicon.png" type="image/x-icon">
|
|
<link rel="shortcut icon" href="/static/favicon.png" type="image/x-icon">
|
|
<link rel="apple-touch-icon" href="/static/favicon.png">
|
|
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/5.15.4/css/all.min.css">
|
|
<link rel="stylesheet" href="/static/style.css">
|
|
<link rel="stylesheet" href="/static/index.css">
|
|
</head>
|
|
<body>
|
|
<div class="layout-container">
|
|
<!-- Floating Config Button and Sidebar -->
|
|
<div class="config-sidebar" id="config-sidebar">
|
|
<div class="config-toggle-btn">
|
|
<i class="fas fa-cogs"></i>
|
|
</div>
|
|
<div class="config-panel">
|
|
<div class="config-header">
|
|
<i class="fas fa-cogs"></i>
|
|
<span>Configuration</span>
|
|
</div>
|
|
<div class="config-content">
|
|
<div class="config-selector">
|
|
<div class="selector-item">
|
|
<label for="config-select">Select Configuration:</label>
|
|
<select id="config-select" onchange="changeConfiguration()">
|
|
<option value="">Loading configurations...</option>
|
|
</select>
|
|
</div>
|
|
</div>
|
|
<div class="config-list">
|
|
<div class="config-item">
|
|
<span class="config-label">Action Space:</span>
|
|
<span class="config-value" id="action-space">Loading...</span>
|
|
</div>
|
|
<div class="config-item">
|
|
<span class="config-label">Observation:</span>
|
|
<span class="config-value" id="observation-type">Loading...</span>
|
|
</div>
|
|
<div class="config-item">
|
|
<span class="config-label">Model:</span>
|
|
<span class="config-value" id="model-name">Loading...</span>
|
|
</div>
|
|
<div class="config-item">
|
|
<span class="config-label">Max Steps:</span>
|
|
<span class="config-value" id="max-steps">Loading...</span>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
|
|
<!-- Main Content -->
|
|
<div class="main-content">
|
|
<h1>OSWorld Monitor <span class="system-status online">System Online</span></h1>
|
|
|
|
<!-- Score Display Banner -->
|
|
<div class="score-banner">
|
|
<div class="score-content">
|
|
<i class="fas fa-star"></i>
|
|
<span class="score-label">Score:</span>
|
|
<span id="score-display" class="score-value">Loading...</span>
|
|
</div>
|
|
</div>
|
|
|
|
<div class="dashboard-stats">
|
|
<div class="stat-card">
|
|
<i class="fas fa-running"></i>
|
|
<span id="active-tasks">Loading...</span>
|
|
<div class="stat-label">Active</div>
|
|
</div>
|
|
<div class="stat-card">
|
|
<i class="fas fa-check-circle"></i>
|
|
<span id="completed-tasks">Loading...</span>
|
|
<div class="stat-label">Completed</div>
|
|
</div>
|
|
<div class="stat-card">
|
|
<i class="fas fa-exclamation-circle"></i>
|
|
<span id="error-tasks">Loading...</span>
|
|
<div class="stat-label">Error</div>
|
|
</div>
|
|
<div class="stat-card">
|
|
<i class="fas fa-tasks"></i>
|
|
<span id="total-tasks">Loading...</span>
|
|
<div class="stat-label">Total Tasks</div>
|
|
</div>
|
|
</div>
|
|
<div id="task-container">
|
|
<div class="loading-spinner">
|
|
<div class="spinner"></div>
|
|
<div>Loading task data...</div>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
<button class="fab fab-refresh" onclick="refreshPage()" title="Refresh"><i class="fas fa-sync-alt"></i></button>
|
|
<script src="/static/index.js"></script>
|
|
</body>
|
|
</html> |