vlalab 0.1.0__tar.gz → 0.1.1__tar.gz

vlalab-0.1.1/MANIFEST.in

@@ -0,0 +1,12 @@
+include LICENSE
+include README.md
+include pyproject.toml
+
+recursive-include src/vlalab *.py
+recursive-include src/vlalab/apps *.py
+
+prune */__pycache__
+prune *.egg-info
+global-exclude *.pyc
+global-exclude *.pyo
+global-exclude .DS_Store

{vlalab-0.1.0/src/vlalab.egg-info → vlalab-0.1.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: vlalab
-Version: 0.1.0
+Version: 0.1.1
 Summary: A toolbox for tracking and visualizing the real-world deployment process of VLA models
 Author: VLA-Lab Contributors
 License: MIT
@@ -44,14 +44,13 @@ Dynamic: license-file
 
 <div align="center">
   
-# 🧪 VLA-Lab
+# 🦾 VLA-Lab
 
 ### The Missing Toolkit for Vision-Language-Action Model Deployment
 
 [![Python 3.8+](https://img.shields.io/badge/python-3.8+-blue.svg)](https://www.python.org/downloads/)
 [![License: MIT](https://img.shields.io/badge/License-MIT-green.svg)](https://opensource.org/licenses/MIT)
 [![PyPI version](https://img.shields.io/badge/pypi-v0.1.0-orange.svg)](https://pypi.org/project/vlalab/)
-[![PRs Welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg)](https://github.com/VLA-Lab/VLA-Lab/pulls)
 
 **Debug • Visualize • Analyze** your VLA deployments in the real world
 
@@ -129,30 +128,21 @@ Explore Zarr-format training/evaluation datasets with intuitive UI.
 </tr>
 </table>
 
-### 🔌 Framework Support
-
-| Framework | Status | Integration |
-|-----------|--------|-------------|
-| **Diffusion Policy** | ✅ Supported | Drop-in logger |
-| **NVIDIA GR00T** | ✅ Supported | Native adapter |
-| **OpenVLA-OFT** | 🚧 Coming Soon | — |
-| **Pi0.5** | 🚧 Coming Soon | — |
 
 ---
 
 ## 🔧 Installation
 
 ```bash
-# Basic installation
 pip install vlalab
+```
 
-# Full installation (with Zarr dataset support)
-pip install "vlalab[full]"
+Or install from source:
 
-# Development installation
+```bash
 git clone https://github.com/VLA-Lab/VLA-Lab.git
 cd VLA-Lab
-pip install -e ".[dev]"
+pip install -e .
 ```
 
 ---
@@ -204,7 +194,7 @@ for step in range(100):
         "images": {"front": obs["front_cam"], "wrist": obs["wrist_cam"]},
         "inference_latency_ms": latency,
     })
-    
+
     robot.execute(action)
 
 # Auto-finishes on exit, or call manually
@@ -345,51 +335,11 @@ vlalab_runs/
     └── run_20240115_103000/        # Run
         ├── meta.json               # Metadata (model, task, robot, cameras)
         ├── steps.jsonl             # Step records (one JSON per line)
-        └── artifacts/
+└── artifacts/
             └── images/             # Saved images
-                ├── step_000000_front.jpg
+        ├── step_000000_front.jpg
                 ├── step_000000_wrist.jpg
-                └── ...
-```
-
----
-
-## 🔗 Framework Integration
-
-### Diffusion Policy
-
-```python
-# In your inference_server.py
-from vlalab import RunLogger
-
-class DPInferenceServer:
-    def __init__(self, checkpoint_path):
-        self.logger = RunLogger(
-            run_dir=f"runs/{datetime.now():%Y%m%d_%H%M%S}",
-            model_name="diffusion_policy",
-            model_path=str(checkpoint_path),
-        )
-    
-    def infer(self, obs):
-        action = self.model(obs)
-        self.logger.log_step(step_idx=self.step, ...)
-        return action
-```
-
-### NVIDIA GR00T
-
-```python
-# In your inference_server_groot.py
-from vlalab import RunLogger
-
-class GrootInferenceServer:
-    def __init__(self, model_path, task_prompt):
-        self.logger = RunLogger(
-            run_dir=f"runs/{datetime.now():%Y%m%d_%H%M%S}",
-            model_name="groot",
-            model_path=str(model_path),
-            task_prompt=task_prompt,
-        )
+        └── ...
 ```
 
 ---
@@ -410,20 +360,12 @@ class GrootInferenceServer:
 
 ## 🤝 Contributing
 
-We welcome contributions! See our [Contributing Guide](CONTRIBUTING.md) for details.
+We welcome contributions! 
 
 ```bash
-# Setup development environment
 git clone https://github.com/VLA-Lab/VLA-Lab.git
 cd VLA-Lab
-pip install -e ".[dev]"
-
-# Run tests
-pytest
-
-# Format code
-black src/
-ruff check src/ --fix
+pip install -e .
 ```
 
 ---

{vlalab-0.1.0 → vlalab-0.1.1}/README.md

@@ -1,13 +1,12 @@
 <div align="center">
   
-# 🧪 VLA-Lab
+# 🦾 VLA-Lab
 
 ### The Missing Toolkit for Vision-Language-Action Model Deployment
 
 [![Python 3.8+](https://img.shields.io/badge/python-3.8+-blue.svg)](https://www.python.org/downloads/)
 [![License: MIT](https://img.shields.io/badge/License-MIT-green.svg)](https://opensource.org/licenses/MIT)
 [![PyPI version](https://img.shields.io/badge/pypi-v0.1.0-orange.svg)](https://pypi.org/project/vlalab/)
-[![PRs Welcome](https://img.shields.io/badge/PRs-welcome-brightgreen.svg)](https://github.com/VLA-Lab/VLA-Lab/pulls)
 
 **Debug • Visualize • Analyze** your VLA deployments in the real world
 
@@ -85,30 +84,21 @@ Explore Zarr-format training/evaluation datasets with intuitive UI.
 </tr>
 </table>
 
-### 🔌 Framework Support
-
-| Framework | Status | Integration |
-|-----------|--------|-------------|
-| **Diffusion Policy** | ✅ Supported | Drop-in logger |
-| **NVIDIA GR00T** | ✅ Supported | Native adapter |
-| **OpenVLA-OFT** | 🚧 Coming Soon | — |
-| **Pi0.5** | 🚧 Coming Soon | — |
 
 ---
 
 ## 🔧 Installation
 
 ```bash
-# Basic installation
 pip install vlalab
+```
 
-# Full installation (with Zarr dataset support)
-pip install "vlalab[full]"
+Or install from source:
 
-# Development installation
+```bash
 git clone https://github.com/VLA-Lab/VLA-Lab.git
 cd VLA-Lab
-pip install -e ".[dev]"
+pip install -e .
 ```
 
 ---
@@ -160,7 +150,7 @@ for step in range(100):
         "images": {"front": obs["front_cam"], "wrist": obs["wrist_cam"]},
         "inference_latency_ms": latency,
     })
-    
+
     robot.execute(action)
 
 # Auto-finishes on exit, or call manually
@@ -301,51 +291,11 @@ vlalab_runs/
     └── run_20240115_103000/        # Run
         ├── meta.json               # Metadata (model, task, robot, cameras)
         ├── steps.jsonl             # Step records (one JSON per line)
-        └── artifacts/
+└── artifacts/
             └── images/             # Saved images
-                ├── step_000000_front.jpg
+        ├── step_000000_front.jpg
                 ├── step_000000_wrist.jpg
-                └── ...
-```
-
----
-
-## 🔗 Framework Integration
-
-### Diffusion Policy
-
-```python
-# In your inference_server.py
-from vlalab import RunLogger
-
-class DPInferenceServer:
-    def __init__(self, checkpoint_path):
-        self.logger = RunLogger(
-            run_dir=f"runs/{datetime.now():%Y%m%d_%H%M%S}",
-            model_name="diffusion_policy",
-            model_path=str(checkpoint_path),
-        )
-    
-    def infer(self, obs):
-        action = self.model(obs)
-        self.logger.log_step(step_idx=self.step, ...)
-        return action
-```
-
-### NVIDIA GR00T
-
-```python
-# In your inference_server_groot.py
-from vlalab import RunLogger
-
-class GrootInferenceServer:
-    def __init__(self, model_path, task_prompt):
-        self.logger = RunLogger(
-            run_dir=f"runs/{datetime.now():%Y%m%d_%H%M%S}",
-            model_name="groot",
-            model_path=str(model_path),
-            task_prompt=task_prompt,
-        )
+        └── ...
 ```
 
 ---
@@ -366,20 +316,12 @@ class GrootInferenceServer:
 
 ## 🤝 Contributing
 
-We welcome contributions! See our [Contributing Guide](CONTRIBUTING.md) for details.
+We welcome contributions! 
 
 ```bash
-# Setup development environment
 git clone https://github.com/VLA-Lab/VLA-Lab.git
 cd VLA-Lab
-pip install -e ".[dev]"
-
-# Run tests
-pytest
-
-# Format code
-black src/
-ruff check src/ --fix
+pip install -e .
 ```
 
 ---

{vlalab-0.1.0 → vlalab-0.1.1}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "vlalab"
-version = "0.1.0"
+version = "0.1.1"
 description = "A toolbox for tracking and visualizing the real-world deployment process of VLA models"
 readme = "README.md"
 license = {text = "MIT"}

{vlalab-0.1.0 → vlalab-0.1.1}/src/vlalab/__init__.py

@@ -38,7 +38,7 @@ Advanced API:
     logger.close()
 """
 
-__version__ = "0.1.0"
+__version__ = "0.1.1"
 
 # Simple API
 from vlalab.core import (
@@ -59,6 +59,9 @@ from vlalab.core import (
 from vlalab.logging import RunLogger
 from vlalab.schema import StepRecord, RunMeta, ImageRef
 
+# Evaluation API
+from vlalab.eval import EvalPolicy, ModalityConfig, OpenLoopEvaluator
+
 __all__ = [
     # Version
     "__version__",
@@ -79,4 +82,8 @@ __all__ = [
     "StepRecord",
     "RunMeta",
     "ImageRef",
+    # Evaluation API
+    "EvalPolicy",
+    "ModalityConfig",
+    "OpenLoopEvaluator",
 ]

vlalab-0.1.1/src/vlalab/apps/streamlit/app.py

@@ -0,0 +1,376 @@
+"""
+VLA-Lab Streamlit Multi-Page Application
+
+Main entry point for the visualization app.
+Usage: streamlit run app.py
+"""
+
+import streamlit as st
+from pathlib import Path
+
+# Page config must be the first Streamlit command
+st.set_page_config(
+    page_title="VLA-Lab",
+    page_icon="🤖",
+    layout="wide",
+    initial_sidebar_state="expanded",
+)
+
+# Custom CSS for styling and hiding default navigation
+st.markdown("""
+<style>
+/* Hide default Streamlit page navigation */
+[data-testid="stSidebarNav"] {
+    display: none !important;
+}
+
+/* Hide hamburger menu on pages */
+header[data-testid="stHeader"] {
+    background: transparent;
+}
+
+/* Sidebar styling */
+[data-testid="stSidebar"] {
+    background: linear-gradient(180deg, #1a1a2e 0%, #16213e 100%);
+}
+
+[data-testid="stSidebar"] .stMarkdown,
+[data-testid="stSidebar"] .stMarkdown p,
+[data-testid="stSidebar"] .stMarkdown h1,
+[data-testid="stSidebar"] .stMarkdown h2,
+[data-testid="stSidebar"] .stMarkdown h3,
+[data-testid="stSidebar"] .stMarkdown h4 {
+    color: #e8e8e8 !important;
+}
+
+[data-testid="stSidebar"] label,
+[data-testid="stSidebar"] .stTextInput label,
+[data-testid="stSidebar"] .stSelectbox label,
+[data-testid="stSidebar"] .stMultiSelect label {
+    color: #d0d0d0 !important;
+}
+
+[data-testid="stSidebar"] .stCaption,
+[data-testid="stSidebar"] small {
+    color: #a0a0a0 !important;
+}
+
+/* Sidebar title */
+.sidebar-title {
+    font-size: 1.8rem;
+    font-weight: 700;
+    background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+    -webkit-background-clip: text;
+    -webkit-text-fill-color: transparent;
+    background-clip: text;
+    margin-bottom: 0.5rem;
+    letter-spacing: -0.5px;
+}
+
+.sidebar-subtitle {
+    color: #a0a0a0;
+    font-size: 0.85rem;
+    margin-bottom: 1.5rem;
+}
+
+/* Navigation styling */
+[data-testid="stSidebar"] .stRadio > label {
+    color: #ffffff !important;
+    font-weight: 500;
+}
+
+[data-testid="stSidebar"] .stRadio > div {
+    gap: 0.3rem;
+}
+
+[data-testid="stSidebar"] .stRadio > div > label {
+    padding: 0.6rem 0.8rem;
+    border-radius: 8px;
+    transition: all 0.2s ease;
+    cursor: pointer;
+    color: #e8e8e8 !important;
+}
+
+[data-testid="stSidebar"] .stRadio > div > label span,
+[data-testid="stSidebar"] .stRadio > div > label p {
+    color: #e8e8e8 !important;
+}
+
+[data-testid="stSidebar"] .stRadio > div > label:hover {
+    background: rgba(102, 126, 234, 0.25);
+}
+
+[data-testid="stSidebar"] .stRadio > div > label[data-checked="true"] {
+    background: linear-gradient(135deg, rgba(102, 126, 234, 0.4) 0%, rgba(118, 75, 162, 0.4) 100%);
+    border-left: 3px solid #667eea;
+}
+
+[data-testid="stSidebar"] .stRadio > div > label[data-checked="true"] span,
+[data-testid="stSidebar"] .stRadio > div > label[data-checked="true"] p {
+    color: #ffffff !important;
+    font-weight: 600;
+}
+
+/* Main content area */
+.main .block-container {
+    padding-top: 2rem;
+    max-width: 1400px;
+}
+
+/* Home page styling */
+.hero-title {
+    font-size: 3rem;
+    font-weight: 800;
+    background: linear-gradient(135deg, #667eea 0%, #764ba2 50%, #f093fb 100%);
+    -webkit-background-clip: text;
+    -webkit-text-fill-color: transparent;
+    background-clip: text;
+    margin-bottom: 0.5rem;
+}
+
+.hero-subtitle {
+    font-size: 1.2rem;
+    color: #666;
+    margin-bottom: 2rem;
+}
+
+/* Feature cards */
+.feature-card {
+    background: linear-gradient(135deg, #f5f7fa 0%, #e4e8eb 100%);
+    border-radius: 12px;
+    padding: 1.5rem;
+    margin: 0.5rem 0;
+    border-left: 4px solid #667eea;
+    transition: transform 0.2s ease, box-shadow 0.2s ease;
+}
+
+.feature-card:hover {
+    transform: translateY(-2px);
+    box-shadow: 0 4px 12px rgba(102, 126, 234, 0.15);
+}
+
+.feature-card h4 {
+    color: #1a1a2e;
+    margin-bottom: 0.5rem;
+}
+
+.feature-card p {
+    color: #555;
+    font-size: 0.9rem;
+    margin: 0;
+}
+
+/* Status badges */
+.status-badge {
+    display: inline-block;
+    padding: 0.2rem 0.6rem;
+    border-radius: 20px;
+    font-size: 0.75rem;
+    font-weight: 600;
+}
+
+.status-supported {
+    background: #d4edda;
+    color: #155724;
+}
+
+/* Footer */
+.sidebar-footer {
+    color: #666;
+    font-size: 0.75rem;
+    padding: 1rem 0;
+    border-top: 1px solid rgba(255,255,255,0.1);
+    margin-top: 2rem;
+}
+
+/* Tabs styling */
+.stTabs [data-baseweb="tab-list"] {
+    gap: 8px;
+}
+
+.stTabs [data-baseweb="tab"] {
+    border-radius: 8px 8px 0 0;
+    padding: 0.5rem 1rem;
+}
+
+/* Metrics */
+[data-testid="stMetric"] {
+    background: linear-gradient(135deg, #f8f9fa 0%, #e9ecef 100%);
+    padding: 1rem;
+    border-radius: 10px;
+    border: 1px solid #dee2e6;
+}
+
+/* Info boxes */
+.stAlert {
+    border-radius: 10px;
+}
+
+/* Dataframe */
+.stDataFrame {
+    border-radius: 10px;
+    overflow: hidden;
+}
+</style>
+""", unsafe_allow_html=True)
+
+# Setup matplotlib fonts
+try:
+    from vlalab.viz.mpl_fonts import setup_matplotlib_fonts
+    setup_matplotlib_fonts(verbose=False)
+except Exception:
+    pass
+
+
+def main():
+    # Sidebar header
+    st.sidebar.markdown("""
+        <div class="sidebar-title">🤖 VLA-Lab</div>
+        <div class="sidebar-subtitle">VLA 部署追踪与可视化</div>
+    """, unsafe_allow_html=True)
+    
+    # Navigation
+    pages = {
+        "🏠 首页": "home",
+        "🔬 推理回放": "inference",
+        "📊 数据集浏览": "dataset",
+        "📈 延迟分析": "latency",
+        "🎯 开环评估": "eval",
+    }
+    
+    selection = st.sidebar.radio(
+        "导航",
+        list(pages.keys()),
+        label_visibility="collapsed",
+    )
+    
+    page_name = pages[selection]
+    
+    if page_name == "home":
+        show_home_page()
+    elif page_name == "inference":
+        from vlalab.apps.streamlit.pages import inference_viewer
+        inference_viewer.render()
+    elif page_name == "dataset":
+        from vlalab.apps.streamlit.pages import dataset_viewer
+        dataset_viewer.render()
+    elif page_name == "latency":
+        from vlalab.apps.streamlit.pages import latency_viewer
+        latency_viewer.render()
+    elif page_name == "eval":
+        from vlalab.apps.streamlit.pages import eval_viewer
+        eval_viewer.render()
+    
+    # Footer
+    st.sidebar.markdown("""
+        <div class="sidebar-footer">
+            VLA-Lab v0.1.1<br>
+            <a href="https://github.com/VLA-Lab/VLA-Lab" style="color: #667eea;">GitHub</a>
+        </div>
+    """, unsafe_allow_html=True)
+
+
+def show_home_page():
+    # Hero section
+    st.markdown("""
+        <div class="hero-title">VLA-Lab</div>
+        <div class="hero-subtitle">
+            专为 VLA (Vision-Language-Action) 模型设计的实机部署追踪与可视化工具箱
+        </div>
+    """, unsafe_allow_html=True)
+    
+    st.divider()
+    
+    # Features section
+    st.markdown("### ✨ 核心功能")
+    
+    col1, col2 = st.columns(2)
+    
+    with col1:
+        st.markdown("""
+        <div class="feature-card">
+            <h4>🔬 推理回放 Inference Replay</h4>
+            <p>
+                逐帧回放策略推理过程，支持多相机视角、3D 轨迹可视化、延迟诊断。
+                快速定位部署问题。
+            </p>
+        </div>
+        """, unsafe_allow_html=True)
+        
+        st.markdown("""
+        <div class="feature-card">
+            <h4>📊 数据集浏览 Dataset Viewer</h4>
+            <p>
+                可视化 Zarr 格式训练数据集，支持 Episode 导航、图像网格、动作轨迹分析。
+            </p>
+        </div>
+        """, unsafe_allow_html=True)
+    
+    with col2:
+        st.markdown("""
+        <div class="feature-card">
+            <h4>📈 延迟分析 Latency Analysis</h4>
+            <p>
+                深度分析控制回路时延：传输延迟、推理延迟、端到端回路时间。
+                多运行对比，识别性能瓶颈。
+            </p>
+        </div>
+        """, unsafe_allow_html=True)
+        
+        st.markdown("""
+        <div class="feature-card">
+            <h4>🎯 开环评估 Open-Loop Eval</h4>
+            <p>
+                对比模型预测动作与真实动作，计算 MSE/MAE 指标，
+                多维度动作可视化。
+            </p>
+        </div>
+        """, unsafe_allow_html=True)
+    
+    st.markdown("### 🔧 支持的框架")
+    
+    col1, col2, col3 = st.columns([1, 1, 2])
+    
+    with col1:
+        st.markdown("""
+        **RealWorld-DP**  
+        <span class="status-badge status-supported">✅ 已支持</span>
+        """, unsafe_allow_html=True)
+    
+    with col2:
+        st.markdown("""
+        **Isaac-GR00T**  
+        <span class="status-badge status-supported">✅ 已支持</span>
+        """, unsafe_allow_html=True)
+    
+    st.markdown("### 🚀 快速开始")
+    
+    st.code("""
+# 1. 安装
+pip install vlalab
+
+# 2. 在推理代码中集成日志
+import vlalab
+
+logger = vlalab.init(
+    project="my_project",
+    model="gr00t-n1",
+    task="pick_and_place",
+)
+
+# 3. 记录每步数据
+logger.log_step(
+    obs={"images": [img], "state": state},
+    action={"values": action},
+    timing={"inference_latency_ms": latency},
+)
+
+# 4. 启动可视化
+# vlalab view
+""", language="python")
+    
+    st.info("👈 从左侧导航栏选择功能开始使用")
+
+
+if __name__ == "__main__":
+    main()