flashstudio 0.0.1__tar.gz

flashstudio-0.0.1/PKG-INFO

@@ -0,0 +1,107 @@
+Metadata-Version: 2.4
+Name: flashstudio
+Version: 0.0.1
+Summary: Interactive Training & Inference UI for FlashDet — runs on Google Colab
+Author: FlashVision
+License: Apache-2.0
+Project-URL: Homepage, https://github.com/FlashVision/FlashStudio
+Project-URL: Repository, https://github.com/FlashVision/FlashStudio
+Keywords: object-detection,flashdet,training,inference,ui,colab
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Science/Research
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Programming Language :: Python :: 3
+Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
+Requires-Python: >=3.10
+Description-Content-Type: text/markdown
+Requires-Dist: streamlit>=1.28.0
+Requires-Dist: plotly>=5.0
+Requires-Dist: pandas>=1.5
+Requires-Dist: pillow>=9.0
+Requires-Dist: numpy>=1.21
+Requires-Dist: opencv-python>=4.5
+Requires-Dist: pyyaml>=6.0
+Requires-Dist: flashdet
+Provides-Extra: full
+Requires-Dist: torch>=2.0; extra == "full"
+Requires-Dist: flashdet[export]; extra == "full"
+Requires-Dist: pyngrok>=6.0; extra == "full"
+Provides-Extra: ngrok
+Requires-Dist: pyngrok>=6.0; extra == "ngrok"
+Provides-Extra: dev
+Requires-Dist: ruff; extra == "dev"
+Requires-Dist: pytest; extra == "dev"
+Requires-Dist: pytest-cov; extra == "dev"
+
+# FlashStudio
+
+**Interactive Training & Inference UI for FlashDet**
+
+---
+
+## Install
+
+```bash
+pip install flashstudio
+pip install git+https://github.com/FlashVision/FlashDet.git
+```
+
+## Quick Start
+
+```bash
+# Launch locally
+flashstudio
+
+# Or with Python
+from flashstudio import launch
+launch()
+```
+
+## Google Colab
+
+```python
+!pip install flashstudio
+!pip install git+https://github.com/FlashVision/FlashDet.git
+
+from flashstudio import launch
+launch(ngrok_token="YOUR_TOKEN")
+```
+
+Get a free ngrok token at: https://dashboard.ngrok.com/get-started/your-authtoken
+
+## Features
+
+- **Training** — Launch and monitor FlashDet training with live loss curves
+- **Inference** — Run detection on images/video with 17 built-in solutions
+- **Model Config** — FlashDet (Pico → X) + YOLOv8/v9/v10/v11/YOLOX
+- **Data** — Upload datasets or download from FlashDet registry
+- **Export** — ONNX export with FP16 support
+- **Dashboard** — Project overview with recent runs and metrics
+
+## Supported Models
+
+| Model | Params | Use Case |
+|-------|--------|----------|
+| FlashDet-Pico | ~298K | Edge / MCU |
+| FlashDet-Nano | ~790K | Embedded / IoT |
+| FlashDet-Small | ~1.8M | General purpose |
+| FlashDet-Medium | ~3.6M | High accuracy |
+| FlashDet-Large | ~5.8M | High accuracy |
+| FlashDet-X | ~9.0M | Server / Max accuracy |
+
+## Requirements
+
+- Python >= 3.10
+- FlashDet (for training/inference)
+- ngrok account (free, for Colab)
+- GPU recommended for training
+
+## Links
+
+- GitHub: https://github.com/FlashVision/FlashStudio
+- FlashDet: https://github.com/FlashVision/FlashDet
+- Organization: https://flashvision.github.io
+
+## License
+
+Apache-2.0

flashstudio-0.0.1/README.md

@@ -0,0 +1,330 @@
+<p align="center">
+  <img src="https://raw.githubusercontent.com/FlashVision/FlashStudio/main/assets/logo.png" width="200" alt="FlashStudio Logo">
+</p>
+
+<h1 align="center">FlashStudio</h1>
+
+<p align="center">
+  <a href="https://pypi.org/project/flashstudio/"><img src="https://badge.fury.io/py/flashstudio.svg" alt="PyPI version"></a>
+  <img src="https://img.shields.io/badge/python-3.9+-blue.svg?logo=python&logoColor=white" alt="Python 3.9+">
+  <img src="https://img.shields.io/badge/Streamlit-UI-FF4B4B?logo=streamlit&logoColor=white" alt="Streamlit">
+  <img src="https://img.shields.io/badge/FlashDet-Powered-00CCFF?logo=lightning&logoColor=white" alt="FlashDet">
+  <img src="https://img.shields.io/badge/License-Apache%202.0-green.svg" alt="License">
+</p>
+
+<p align="center">
+  <b>Interactive Training & Inference UI for FlashDet — runs locally or on Google Colab with a Streamlit interface.</b>
+</p>
+
+<p align="center">
+  <img src="docs/mockups/flashstudio_streamlit_mockup.png" width="800" alt="FlashStudio UI"/>
+</p>
+
+## Features
+
+- 🏋️ **Training Dashboard** — Real-time monitoring with live loss curves, per-epoch visualizations, GT verification
+- 🧠 **Model Config** — All 6 FlashDet sizes + YOLOv8/v9/v10/v11/YOLOX with accurate params
+- 🔍 **Inference Pipeline** — 4-step wizard: Model → Data → Zone → Run (17 solutions, 6 trackers)
+- 📤 **Export** — ONNX export with FP16 auto-generated weights
+- 📦 **Data** — Native `flashdet download` datasets + custom upload (COCO/VOC/YOLO formats)
+- 📊 **Dashboard** — Overview with recent training runs from workspace
+- 🚀 **Colab Support** — Works out-of-the-box, no token or signup needed
+
+---
+
+## UI Tour
+
+### Dashboard
+Project overview with key metrics (GPU, Dataset, Model, mAP), pipeline status, quick actions, and recent training runs.
+
+<p align="center">
+  <img src="demos/01_dashboard.gif" width="800" alt="Dashboard"/>
+</p>
+
+### Data Management
+Upload datasets (ZIP/TAR), download from `flashdet` registry, preview images with navigation, and verify annotations.
+
+<p align="center">
+  <img src="demos/02_data.gif" width="800" alt="Data Management"/>
+</p>
+
+### Model Configuration
+Choose from 6 FlashDet sizes + YOLO variants. Configure hyperparameters, augmentations, and advanced settings (memory optimization, distributed training).
+
+<p align="center">
+  <img src="demos/03_model.gif" width="800" alt="Model Configuration"/>
+</p>
+
+### Training
+Create experiments, launch training with preflight checks, and monitor runs with real-time curves, visualizations, ground truth, logs, and file browser.
+
+<p align="center">
+  <img src="demos/04_training.gif" width="800" alt="Training"/>
+</p>
+
+### Export
+Export trained models to ONNX format with configurable opset, image size, and dynamic batching. Auto-detects saved weights from training.
+
+<p align="center">
+  <img src="demos/05_export.gif" width="800" alt="Export"/>
+</p>
+
+### Inference
+Full inference pipeline with 17 built-in solutions (object counting, speed estimation, heatmaps, security alarms, etc.). Interactive zone drawing with Polygon, Rectangle, and Line tools.
+
+<p align="center">
+  <img src="demos/06_inference.gif" width="800" alt="Inference"/>
+</p>
+
+---
+
+## Installation
+
+```bash
+pip install flashstudio
+```
+
+This automatically installs FlashDet and all other dependencies.
+
+<details>
+<summary><b>Development install (from source)</b></summary>
+
+```bash
+git clone https://github.com/FlashVision/FlashStudio.git
+cd FlashStudio
+pip install -e .
+```
+
+</details>
+
+---
+
+## Usage — Local Machine
+
+### Option 1: CLI
+
+```bash
+flashstudio --port 8501
+```
+
+### Option 2: Streamlit directly
+
+```bash
+cd FlashStudio
+streamlit run flashstudio/app.py
+```
+
+Then open **http://localhost:8501** in your browser.
+
+---
+
+## Usage — Google Colab
+
+No signup or token needed. Just 2 steps:
+
+### Step 1: Install
+
+```python
+!pip install -q flashstudio
+```
+
+### Step 2: Launch
+
+```python
+from flashstudio import launch
+launch()
+```
+
+A clickable **"Open FlashStudio"** button appears in the cell output. Click it to open the UI.
+
+FlashStudio auto-detects Colab and uses the built-in proxy to create a working URL — no `localhost`, no ngrok, no token.
+
+<details>
+<summary><b>Optional: ngrok for a stable public URL</b></summary>
+
+If the built-in proxy doesn't work for your setup, you can use ngrok as a fallback:
+
+1. Get a free token at https://dashboard.ngrok.com/get-started/your-authtoken
+2. Pass it when launching:
+
+```python
+launch(ngrok_token="YOUR_TOKEN")
+```
+
+Or set it as an environment variable:
+
+```python
+import os
+os.environ["NGROK_TOKEN"] = "YOUR_TOKEN"
+launch()
+```
+
+</details>
+
+---
+
+## Google Colab Notebooks (Ready to Use)
+
+| Notebook | Description | Link |
+|----------|-------------|------|
+| Training | Train FlashDet models | [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/FlashVision/FlashStudio/blob/main/notebooks/FlashStudio_Train.ipynb) |
+| Inference | Run detection on images/video | [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/FlashVision/FlashStudio/blob/main/notebooks/FlashStudio_Inference.ipynb) |
+
+---
+
+## Supported Models
+
+| Model | Params | Best For |
+|-------|--------|----------|
+| FlashDet-Pico | ~298K | Edge / MCU |
+| FlashDet-Nano | ~790K | Embedded / IoT |
+| FlashDet-Small | ~1.8M | General purpose |
+| FlashDet-Medium | ~3.6M | High accuracy |
+| FlashDet-Large | ~5.8M | High accuracy |
+| FlashDet-X | ~9.0M | Max accuracy / Server |
+| YOLOv8/v9/v10/v11/YOLOX | Varies | General YOLO |
+
+---
+
+## Architecture
+
+```
+FlashStudio/
+├── flashstudio/
+│   ├── __init__.py                  # Package init + launch() export
+│   ├── app.py                       # Main Streamlit app (wizard flow)
+│   ├── launcher.py                  # Colab/local launcher (auto-tunnel)
+│   ├── cli.py                       # CLI entrypoint
+│   ├── constants.py                 # Centralized constants (paths, defaults, models)
+│   │
+│   ├── pages/                       # Each page is a sub-package
+│   │   ├── dashboard/
+│   │   │   ├── page.py              # Main dashboard render
+│   │   │   ├── pipeline_status.py   # Pipeline status cards
+│   │   │   └── recent_runs.py       # Recent training runs table
+│   │   ├── data/
+│   │   │   ├── page.py              # Main data page render
+│   │   │   ├── upload.py            # Upload tab (ZIP/TAR + class config)
+│   │   │   ├── download.py          # Download tab (quick start + registries)
+│   │   │   ├── preview.py           # Preview tab (image grid + annotations)
+│   │   │   ├── verify.py            # Verify tab (dataset validation)
+│   │   │   └── helpers.py           # Shared helpers (extract, detect, convert)
+│   │   ├── model/
+│   │   │   ├── page.py              # Main model page render
+│   │   │   ├── architecture.py      # Architecture tab (FlashDet/YOLO selection)
+│   │   │   ├── hyperparams.py       # Hyperparameters tab
+│   │   │   ├── augmentation.py      # Augmentation tab
+│   │   │   ├── advanced.py          # Advanced tab (memory, distributed)
+│   │   │   └── summary.py           # Config summary bar + YAML save/load
+│   │   ├── training/
+│   │   │   ├── page.py              # Main training page render
+│   │   │   ├── _common.py           # Shared utilities (_get_save_dir)
+│   │   │   ├── launch/              # Launch sub-package
+│   │   │   │   ├── tab.py           # Launch tab entry point
+│   │   │   │   ├── preflight.py     # Pre-flight checks
+│   │   │   │   ├── runner.py        # FlashDet Trainer subprocess
+│   │   │   │   ├── controls.py      # Start/stop/pause/resume buttons
+│   │   │   │   └── dialogs.py       # Clean/resume/config dialogs
+│   │   │   └── monitor/             # Monitor sub-package
+│   │   │       ├── tab.py           # Monitor tab entry point
+│   │   │       ├── run_meta.py      # Run metadata extraction
+│   │   │       ├── parsers.py       # CSV/log parsing
+│   │   │       ├── dashboard.py     # Run dashboard + metrics
+│   │   │       ├── curves.py        # Plotly training curves
+│   │   │       ├── visualizations.py # Epoch visualizations
+│   │   │       ├── gt_verification.py # Ground truth verification
+│   │   │       ├── log_viewer.py    # Full log viewer
+│   │   │       └── checkpoints.py   # File browser + checkpoints
+│   │   ├── export/
+│   │   │   └── page.py              # Export page (ONNX/TorchScript)
+│   │   └── inference/
+│   │       ├── page.py              # Main inference page render
+│   │       ├── model_tab.py         # Model selection tab
+│   │       ├── data_tab.py          # Data input tab (images/video/RTSP)
+│   │       ├── solution_tab.py      # Solution selection + zone drawing
+│   │       ├── run_tab.py           # Run tab + results display
+│   │       ├── detection.py         # Detection utilities (real + demo)
+│   │       └── video.py             # Video/image inference runners
+│   │
+│   ├── components/
+│   │   ├── sidebar.py               # Navigation sidebar
+│   │   ├── styles.py                # Custom CSS + UI helpers
+│   │   ├── wizard.py                # Step indicator + navigation
+│   │   ├── project_manager.py       # Project CRUD + state persistence
+│   │   └── zone_drawer/             # Interactive canvas zone drawing
+│   │
+│   └── utils/
+│       ├── __init__.py              # Shared helpers (defaults, state, flash)
+│       ├── device.py                # GPU/environment detection
+│       ├── jobs.py                  # Background process tracking
+│       ├── filesystem.py            # Directory size/listing utilities
+│       ├── config.py                # Training config build/save/load
+│       └── training_hooks.py        # FlashDet training callbacks
+│
+├── tests/                           # Pytest test suite
+│   ├── conftest.py                  # Shared fixtures
+│   ├── test_constants.py            # Constants validation
+│   ├── test_cli.py                  # CLI tests
+│   ├── test_utils/                  # Utils sub-package tests
+│   │   ├── test_init.py
+│   │   ├── test_device.py
+│   │   ├── test_filesystem.py
+│   │   ├── test_config.py
+│   │   └── test_jobs.py
+│   ├── test_pages/                  # Page logic tests
+│   │   └── test_training_parsers.py
+│   └── test_components/
+│       └── test_project_manager.py
+│
+├── .github/workflows/ci.yml        # GitHub Actions CI
+├── notebooks/
+├── .streamlit/config.toml
+├── pyproject.toml
+└── README.md
+```
+
+---
+
+## Troubleshooting
+
+### Colab: "localhost can't be reached" or ngrok error
+
+`launch()` uses Colab's built-in proxy automatically. If it doesn't work, run this in a **new cell**:
+
+```python
+import subprocess, sys, time
+
+proc = subprocess.Popen(
+    [sys.executable, "-m", "streamlit", "run",
+     __import__('flashstudio.launcher', fromlist=['_get_app_path'])._get_app_path(),
+     "--server.port", "8501", "--server.headless", "true",
+     "--server.enableCORS", "false", "--server.enableXsrfProtection", "false"],
+    stdout=subprocess.PIPE, stderr=subprocess.PIPE
+)
+time.sleep(5)
+
+from google.colab.output import eval_js
+print(eval_js("google.colab.kernel.proxyPort(8501)"))
+```
+
+Open the printed URL in a new tab. No token or signup needed.
+
+### Streamlit port already in use
+
+```bash
+pkill -f "streamlit run"
+flashstudio --port 8501
+```
+
+---
+
+## Requirements
+
+- Python >= 3.10
+- GPU recommended for training (T4 or better)
+- Google Colab works out-of-the-box (no extra accounts needed)
+- FlashDet installs automatically with `pip install flashstudio`
+
+## License
+
+Apache-2.0

flashstudio-0.0.1/README_PYPI.md

@@ -0,0 +1,72 @@
+# FlashStudio
+
+**Interactive Training & Inference UI for FlashDet**
+
+---
+
+## Install
+
+```bash
+pip install flashstudio
+pip install git+https://github.com/FlashVision/FlashDet.git
+```
+
+## Quick Start
+
+```bash
+# Launch locally
+flashstudio
+
+# Or with Python
+from flashstudio import launch
+launch()
+```
+
+## Google Colab
+
+```python
+!pip install flashstudio
+!pip install git+https://github.com/FlashVision/FlashDet.git
+
+from flashstudio import launch
+launch(ngrok_token="YOUR_TOKEN")
+```
+
+Get a free ngrok token at: https://dashboard.ngrok.com/get-started/your-authtoken
+
+## Features
+
+- **Training** — Launch and monitor FlashDet training with live loss curves
+- **Inference** — Run detection on images/video with 17 built-in solutions
+- **Model Config** — FlashDet (Pico → X) + YOLOv8/v9/v10/v11/YOLOX
+- **Data** — Upload datasets or download from FlashDet registry
+- **Export** — ONNX export with FP16 support
+- **Dashboard** — Project overview with recent runs and metrics
+
+## Supported Models
+
+| Model | Params | Use Case |
+|-------|--------|----------|
+| FlashDet-Pico | ~298K | Edge / MCU |
+| FlashDet-Nano | ~790K | Embedded / IoT |
+| FlashDet-Small | ~1.8M | General purpose |
+| FlashDet-Medium | ~3.6M | High accuracy |
+| FlashDet-Large | ~5.8M | High accuracy |
+| FlashDet-X | ~9.0M | Server / Max accuracy |
+
+## Requirements
+
+- Python >= 3.10
+- FlashDet (for training/inference)
+- ngrok account (free, for Colab)
+- GPU recommended for training
+
+## Links
+
+- GitHub: https://github.com/FlashVision/FlashStudio
+- FlashDet: https://github.com/FlashVision/FlashDet
+- Organization: https://flashvision.github.io
+
+## License
+
+Apache-2.0

flashstudio-0.0.1/flashstudio/__init__.py

@@ -0,0 +1,5 @@
+"""FlashStudio — Interactive Training & Inference UI for FlashDet."""
+
+__version__ = "0.1.2"
+
+from flashstudio.launcher import launch  # noqa: F401