### Install Dependencies from Source

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/segment_anything_2_image/README.md

Clone the repository, install the ML backend, and then install specific requirements for the segment anything example. Ensure you are in the correct directory before installing.

```bash
git clone https://github.com/HumanSignal/label-studio-ml-backend.git
cd label-studio-ml-backend
pip install -e .
cd label_studio_ml/examples/segment_anything_2_image
pip install -r requirements.txt
```

--------------------------------

### Install Dependencies for SAM2 Video Example

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/segment_anything_2_video/README.md

Clone the repository and install necessary Python packages for the SAM2 video example. This includes the main library and specific requirements for the example.

```bash
git clone https://github.com/HumanSignal/label-studio-ml-backend.git
cd label-studio-ml-backend
pip install -e .
cd label_studio_ml/examples/segment_anything_2_video
pip install -r requirements.txt
```

--------------------------------

### Start ML Backend from Source

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/nemo_asr/README.md

Run the Label Studio ML backend from source after installing dependencies, specifying the example configuration to use.

```bash
label-studio-ml start ./nemo_asr
```

--------------------------------

### Start ML Backend from Source

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/default_configs/README.md

After setting up the environment and installing dependencies, start the ML backend using this command, specifying the directory containing your models.

```bash
label-studio-ml start ./dir_with_your_model
```

--------------------------------

### Install Dependencies and Run ML Backend

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/huggingface_ner/README.md

Steps to set up a Python virtual environment, install dependencies, and start the ML backend from source without Docker.

```bash
python -m venv ml-backend
source ml-backend/bin/activate
pip install -r requirements.txt
```

```bash
label-studio-ml start ./huggingface_ner
```

--------------------------------

### Start EasyOCR ML Backend

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/easyocr/README.md

Run this command to start the EasyOCR ML backend after installing dependencies.

```bash
label-studio-ml start ./easyocr
```

--------------------------------

### Start ML Backend from Source

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/segment_anything_2_image/README.md

After cloning the SAM2 repository and installing dependencies, start the ML backend. This command must be run from the `~/sam2` directory due to specific path requirements.

```bash
cd ~/sam2
label-studio-ml start ../label-studio-ml-backend/label_studio_ml/examples/segment_anything_2_image
```

--------------------------------

### Start ML Backend with Docker

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/default_configs/README.md

Use this command to start the ML backend using a prebuilt Docker image. Ensure Docker is installed and running.

```bash
docker-compose up
```

--------------------------------

### Clone and Install SAM2

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/segment_anything_video_interactive/README.md

Clone the Segment Anything 2 repository next to the example and install its dependencies.

```bash
cd label_studio_ml/examples/segment_anything_video_interactive
git clone https://github.com/facebookresearch/segment-anything-2.git
(cd segment-anything-2 && pip install -e . && cd checkpoints && ./download_ckpts.sh)
```

--------------------------------

### Install Label Studio and Clone Repository

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/yolo/README.md

Install Label Studio using pip and clone the ML backend repository to access the YOLO example.

```bash
pip install label-studio
label-studio

```

```bash
git clone https://github.com/HumanSignal/label-studio-ml-backend.git
cd label-studio-ml-backend/examples/yolo

```

--------------------------------

### Install Dependencies and Run ML Backend

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/gliner/README.md

This sequence of commands installs the ML backend dependencies in a virtual environment and then starts the server. This is an advanced option for running without Docker.

```bash
python -m venv ml-backend
source ml-backend/bin/activate
pip install -r requirements.txt
label-studio-ml start ./dir_with_your_model
```

--------------------------------

### Start ML Backend for SAM2 Video

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/segment_anything_2_video/README.md

Run the ML backend from source, pointing it to the segment_anything_2_video example directory. Ensure environment variables for Label Studio URL and API key are set.

```bash
cd ../
label-studio-ml start ./segment_anything_2_video
```

--------------------------------

### Start PP-OCR Server (label-studio-ml)

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/ppocr/README.md

Starts the PP-OCR server using the label-studio-ml command-line tool.

```bash
label-studio-ml start ./ppocr
```

--------------------------------

### Set Up Virtual Environment and Install Dependencies

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/default_configs/README.md

Run the ML backend without Docker by cloning the repository, creating a virtual environment, and installing dependencies using pip.

```bash
python -m venv ml-backend
source ml-backend/bin/activate
pip install -r requirements.txt
```

--------------------------------

### Install Dependencies and Run ML Backend

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/interactive_substring_matching/README.md

Install ML backend dependencies using pip within a virtual environment and then start the server. This is an advanced option for running without Docker.

```bash
python -m venv ml-backend
source ml-backend/bin/activate
pip install -r requirements.txt
label-studio-ml start ./interactive_substring_matching
```

--------------------------------

### Start Local Dev Server

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/segment_anything_video_interactive/README.md

Start the development server locally, setting environment variables for the device and Label Studio connection.

```bash
cd label_studio_ml/examples/segment_anything_video_interactive
export DEVICE=mps            # on Apple Silicon, or keep cuda on NVIDIA
export LABEL_STUDIO_URL=http://host.docker.internal:8080
export LABEL_STUDIO_API_KEY=<your LSE API key>
python _wsgi.py -p 9090
```

--------------------------------

### Install Dependencies and Run ML Backend without Docker

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/huggingface_llm/README.md

Install the necessary Python dependencies and start the ML backend from source if you are not using Docker.

```bash
python -m venv ml-backend
source ml-backend/bin/activate
pip install -r requirements.txt
```

```bash
label-studio-ml start ./huggingface_llm
```

--------------------------------

### Install Label Studio with Docker

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/tesseract/README.md

Launches Label Studio using Docker. Ensure git and Docker Compose are installed beforehand.

```bash
docker run -it \
   -p 8080:8080 \
   -v `pwd`/mydata:/label-studio/data \
   heartexlabs/label-studio:latest
```

--------------------------------

### Install Label Studio ML Backend

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/README.md

Clone the repository and install the label-studio-ml package in editable mode.

```bash
git clone https://github.com/HumanSignal/label-studio-ml-backend.git
cd label-studio-ml-backend/
pip install -e .
```

--------------------------------

### Quick Docker Setup for MMDetection

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/mmdetection-3/README.md

Use this command to quickly start the MMDetection model backend with Docker Compose.

```bash
docker-compose up -d
```

--------------------------------

### Start ML Backend Server with Docker Compose

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/README.md

Use this command to clone the repository and start serving the ML backend. Replace `{MODEL_NAME}` with the desired model.

```bash
git clone https://github.com/HumanSignal/label-studio-ml-backend.git
cd label-studio-ml-backend/label_studio_ml/examples/{MODEL_NAME}
docker-compose up
```

--------------------------------

### Start ML Backend with label-studio-ml CLI

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/mmdetection-3/README.md

Run this command to start the ML backend using the 'label-studio-ml' command-line interface, specifying the root directory and the model configuration.

```bash
label-studio-ml start --root-dir .. mmdetection-3
```

--------------------------------

### Install Project Requirements

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/ppocr/README.md

Installs the specific dependencies for the project.

```bash
pip install -r requirements.txt
```

--------------------------------

### Start ML Backend Without Docker

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/README.md

Run the ML backend directly using the label-studio-ml start command for debugging or development.

```bash
label-studio-ml start my_ml_backend
```

--------------------------------

### Install Backend Requirements

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/segment_anything_video_interactive/README.md

Install the necessary Python packages for the Label Studio ML backend from the repository root.

```bash
pip install -e .
pip install -r label_studio_ml/examples/segment_anything_video_interactive/requirements.txt
```

--------------------------------

### Start PP-OCR Server (WSGI)

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/ppocr/README.md

Starts the PP-OCR server using the WSGI interface.

```bash
python _wsgi.py
```

--------------------------------

### Start ML Backend on a Custom Port

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/README.md

Specify a custom port for the ML backend using the -p parameter with the start command.

```bash
label-studio-ml start my_ml_backend -p 9091
```

--------------------------------

### Install Base Requirements

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/ppocr/README.md

Installs the base dependencies required for the project.

```bash
pip install -r requirements-base.txt
```

--------------------------------

### Clone Label Studio ML Backend Repository

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/mmdetection-3/README.md

Clone the repository to access the example MMDetection setup and other ML backend functionalities.

```bash
git clone https://github.com/heartexlabs/label-studio-ml-backend
cd label-studio-ml-backend/label_studio_ml/examples/mmdetection-3
```

--------------------------------

### Install PaddleOCR

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/ppocr/README.md

Installs the PaddleOCR library after PaddlePaddle has been installed.

```bash
python -m pip install paddleocr
```

--------------------------------

### Install Test Requirements

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/ppocr/README.md

Installs the dependencies required for running unit tests.

```bash
pip install -r requirements-test.txt
```

--------------------------------

### Install Dependencies and Set Environment Variables

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/docling/README.md

Installs necessary Python packages and sets environment variables for running the Docling service locally. Adjust LABEL_STUDIO_URL if Label Studio is not running in Docker.

```bash
pip install -r requirements-base.txt -r requirements.txt
export DOCLING_SERVICE_URL=https://api.aws-c1.dcls.saas.ibm.com/your-instance/v1
export DOCLING_SERVE_API_KEY=your-api-key
export LABEL_STUDIO_URL=http://host.docker.internal:8080
export LABEL_STUDIO_API_KEY=your-label-studio-token
python _wsgi.py -p 9090
```

--------------------------------

### Build and Run ML Backend

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/timeseries_segmenter/README.md

Use this command to build the Docker image and start the ML backend service.

```bash
docker-compose up --build
```

--------------------------------

### Start ML Backend with MobileSAM Checkpoint

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/segment_anything_model/README.md

Start the ML backend, specifying the path to the MobileSAM checkpoint file using an environment variable. This command is used for manual backend startup with a specific model.

```bash
MOBILESAM_CHECKPOINT=path/to/mobile_sam.pt label-studio-ml start segment_anything_model/
```

--------------------------------

### Labeling Interface Example

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/langchain_search_agent/README.md

An example Label Studio labeling interface configuration that includes input fields for text, prompt, response, search snippets, and classification choices.

```xml
<View>
    <Style>
        .lsf-main-content.lsf-requesting .prompt::before { content: ' loading...'; color: #808080; }
    </Style>
    <Text name="input" value="$text"/>
    <View className="prompt">
        <TextArea name="prompt" toName="input" maxSubmissions="1" editable="true"/>
    </View>
    <TextArea name="response" toName="input" maxSubmissions="1" editable="true"/>
    <TextArea name="snippets" toName="input"/>
    <Choices name="classification" toName="input" choice="single" showInLine="true">
        <Choice value="Good"/>
        <Choice value="Bad"/>
    </Choices>
</View>
```

--------------------------------

### Create and Activate Virtual Environment

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/mmdetection-3/README.md

Set up a Python virtual environment for managing dependencies and activate it before installing packages.

```bash
python -m venv ml-backend
source ml-backend/bin/activate
```

--------------------------------

### Start Tesseract and MinIO Servers with Docker Compose

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/tesseract/README.md

Starts the Tesseract OCR and MinIO servers using Docker Compose. This command assumes you have a `docker-compose.yml` file configured.

```bash
docker compose up
```

--------------------------------

### Custom Tracker Parameters Example

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/yolo/README.md

This example shows how to pass custom parameters to a tracker, such as BoT-SORT, directly from the labeling configuration. Parameters are prefixed with the tracker name (e.g., `botsort_`).

```xml
<VideoRectangle 
    name="label" toName="video" 
    model_tracker="botsort" 
    botsort_track_low_thresh="0.1" botsort_track_high_thresh="0.2" 
/>
```

--------------------------------

### Install Dependencies for ML Backend (No Docker)

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/yolo/README.md

Install the necessary Python dependencies for running the ML backend without Docker. This involves creating a virtual environment and using pip.

```bash
python -m venv ml-backend
source ml-backend/bin/activate
pip install -r requirements-base.txt
pip install -r requirements.txt
```

--------------------------------

### Install System and Python Dependencies

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/mmdetection-3/README.md

Install necessary system packages and Python libraries, including PyTorch, MMDetection, and their dependencies. Ensure CUDA and cuDNN versions are set correctly if using GPU.

```bash
# Set environment variables
export PYTORCH="1.13.0"
export CUDA="11.6"
export CUDNN="8"

# To fix GPG key error when running apt-get update
apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/3bf863cc.pub
apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64/7fa2af80.pub

# Update the base OS
apt-get update
apt install -y git

# Install system dependencies for opencv-python
apt-get install -y libgl1 libglib2.0-0

# Upgrade pip
pip install -U pip

# Install base requirements
pip install -r requirements-base.txt

# Install custom requirements
pip install -r requirements.txt

# Mim install commands
pip install mim
mim install mmengine==0.10.3
mim install mmdet==3.3.0
mim download mmdet --config yolov3_mobilenetv2_320-300e_coco --dest .
```

--------------------------------

### Task Data JSON File Example

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/yolo/README.md

Example of a JSON file containing task objects with IDs and data. This format is also compatible with the `--tasks` argument of the CLI.

```json
[{"id": 1, "data": {"image": "https://example.com/1.jpg"}}, {"id": 2, "data": {"image": "https://example.com/2.jpg"}}]
```

--------------------------------

### Start Korean OCR with CPU

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/ppocr/README.md

Launches the PP-OCR service for Korean OCR, explicitly using the CPU for processing.

```bash
OCR_LANG=korean docker-compose up -d
```

--------------------------------

### Start Chinese OCR (Default)

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/ppocr/README.md

Launches the PP-OCR service with default settings for Chinese OCR using the server model variant.

```bash
MODEL_VARIANT=server OCR_LANG=ch docker-compose up -d
```

--------------------------------

### Clone Label Studio ML Backend Repository

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/tesseract/README.md

Clones the Label Studio ML backend repository and navigates into the Tesseract example directory. Ensure git is installed.

```bash
git clone https://github.com/humansignal/label-studio-ml-backend
cd label-studio-ml-backend/label_studio_ml/examples/tesseract
```

--------------------------------

### Run Label Studio with Docker and Local File Serving

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/segment_anything_model/README.md

Launch Label Studio using Docker, mounting a local data volume and enabling local file serving. This example demonstrates how to map local directories and set environment variables for file serving.

```bash
docker run -it -p 8080:8080 \
               -v $(pwd)/mydata:/label-studio/data \
               --env LABEL_STUDIO_LOCAL_FILES_SERVING_ENABLED=true \
               --env LABEL_STUDIO_LOCAL_FILES_DOCUMENT_ROOT=/label-studio/data/images \
               heartexlabs/label-studio:latest
```

--------------------------------

### LLM Prompt for YOLO Label Mapping

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/yolo/README.md

An example prompt to guide an LLM in automatically generating a mapping between Label Studio labels and YOLO model labels. This is useful for large numbers of labels or complex naming conventions.

```text
**Task:**

1. **ML Model Labels:**
   - I have the following labels in my ML model:
['tench', 'goldfish', 'great_white_shark', 'tiger_shark', 'hammerhead', 'electric_ray', 'stingray', 'cock', 'hen', 'ostrich', 'brambling', 'goldfinch', 'house_finch', 'junco', 'indigo_bunting', 'robin', 'bulbul', 'jay', 'magpie', 'chickadee', 'water_ouzel', 'kite', 'bald_eagle', 'vulture', 'great_grey_owl', 'European_fire_salamander', 'common_newt', 'eft', 'spotted_salamander', 'axolotl', 'bullfrog', 'tree_frog', 'tailed_frog', 'loggerhead', 'leatherback_turtle', 'mud_turtle', 'terrapin', 'box_turtle', 'banded_gecko', 'common_iguana', 'American_chameleon', 'whiptail', 'agama', 'frilled_lizard', 'alligator_lizard', 'Gila_monster', 'green_lizard', 'African_chameleon', 'Komodo_dragon', 'African_crocodile', 'American_alligator', 'triceratops', 'thunder_snake', 'ringneck_snake', 'hognose_snake', 'green_snake', 'king_snake', 'garter_snake', 'water_snake', 'vine_snake', 'night_snake', 'boa_constrictor', 'rock_python', 'Indian_cobra', 'green_mamba', 'sea_snake', 'horned_viper', 'diamondback', 'sidewinder', 'trilobite', 'harvestman', 'scorpion', 'black_and_gold_garden_spider', 'barn_spider', 'garden_spider', 'black_widow', 'tarantula', 'wolf_spider', 'tick', 'centipede', 'black_grouse', 'ptarmigan', 'ruffed_grouse', 'prairie_chicken', 'peacock', 'quail', 'partridge', 'African_grey', 'macaw', 'sulphur-crested_cockatoo', 'lorikeet', 'coucal', 'bee_eater', 'hornbill', 'hummingbird', 'jacamar', 'toucan', 'drake', 'red-breasted_merganser', 'goose', 'black_swan', 'tusker', 'echidna', 'platypus', 'wallaby', 'koala', 'wombat', 'jellyfish', 'sea_anemone', 'brain_coral', 'flatworm', 'nematode', 'conch', 'snail', 'slug', 'sea_slug', 'chiton', 'chambered_nautilus', 'Dungeness_crab', 'rock_crab', 'fiddler_crab', 'king_crab', 'American_lobster', 'spiny_lobster', 'crayfish', 'hermit_crab', 'isopod', 'white_stork', 'black_stork', 'spoonbill', 'flamingo', 'little_blue_heron', 'American_egret', 'bittern', 'crane_(bird)', 'limpkin', 'European_gallinule', 'American_coot', 'bustard', 'ruddy_turnstone', 'red-backed_sandpiper', 'redshank', 'dowitcher', 'oystercatcher', 'pelican', 'king_penguin', 'albatross', 'grey_whale', 'killer_whale', 'dugong', 'sea_lion', 'Chihuahua', 'Japanese_spaniel', 'Maltese_dog', 'Pekinese', 'Shih-Tzu', 'Blenheim_spaniel', 'papillon', 'toy_terrier', 'Rhodesian_ridgeback', 'Afghan_hound', 'basset', 'beagle', 'bloodhound', 'bluetick', 'black-and-tan_coonhound', 'Walker_hound', 'English_foxhound', 'redbone', 'borzoi', 'Irish_wolfhound', 'Italian_greyhound', 'whippet', 'Ibizan_hound', 'Norwegian_elkhound', 'otterhound', 'Saluki', 'Scottish_deerhound', 'Weimaraner', 'Staffordshire_bullterrier', 'American_Staffordshire_terrier', 'Bedlington_terrier', 'Border_terrier', 'Kerry_blue_terrier', 'Irish_terrier', 'Norfolk_terrier', 'Norwich_terrier', 'Yorkshire_terrier', 'wire-haired_fox_terrier', 'Lakeland_terrier', 'Sealyham_terrier', 'Airedale', 'cairn', 'Australian_terrier', 'Dandie_Dinmont', 'Boston_bull', 'miniature_schnauzer', 'giant_schnauzer', 'standard_schnauzer', 'Scotch_terrier', 'Tibetan_terrier', 'silky_terrier', 'soft-coated_wheaten_terrier', 'West_Highland_white_terrier', 'Lhasa', 'flat-coated_retriever', 'curly-coated_retriever', 'golden_retriever', 'Labrador_retriever', 'Chesapeake_Bay_retriever', 'German_short-haired_pointer', 'vizsla', 'English_setter', 'Irish_setter', 'Gordon_setter', 'Brittany_spaniel', 'clumber', 'English_springer', 'Welsh_springer_spaniel', 'cocker_spaniel', 'Sussex_spaniel', 'Irish_water_spaniel', 'kuvasz', 'schipperke', 'groenendael', 'malinois', 'briard', 'kelpie', 'komondor', 'Old_English_sheepdog', 'Shetland_sheepdog', 'collie', 'Border_collie', 'Bouvier_de']

2. **Label Studio Labels:**
   - I have the following labels in my Label Studio project:
['Car', 'Truck', 'Motorcycle', 'Bicycle', 'Pedestrian', 'Traffic Light', 'Stop Sign']

3. **Mapping:**
   - Create a JSON object where keys are my Label Studio labels and values are the most relevant ML model labels from the provided list. If a direct match is not obvious, choose the closest semantic match. For example, 'Car' could map to 'automobile' or 'sedan' if available, or a more general term like 'vehicle'.

**Output Format:**
```json
{
  "Label Studio Label": "ML Model Label",
  "Another Label": "Another ML Model Label"
}
```
```

--------------------------------

### Create a New ML Backend Project

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/README.md

Use the label-studio-ml create command to scaffold a new ML backend project.

```bash
label-studio-ml create my_ml_backend
```

--------------------------------

### Install Label Studio with Local File Serving Enabled

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/tesseract/README.md

Launches Label Studio with local file serving enabled. This requires setting environment variables for the image root directory.

```bash
docker run -it \
   -p 8080:8080 \
   -v `pwd`/mydata:/label-studio/data \
   --env LABEL_STUDIO_LOCAL_FILES_SERVING_ENABLED=true \
   --env LABEL_STUDIO_LOCAL_FILES_DOCUMENT_ROOT=/label-studio/data/images \
   heartexlabs/label-studio:latest
```

--------------------------------

### Install PaddlePaddle CPU Version

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/ppocr/README.md

Installs the CPU version of PaddlePaddle, a deep learning framework required for PP-OCR.

```bash
# CPU version
python -m pip install paddlepaddle==3.2.0 -i https://www.paddlepaddle.org.cn/packages/stable/cpu/
```

--------------------------------

### Clone ML Backend Repository

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/segment_anything_model/README.md

Clone the label-studio-ml-backend repository and navigate to the segment anything model example directory. This is the first step for setting up the SAM backend manually or with Docker Compose.

```bash
git clone https://github.com/humansignal/label-studio-ml-backend
cd label-studio-ml-backend/label_studio_ml/examples/segment_anything_model
```

--------------------------------

### YOLO Object Detection Model Configuration Example

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/yolo/README.md

Example of configuring RectangleLabels for YOLO object detection, including model_score_threshold and model_path.

```xml
<RectangleLabels name="label" toName="image" model_score_threshold="0.25" model_path="my_model.pt">

```

--------------------------------

### Docker Compose Up

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/docling/README.md

Use this command to build and start the Docling ML backend service using Docker Compose. Ensure environment variables are set in your docker-compose.yml file.

```bash
cd label_studio_ml/examples/docling
# Set DOCLING_SERVICE_URL, DOCLING_SERVE_API_KEY, LABEL_STUDIO_URL, LABEL_STUDIO_API_KEY in docker-compose.yml
docker compose up --build
```

--------------------------------

### Image Document Data Input Example

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/llm_interactive/README.md

Example JSON data structure for an image document, specifying the URL of the image to be analyzed.

```json
{
  "image": "https://sandbox2-test-bucket.s3.amazonaws.com/receipts/113494_page1.png"
}
```

--------------------------------

### LLM Dialogue Data Input Example

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/llm_interactive/README.md

Example JSON data structure for an LLM dialogue, suitable for use with the Paragraphs tag in Label Studio.

```json
{
  "dialogue": [
    {
      "role": "user",
      "content": "What is the capital of France?"
    },
    {
      "role": "assistant",
      "content": "The capital of France is Paris."
    },
    {
      "role": "user",
      "content": "Tell me a joke."
    }
  ]
}
```

--------------------------------

### Install PaddlePaddle GPU Version (CUDA 12.6)

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/ppocr/README.md

Installs the GPU version of PaddlePaddle for CUDA 12.6. Ensure your NVIDIA drivers are version 550.54.14 or higher.

```bash
# GPU version (CUDA 12.6, requires driver >= 550.54.14)
python -m pip install paddlepaddle-gpu==3.2.0 -i https://www.paddlepaddle.org.cn/packages/stable/cu126/
```

--------------------------------

### Run Backend with Docker Compose

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/segment_anything_model/README.md

Use Docker Compose to build and run the ML backend. Ensure you have edited the `docker-compose.yml` file to include your Label Studio host and access token.

```bash
docker compose up --build
```

--------------------------------

### Start ML Backend (GPU Version - CUDA 11.8)

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/ppocr/README.md

Deploy the ML backend with PP-OCR support on a GPU using CUDA 11.8. This command utilizes a specific Docker Compose file for GPU acceleration.

```bash
docker-compose -f docker-compose.gpu.yml up -d
```

--------------------------------

### Install PaddlePaddle GPU Version (CUDA 11.8)

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/ppocr/README.md

Installs the GPU version of PaddlePaddle for CUDA 11.8. Ensure your NVIDIA drivers are version 450.80.02 or higher.

```bash
# GPU version (CUDA 11.8, requires driver >= 450.80.02)
python -m pip install paddlepaddle-gpu==3.2.0 -i https://www.paddlepaddle.org.cn/packages/stable/cu118/
```

--------------------------------

### Task ID JSON File Example

Source: https://github.com/humansignal/label-studio-ml-backend/blob/master/label_studio_ml/examples/yolo/README.md

Example of a JSON file containing a list of task IDs. This format can be used with the `--tasks` argument of the CLI.

```json
[1,2,3]
```