​
Uploading Pre-annotations

Pre-annotations are labels which can be pre-loaded to a project which can speed up manual annotation. Size limit for uploading pre-annotations is 25 MB.

​
Example Usage (Synchronous):

Synchronous Pre-annotation Upload
from labellerr.client import LabellerrClient
from labellerr.exceptions import LabellerrError

# Initialize the client with your API credentials
client = LabellerrClient('your_api_key', 'your_api_secret')
project_id = 'project_123'
client_id = '12345'
annotation_format = 'coco'
annotation_file = '/path/to/annotations.json'

try:
    # Upload and wait for processing to complete
    result = client.upload_preannotation_by_project_id(project_id, client_id, annotation_format, annotation_file)
    # Check the final status
    if result['response']['status'] == 'completed':
        print("Pre-annotations processed successfully")
        # Access additional metadata if needed
        metadata = result['response'].get('metadata', {})
        print("metadata",metadata)
except LabellerrError as e:
    print(f"Pre-annotation upload failed: {str(e)}")

​
Example Usage (Asynchronous):

Asynchronous Pre-annotation Upload
from labellerr.client import LabellerrClient
from labellerr.exceptions import LabellerrError

# Initialize the client with your API credentials
client = LabellerrClient('your_api_key', 'your_api_secret')
project_id = 'project_123'
client_id = '12345'
annotation_format = 'coco'
annotation_file = '/path/to/annotations.json'

try:
    # Start the async upload - returns immediately
    future = client.upload_preannotation_by_project_id_async(project_id, client_id, annotation_format, annotation_file)
    print("Upload started, you can do other work here...")
    # When you need the result, wait for completion
    try:
        result = future.result(timeout=300)  # 5 minutes timeout
        if result['response']['status'] == 'completed':
            print("Pre-annotations processed successfully")
            metadata = result['response'].get('metadata', {})
            print("metadata",metadata)
    except TimeoutError:
        print("Processing took too long")
    except Exception as e:
        print(f"Error in processing: {str(e)}")
except LabellerrError as e:
    print(f"Failed to start upload: {str(e)}")

​
Choosing Between Sync and Async

  1. Synchronous Method:
    1. Simpler to use - just call and wait for result
    2. Blocks until processing is complete
    3. Good for scripts and sequential processing
  2. Asynchronous Method:
    1. Returns immediately with a Future object
    2. Allows you to do other work while processing
    3. Can set timeouts and handle long-running uploads
    4. Better for applications that need to stay responsive
Both methods will:
  1. Upload your annotation file
  2. Monitor the processing status
  3. Return the final result with complete status information
Note: The processing time depends on the size of your annotation file and the number of annotations. For the sync method, this means waiting time. For the async method, you can do other work during this time.