Implementation:Bentoml BentoML Framework EasyOCR
| Knowledge Sources | |
|---|---|
| Domains | ML Framework, OCR, Model Serialization |
| Last Updated | 2026-02-13 15:00 GMT |
Overview
Provides the BentoML framework integration for EasyOCR, enabling saving, loading, and serving EasyOCR Reader models through the BentoML model store.
Description
This module implements the standard BentoML framework adapter pattern for EasyOCR. It exposes four public functions: get(), load_model(), save_model(), and get_runnable().
save_model() serializes an easyocr.Reader instance using cloudpickle and stores it in the BentoML model store. Default model signatures include non-batchable methods (detect, readtext, readtextlang, recognize) and a batchable method (readtext_batched). The model context records the easyocr framework version.
load_model() retrieves the model from the store by tag and deserializes it with cloudpickle.load(), returning an easyocr.Reader instance.
get() retrieves the BentoML Model metadata object for a given tag, validating that it was saved with the EasyOCR module.
get_runnable() creates a dynamic EasyOCRRunnable class (subclass of bentoml.legacy.Runnable) that loads the model on initialization and maps each signature method to a runnable method with the appropriate batchable and batch_dim settings. The runnable supports both GPU and CPU resources.
Usage
Use this module to save pre-trained EasyOCR reader models to the BentoML model store and serve them via BentoML services. Access via bentoml.easyocr.
Code Reference
Source Location
- Repository: Bentoml_BentoML
- File: src/bentoml/_internal/frameworks/easyocr.py
- Lines: 1-208
Signature
def get(tag_like: str | Tag) -> Model: ...
def load_model(bento_model: str | Tag | Model) -> easyocr.Reader: ...
def save_model(
name: Tag | str,
reader: easyocr.Reader,
*,
signatures: ModelSignaturesType | None = None,
labels: dict[str, str] | None = None,
custom_objects: dict[str, t.Any] | None = None,
external_modules: t.List[ModuleType] | None = None,
metadata: dict[str, t.Any] | None = None,
) -> bentoml.Model: ...
def get_runnable(bento_model: bentoml.Model) -> type[bentoml.legacy.Runnable]: ...
Import
import bentoml
# Via the public API
model = bentoml.easyocr.save_model("reader", reader)
reader = bentoml.easyocr.load_model("reader:latest")
# Direct import
from bentoml._internal.frameworks.easyocr import save_model, load_model, get, get_runnable
I/O Contract
Inputs
| Name | Type | Required | Description |
|---|---|---|---|
| name | Tag or str | Yes (save_model) | Name or tag for the model in the store |
| reader | easyocr.Reader | Yes (save_model) | The EasyOCR Reader instance to save |
| tag_like | str or Tag | Yes (get) | Tag to retrieve from the model store |
| bento_model | str, Tag, or Model | Yes (load_model) | Model identifier or object to load |
| signatures | ModelSignaturesType or None | No | Method signatures; defaults to detect, readtext, readtextlang, recognize (non-batchable) and readtext_batched (batchable) |
| labels | dict[str, str] or None | No | User-defined labels for model management |
| custom_objects | dict[str, Any] or None | No | Additional objects to save with the model |
| external_modules | List[ModuleType] or None | No | Additional Python modules to bundle |
| metadata | dict[str, Any] or None | No | Custom metadata for the model |
Outputs
| Name | Type | Description |
|---|---|---|
| bentoml.Model (save_model) | bentoml.Model | The saved model reference in the BentoML store |
| easyocr.Reader (load_model) | easyocr.Reader | The deserialized EasyOCR reader instance |
| Model (get) | Model | BentoML Model metadata object |
| EasyOCRRunnable (get_runnable) | type[Runnable] | A Runnable class for serving the model |
Usage Examples
import bentoml
import easyocr
# Create and save an EasyOCR reader
reader = easyocr.Reader(['en'])
bento_model = bentoml.easyocr.save_model('en_reader', reader)
print(f"Saved: {bento_model.tag}")
# Load the model back
loaded_reader = bentoml.easyocr.load_model('en_reader:latest')
results = loaded_reader.readtext('image.png')
# Retrieve model metadata
model = bentoml.easyocr.get('en_reader:latest')
print(model.info.signatures)
# Create a runnable for serving
runnable_cls = bentoml.easyocr.get_runnable(model)