Implementation done

.gitignore

@@ -0,0 +1,2 @@
+trocr-env/
+.env

Dockerfile

@@ -0,0 +1,21 @@
+# read the doc: https://huggingface.co/docs/hub/spaces-sdks-docker
+# you will also find guides on how best to write your Dockerfile
+
+FROM python:3.10
+
+WORKDIR /code
+
+COPY ./requirements.txt /code/requirements.txt
+
+RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt
+
+RUN useradd -m -u 1000 user
+USER user
+ENV HOME=/home/user \
+	PATH=/home/user/.local/bin:$PATH
+
+WORKDIR $HOME/app
+
+COPY --chown=user . $HOME/app
+
+CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]

config.py

@@ -0,0 +1,8 @@
+from pydantic_settings import BaseSettings, SettingsConfigDict
+from pydantic import Field
+
+class Settings(BaseSettings):
+    model_config = SettingsConfigDict(env_file='.env', env_file_encoding='utf-8')
+    OCR_MODEL: str
+    TOKENIZER: str
+    FEATURE_EXTRACTOR: str

main.py

@@ -0,0 +1,47 @@
+from config import Settings
+import torch
+from PIL import Image
+import io
+from contextlib import asynccontextmanager
+from transformers import VisionEncoderDecoderModel
+from fastapi import FastAPI, UploadFile, Form, HTTPException
+from transformers import TrOCRProcessor, AutoTokenizer, ViTImageProcessor
+
+config = {}
+
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    settings = Settings()
+    config['settings'] = settings
+    config['device'] = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+    tokenizer = AutoTokenizer.from_pretrained(settings.TOKENIZER)
+    feature_extractor = ViTImageProcessor.from_pretrained(settings.FEATURE_EXTRACTOR)
+    config['processor'] = TrOCRProcessor(image_processor=feature_extractor, tokenizer=tokenizer)
+    config['ocr_model'] = VisionEncoderDecoderModel.from_pretrained(settings.OCR_MODEL)
+    
+    yield
+    # Clean up and release the resources
+    config.clear()
+
+app = FastAPI(lifespan=lifespan)
+
+@app.get("/")
+def api_home():
+    return {'detail': 'Welcome to Sinhala OCR Space'}
+
+app.post("/apply-trocr")
+async def ApplyOCR(file: UploadFile):
+    try:
+        # Read the uploaded image file
+        contents = await file.read()
+        image = Image.open(io.BytesIO(contents))
+
+        pixel_values = config['processor'](image, return_tensors="pt").pixel_values
+        generated_ids = config['ocr_model'].generate(pixel_values)
+        generated_text =  config['processor'].batch_decode(generated_ids, skip_special_tokens=True)[0]
+
+        # Return the extracted text as the response
+        return {"ocr_result": generated_text}
+    except Exception as e:
+        # Handle any exceptions that may occur
+        return {"error": str(e)}