push fix
Browse files
app.py
CHANGED
|
@@ -3,12 +3,12 @@ import json
|
|
| 3 |
import os
|
| 4 |
import tempfile
|
| 5 |
import io
|
|
|
|
| 6 |
from pathlib import Path
|
| 7 |
from typing import Optional, Tuple
|
| 8 |
from fastapi import FastAPI, UploadFile, File, HTTPException
|
| 9 |
from fastapi.responses import JSONResponse
|
| 10 |
from transformers import AutoTokenizer, AutoModelForCausalLM
|
| 11 |
-
from pdfminer.high_level import extract_text as extract_pdf_text
|
| 12 |
from docx import Document as DocxDocument
|
| 13 |
from pptx import Presentation
|
| 14 |
import logging
|
|
@@ -17,9 +17,14 @@ import pytesseract
|
|
| 17 |
from pdf2image import convert_from_path
|
| 18 |
import easyocr
|
| 19 |
|
|
|
|
|
|
|
|
|
|
| 20 |
logging.basicConfig(level=logging.INFO)
|
| 21 |
logger = logging.getLogger(__name__)
|
| 22 |
|
|
|
|
|
|
|
| 23 |
app = FastAPI(
|
| 24 |
title="Deckgpt",
|
| 25 |
description="Upload your startup pitch deck (PDF, PPT, DOCX) and get an investor-style review",
|
|
@@ -49,13 +54,22 @@ async def load_model():
|
|
| 49 |
try:
|
| 50 |
ocr_model_dir = os.path.join(tempfile.gettempdir(), '.EasyOCR')
|
| 51 |
os.makedirs(ocr_model_dir, exist_ok=True)
|
|
|
|
| 52 |
|
| 53 |
-
|
| 54 |
-
|
| 55 |
-
|
| 56 |
-
|
| 57 |
-
|
| 58 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 59 |
except Exception as e:
|
| 60 |
logger.warning(f"⚠️ EasyOCR failed to load, will use pytesseract fallback: {e}")
|
| 61 |
ocr_reader = None
|
|
|
|
| 3 |
import os
|
| 4 |
import tempfile
|
| 5 |
import io
|
| 6 |
+
import warnings
|
| 7 |
from pathlib import Path
|
| 8 |
from typing import Optional, Tuple
|
| 9 |
from fastapi import FastAPI, UploadFile, File, HTTPException
|
| 10 |
from fastapi.responses import JSONResponse
|
| 11 |
from transformers import AutoTokenizer, AutoModelForCausalLM
|
|
|
|
| 12 |
from docx import Document as DocxDocument
|
| 13 |
from pptx import Presentation
|
| 14 |
import logging
|
|
|
|
| 17 |
from pdf2image import convert_from_path
|
| 18 |
import easyocr
|
| 19 |
|
| 20 |
+
warnings.filterwarnings("ignore", category=UserWarning, module="pdfminer")
|
| 21 |
+
warnings.filterwarnings("ignore", category=FutureWarning, module="transformers")
|
| 22 |
+
|
| 23 |
logging.basicConfig(level=logging.INFO)
|
| 24 |
logger = logging.getLogger(__name__)
|
| 25 |
|
| 26 |
+
from pdfminer.high_level import extract_text as extract_pdf_text
|
| 27 |
+
|
| 28 |
app = FastAPI(
|
| 29 |
title="Deckgpt",
|
| 30 |
description="Upload your startup pitch deck (PDF, PPT, DOCX) and get an investor-style review",
|
|
|
|
| 54 |
try:
|
| 55 |
ocr_model_dir = os.path.join(tempfile.gettempdir(), '.EasyOCR')
|
| 56 |
os.makedirs(ocr_model_dir, exist_ok=True)
|
| 57 |
+
os.chmod(ocr_model_dir, 0o777)
|
| 58 |
|
| 59 |
+
old_home = os.environ.get('HOME')
|
| 60 |
+
try:
|
| 61 |
+
os.environ['HOME'] = tempfile.gettempdir()
|
| 62 |
+
ocr_reader = easyocr.Reader(
|
| 63 |
+
['en'],
|
| 64 |
+
gpu=torch.cuda.is_available(),
|
| 65 |
+
model_storage_directory=ocr_model_dir
|
| 66 |
+
)
|
| 67 |
+
logger.info("✅ OCR reader loaded successfully!")
|
| 68 |
+
finally:
|
| 69 |
+
if old_home:
|
| 70 |
+
os.environ['HOME'] = old_home
|
| 71 |
+
elif 'HOME' in os.environ:
|
| 72 |
+
del os.environ['HOME']
|
| 73 |
except Exception as e:
|
| 74 |
logger.warning(f"⚠️ EasyOCR failed to load, will use pytesseract fallback: {e}")
|
| 75 |
ocr_reader = None
|