Update app.py

app.py

@@ -25,13 +25,13 @@ app = FastAPI()
 MODEL_NAME = "facebook/bart-large-cnn"
 tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
 model = AutoModelForSeq2SeqLM.from_pretrained(MODEL_NAME)
-summarizer = pipeline("summarization", model=model, tokenizer=tokenizer, device=-1)
+summarizer = pipeline("summarization", model=model, tokenizer=tokenizer, device=-1, batch_size=4)
 
-reader = easyocr.Reader(['en'])
+reader = easyocr.Reader(['en'], gpu=torch.cuda.is_available())
 
 def clean_text(text: str) -> str:
     text = re.sub(r'\s+', ' ', text)
-    text = re.sub(r'•\s*|\d\.\s+', '', text)
+    text = re.sub(r'\u2022\s*|\d\.\s+', '', text)
     text = re.sub(r'\[.*?\]|\(.*?\)', '', text)
     text = re.sub(r'\bPage\s*\d+\b', '', text, flags=re.IGNORECASE)
     return text.strip()
@@ -72,7 +72,7 @@ def extract_text(file_path: str, file_extension: str):
     except Exception as e:
         return "", f"Error reading {file_extension.upper()} file: {str(e)}"
 
-def chunk_text(text: str, max_tokens: int = 768):
+def chunk_text(text: str, max_tokens: int = 950):
     try:
         sentences = sent_tokenize(text)
     except:
@@ -100,23 +100,22 @@ def generate_summary(text: str, length: str = "medium") -> str:
         "long": {"max_length": 300, "min_length": 210}
     }
     chunks = chunk_text(text)
-    summaries = []
-    for chunk in chunks:
-        try:
-            summary = summarizer(
-                chunk,
-                max_length=length_params[length]["max_length"],
-                min_length=length_params[length]["min_length"],
-                do_sample=False,
-                truncation=True,
-                no_repeat_ngram_size=2,
-                num_beams=2,
-                early_stopping=True
-            )
-            summaries.append(summary[0]['summary_text'])
-        except Exception as e:
-            summaries.append(f"[Chunk error: {str(e)}]")
-    final_summary = " ".join(summaries)
+    try:
+        summaries = summarizer(
+            chunks,
+            max_length=length_params[length]["max_length"],
+            min_length=length_params[length]["min_length"],
+            do_sample=False,
+            truncation=True,
+            no_repeat_ngram_size=2,
+            num_beams=2,
+            early_stopping=True
+        )
+        summary_texts = [s['summary_text'] for s in summaries]
+    except Exception as e:
+        summary_texts = [f"[Batch error: {str(e)}]"]
+
+    final_summary = " ".join(summary_texts)
     final_summary = ". ".join(s.strip().capitalize() for s in final_summary.split(". ") if s.strip())
     return final_summary if len(final_summary) > 25 else "Summary too short - document may be too brief"