percentage

app.py

@@ -21,25 +21,27 @@ CPU_COUNT = multiprocessing.cpu_count()
 MAX_WORKERS = min(32, CPU_COUNT)  # Use CPU count directly for processes
 
 
-def process_pdf(pdf_file, sample_size, temp_dir):
+def process_pdf(pdf_file, sample_percentage, temp_dir):
     try:
         pdf_path = pdf_file.name
         doc = fitz.open(pdf_path)
         total_pages = len(doc)
 
-        pages_to_convert = (
-            total_pages if sample_size == 0 else min(sample_size, total_pages)
-        )
+        pages_to_convert = int(total_pages * (sample_percentage / 100))
+        pages_to_convert = max(
+            1, min(pages_to_convert, total_pages)
+        )  # Ensure at least one page and not more than total pages
+
         selected_pages = (
             sorted(random.sample(range(total_pages), pages_to_convert))
-            if sample_size > 0 and sample_size < total_pages
+            if 0 < sample_percentage < 100
             else range(total_pages)
         )
 
         images = []
         for page_num in selected_pages:
             page = doc[page_num]
-            pix = page.get_pixmap(matrix=fitz.Matrix(2, 2))  # Increase resolution
+            pix = page.get_pixmap()  # Remove the Matrix scaling
             image = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
             image_path = os.path.join(
                 temp_dir, f"{os.path.basename(pdf_path)}_page_{page_num+1}.jpg"
@@ -53,7 +55,7 @@ def process_pdf(pdf_file, sample_size, temp_dir):
         return [], f"Error processing {pdf_file.name}: {str(e)}", 0
 
 
-def pdf_to_images(pdf_files, sample_size, temp_dir, progress=gr.Progress()):
+def pdf_to_images(pdf_files, sample_percentage, temp_dir, progress=gr.Progress()):
     if not os.path.exists(temp_dir):
         os.makedirs(temp_dir)
 
@@ -66,7 +68,7 @@ def pdf_to_images(pdf_files, sample_size, temp_dir, progress=gr.Progress()):
 
     with ThreadPoolExecutor(max_workers=MAX_WORKERS) as executor:
         future_to_pdf = {
-            executor.submit(process_pdf, pdf, sample_size, temp_dir): pdf
+            executor.submit(process_pdf, pdf, sample_percentage, temp_dir): pdf
             for pdf in pdf_files
         }
 
@@ -103,7 +105,7 @@ def get_size_category(num_images):
 
 def process_pdfs(
     pdf_files,
-    sample_size,
+    sample_percentage,
     hf_repo,
     create_zip,
     private_repo,
@@ -134,7 +136,7 @@ def process_pdfs(
         os.makedirs(images_dir)
 
         progress(0, desc="Starting PDF processing")
-        images, message = pdf_to_images(pdf_files, sample_size, images_dir)
+        images, message = pdf_to_images(pdf_files, sample_percentage, images_dir)
 
         # Create a new directory for sampled images
         sampled_images_dir = os.path.join(temp_dir, "sampled_images")
@@ -195,7 +197,9 @@ def process_pdfs(
                     hf_repo=hf_repo,
                     num_images=len(images),
                     num_pdfs=len(pdf_files),
-                    sample_size=sample_size if sample_size > 0 else "All pages",
+                    sample_size=sample_percentage
+                    if sample_percentage > 0
+                    else "All pages",
                     creation_date=datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
                 )
 
@@ -204,7 +208,9 @@ def process_pdfs(
                     hf_repo=hf_repo,
                     num_images=len(images),
                     num_pdfs=len(pdf_files),
-                    sample_size=sample_size if sample_size > 0 else "All pages",
+                    sample_size=sample_percentage
+                    if sample_percentage > 0
+                    else "All pages",
                     creation_date=datetime.now().strftime("%Y-%m-%d %H:%M:%S"),
                     size_category=size_category,
                 )
@@ -248,10 +254,13 @@ with gr.Blocks() as demo:
             file_count="multiple", label="Upload PDF(s)", file_types=["*.pdf"]
         )
     with gr.Row():
-        sample_size = gr.Number(
-            value=None,
-            label="Pages per PDF (0 for all pages)",
-            info="Specify how many pages to convert from each PDF. Use 0 to convert all pages.",
+        sample_percentage = gr.Slider(
+            minimum=0,
+            maximum=100,
+            value=100,
+            step=1,
+            label="Percentage of pages to sample per PDF",
+            info="0% for no sampling (all pages), 100% for all pages",
         )
         hf_repo = gr.Textbox(
             label="Hugging Face Repo",
@@ -269,7 +278,7 @@ with gr.Blocks() as demo:
     submit_button = gr.Button("Convert PDFs to page images")
     submit_button.click(
         process_pdfs,
-        inputs=[pdf_files, sample_size, hf_repo, create_zip, private_repo],
+        inputs=[pdf_files, sample_percentage, hf_repo, create_zip, private_repo],
         outputs=[output_gallery, download_button, status_text],
     )