Update

app.py

@@ -4,14 +4,15 @@ from tqdm import tqdm
 import re
 import gradio as gr
 import os
-# from llama_cpp import Llama
-# from gpt4all import GPT4All
+from llama_cpp import Llama
+from gpt4all import GPT4All
 import transformers
 # from transformers import GemmaTokenizer, AutoModelForCausalLM
 # from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
 import accelerate
 import torch
 
+# HF_TOKEN = os.environ.get("HF_TOKEN", None)
 
 def process_document(pdf_path, page_ids=None):
 
@@ -68,12 +69,9 @@ def txt_to_html(text):
     html_content += "</body></html>"
     return html_content
 
-def deidentify_doc(pdftext="", prompt="", maxtokens=600, temperature=1.2, top_probability=0.95):
+def deidentify_doc(pdftext="", prompt="", maxtokens=600, temperature=0.0001, top_probability=0.95):
 
-    # prompt = "Please anonymize the following clinical note. Replace all the following information with the term '[redacted]': Redact any strings that might be a name or initials, patients’ names, doctors’ names, the names Dr., redact any medical staff names, redact any strings that might be a location or address, such as '3970 Longview Drive', redact any strings that look like 'age 37', redact any dates and registration numbers, redact professions such as 'manager', redact any contact information."
-
-    print('Max Tokens is  ',maxtokens)
-    print('Temperature is  ',temperature)
+    prompt = "Task: Please anonymize the following clinical note. Specific Rules: Replace all the following information with the term \"[redacted]\": 1. Redact any strings that is person name 2. Redact any medical staff names 3. Redact any strings that is location or address, such as \"3970 Longview Drive\" 4. Redact any strings that is age of person 5. Redact any dates and IDs 6. Redact clinic and hospital names 7. Redact professions such as \"manager\" 8. Redact any contact information"
 
     # output = model.create_chat_completion(
     #                 messages = [
@@ -118,19 +116,35 @@ def deidentify_doc(pdftext="", prompt="", maxtokens=600, temperature=1.2, top_pr
 
     return output
 
-def pdf_to_text(file, prompt, maxtokens=600, temperature=1.2, top_probability=0.95):
-    pdftext=""
-    if(file):
-       page2content = process_document(file, page_ids=[0])
-       pdftext = page2content[1]
-    display_text = deidentify_doc(pdftext, prompt, maxtokens, temperature, top_probability)
-    html = txt_to_html(display_text)
-    with open('out.html', "w", encoding="utf-8") as file:
-        file.write(html)
-    return html
+def mkdir(dir):
+    if not os.path.exists(dir):
+        os.makedirs(dir)
+
+def pdf_to_text(files, output_folder, prompt="", maxtokens=600, temperature=1.2, top_probability=0.95):
+    if not output_folder:
+        output_folder='C:/'
+    output_folder = output_folder.replace('\\','/')
+    for file in files:
+        file_name = os.path.basename(file)
+        file_name_splt = file_name.split('.')
+        print('File name is ', file_name)
+        print('output folder is ', output_folder)
+        if(len(file_name_splt)>1 and file_name_splt[1]=='pdf'):
+           page2content = process_document(file, page_ids=[0])
+           pdftext = page2content[1]
+           if(pdftext):
+               anonymized_text = deidentify_doc(pdftext, prompt, maxtokens, temperature, top_probability)
+               # html = txt_to_html(display_text)
+               # with open('out.html', "w", encoding="utf-8") as file:
+               #     file.write(html)
+               with open(os.path.join(output_folder, file_name_splt[0]+'.txt'), 'w') as file:
+                # Write some text to the file
+                file.write(anonymized_text)
+    display_text = "All selected reports are anonymized and results are saved in "
+    return display_text
 
 # model_id = "D:/llama/meta-llama/Meta-Llama-3-8B-Instruct.Q5_K_M.gguf"
-# model = Llama(model_path=model_id, n_ctx=2048, n_threads=8, n_gpu_layers=-1, n_batch=64)
+# model = Llama(model_path=model_id, n_ctx=2048, n_threads=8, n_gpu_layers=32, n_batch=64)
 
 # model = GPT4All("Meta-Llama-3-8B-Instruct.Q4_0.gguf", n_threads=8, device='gpu')
 # model.chat_session()
@@ -140,17 +154,19 @@ model = transformers.pipeline(
     "text-generation",
     model=model_id,
     model_kwargs={"torch_dtype": torch.bfloat16},
-    device="cuda"
+    device="cuda",
 )
 
-css = ".gradio-container {background: 'logo.png'}"
 temp_slider = gr.Slider(minimum=0, maximum=2, value=0.2, label="Temperature Value")
 prob_slider = gr.Slider(minimum=0, maximum=1, value=0.95, label="Max Probability Value")
 max_tokens = gr.Number(value=600, label="Max Tokens")
+input_folder = gr.File(file_count='multiple')
+output_text = gr.Textbox()
+output_path_component = gr.File(label="Select Output Path")
 iface = gr.Interface(
     fn = pdf_to_text,
-    inputs = ['file', "textbox", max_tokens, temp_slider, prob_slider],
-    outputs="html",
+    inputs = ['files', gr.Textbox(label='Enter output folder path')],
+    outputs=output_text,
     title='COBIx Endoscopy Report De-Identification',
     description="This application assists to remove personal information from the uploaded clinical report",
     theme=gr.themes.Soft(),