Update app.py

app.py

@@ -1,24 +1,20 @@
 from huggingface_hub import InferenceClient
-import pandas as pd
-import re
-import random
-import csv
-import tempfile
 import gradio as gr
+import PyPDF2
+import random
+import pandas as pd
+from io import BytesIO
+
 client = InferenceClient("mistralai/Mixtral-8x7B-Instruct-v0.1")
 
-def extract_sentences_from_excel(file):
+def extract_text_from_excel(file):
     df = pd.read_excel(file)
-    sentences = []
-    for row in df.values.tolist():
-        text = ' '.join(str(x) for x in row)
-        new_sentences = re.split(r'(?<=[^.!?])(?=[.!?])', text)
-        sentences.extend([s.strip() for s in new_sentences if s.strip()])
-    return sentences
+    text = ' '.join(df['data'].astype(str))
+    return text
 
-def generate_synthetic_data(file, temperature, max_new_tokens, top_p, repetition_penalty):
-    sentences = extract_sentences_from_excel(file)
-    random.shuffle(sentences)
+def generate_sentences(text, temperature, max_new_tokens, top_p, repetition_penalty):
+    sentences = text.split('.')
+    random.shuffle(sentences)  # Shuffle sentences
 
     generated_data = []
 
@@ -37,41 +33,42 @@ def generate_synthetic_data(file, temperature, max_new_tokens, top_p, repetition
         }
 
         try:
-            output = client.generate(sentence, **generate_kwargs, return_full_text=True)
-            synthetic_data = output.text.strip()
-
-            generated_sentences = re.split(r'(?<=[\.\!\?:])[\s\n]+', synthetic_data)
-            generated_sentences = [s.strip() for s in generated_sentences if s.strip() and s != '.']
+            stream = client.text_generation(sentence, **generate_kwargs, stream=True, details=True, return_full_text=False)
+            output = ""
+            for response in stream:
+                output += response.token.text
 
-            for generated_sentence in generated_sentences:
-                generated_data.append({'Original Sentence': sentence, 'Synthetic Data': generated_sentence})
+            generated_sentences = [s.strip() for s in output.split('.') if s.strip()]
+            generated_data.extend([(sentence, generated_sentence) for generated_sentence in generated_sentences])
 
         except Exception as e:
             print(f"Error generating data for sentence '{sentence}': {e}")
 
-    with tempfile.NamedTemporaryFile(mode='w', newline='', delete=False, suffix='.csv') as tmp:
-        fieldnames = ['Original Sentence', 'Synthetic Data']
-        writer = csv.DictWriter(tmp, fieldnames=fieldnames)
-        writer.writeheader()
-
-        for data in generated_data:
-            writer.writerow(data)
+    return generated_data
 
-        tmp_path = tmp.name
+def save_to_csv(data, filename="synthetic_data.csv"):
+    with open(filename, mode='w', newline='', encoding='utf-8') as file:
+        writer = csv.writer(file)
+        writer.writerow(['Original Sentence', 'Generated Sentence'])
+        writer.writerows(data)
 
-    return tmp_path
+def generate(file, temperature, max_new_tokens, top_p, repetition_penalty):
+    text = extract_text_from_excel(file)
+    data = generate_sentences(text, temperature, max_new_tokens, top_p, repetition_penalty)
+    save_to_csv(data)
+    return gr.File.update(value=filename, visible=True)
 
 gr.Interface(
-    fn=generate_synthetic_data,
+    fn=generate,
     inputs=[
-        gr.File(label="Upload Excel File", file_count="single", file_types=[".xlsx", ".xls"]),
+        gr.File(label="Upload Excel File", file_count="single", file_types=[".xlsx"]),
         gr.Slider(label="Temperature", value=0.9, minimum=0.0, maximum=1.0, step=0.05, interactive=True, info="Higher values produce more diverse outputs"),
         gr.Slider(label="Max new tokens", value=256, minimum=0, maximum=5120, step=64, interactive=True, info="The maximum numbers of new tokens"),
         gr.Slider(label="Top-p (nucleus sampling)", value=0.95, minimum=0.0, maximum=1, step=0.05, interactive=True, info="Higher values sample more low-probability tokens"),
         gr.Slider(label="Repetition penalty", value=1.0, minimum=1.0, maximum=2.0, step=0.1, interactive=True, info="Penalize repeated tokens"),
     ],
-    outputs=gr.File(label="Synthetic Data CSV"),
-    title="Synthetic Data Generation",
-    description="Generate synthetic data from sentences in an Excel file and save it to a CSV file.",
+    outputs=gr.File(label="Synthetic Data"),
+    title="SDG",
+    description="AYE QABIL.",
     allow_flagging="never",
 ).launch()