Upload app1.py

app1.py

@@ -0,0 +1,181 @@
+from transformers import RobertaForSequenceClassification, AutoTokenizer
+import torch
+import docx2txt
+import pandas as pd
+import matplotlib.pyplot as plt
+import openpyxl
+from openpyxl.styles import Font, Color, PatternFill
+from openpyxl.styles.colors import WHITE
+import gradio as gr
+import underthesea
+
+# Load the model and tokenizer
+senti_model = RobertaForSequenceClassification.from_pretrained("wonrax/phobert-base-vietnamese-sentiment")
+senti_tokenizer = AutoTokenizer.from_pretrained("wonrax/phobert-base-vietnamese-sentiment", use_fast=False)
+
+
+def segmentation(sentences):
+    segmented_sentences = []
+    for sentence in sentences:
+        segmented_sentence = underthesea.word_tokenize(sentence)
+        segmented_sentences.append(' '.join(segmented_sentence))
+    return segmented_sentences
+
+
+def analyze(sentence):
+    input_ids = torch.tensor([senti_tokenizer.encode(sentence)])
+    with torch.no_grad():
+        out = senti_model(input_ids)
+        results = out.logits.softmax(dim=-1).tolist()
+        return results[0]
+
+
+def read_file(docx):
+    try:
+        text = docx2txt.process(docx)
+        lines = text.split('\n')
+        lines = [line.strip() for line in lines]
+        lines = [line for line in lines if line]
+        return lines
+    except Exception as e:
+        print(f"Error reading file: {e}")
+
+
+def process_file(docx):
+    # Read the file and segment the sentences
+    sentences = read_file(docx)
+    segmented_sentences = segmentation(sentences)
+
+    # Analyze the sentiment of each sentence
+    results = []
+    for sentence in segmented_sentences:
+        results.append(analyze(sentence))
+
+    # Create a DataFrame from the results
+    df = pd.DataFrame(results, columns=['Negative', 'Neutral', 'Positive'])
+    df['Text'] = read_file(docx)
+
+    # Generate the pie chart and excel file
+    pie_chart_name = generate_pie_chart(df)
+    excel_file_path = generate_excel_file(df)
+
+    return excel_file_path, pie_chart_name
+
+
+def analyze_text(text, docx_file):
+    if text:
+        # Perform analysis on the text
+        segmented_text = segmentation([text])
+        results = []
+        for sentence in segmented_text:
+            results.append(analyze(sentence))
+
+        df = pd.DataFrame(results, columns=['Negative', 'Neutral', 'Positive'])
+        df['Text'] = [text]
+        pie_chart_name = generate_pie_chart(df)
+        excel_file_path = generate_excel_file(df)
+        return excel_file_path, pie_chart_name
+
+    elif docx_file:
+        return process_file(docx_file.name)
+
+    else:
+        # No input provided
+        return None
+
+
+def generate_pie_chart(df):
+    # Calculate the average scores
+    neg_avg = df['Negative'].mean()
+    neu_avg = df['Neutral'].mean()
+    pos_avg = df['Positive'].mean()
+
+    # Create a new DataFrame with the average scores
+    avg_df = pd.DataFrame({'Sentiment': ['Negative', 'Neutral', 'Positive'],
+                           'Score': [neg_avg, neu_avg, pos_avg]})
+
+    # Set custom colors for the pie chart
+    colors = ['#BDBDBD', '#87CEFA', '#9ACD32']
+
+    # Create a pie chart showing the average scores
+    plt.pie(avg_df['Score'], labels=avg_df['Sentiment'], colors=colors, autopct='%1.1f%%')
+    plt.title('Average Scores by Sentiment')
+
+    # Save the pie chart as an image file in the static folder
+    pie_chart_name = 'pie_chart.png'
+    plt.savefig(pie_chart_name)
+    plt.close()
+
+    return pie_chart_name
+
+
+def generate_excel_file(df):
+    # Create a new workbook and worksheet
+    wb = openpyxl.Workbook()
+    ws = wb.active
+
+    # Add column headers to the worksheet
+    headers = ['Negative', 'Neutral', 'Positive', 'Text']
+    for col_num, header in enumerate(headers, 1):
+        cell = ws.cell(row=1, column=col_num)
+        cell.value = header
+        cell.font = Font(bold=True)
+
+    # Set up cell formatting for each sentiment
+    fill_dict = {
+        'Negative': PatternFill(start_color='BDBDBD', end_color='BDBDBD', fill_type='solid'),
+        'Neutral': PatternFill(start_color='87CEFA', end_color='87CEFA', fill_type='solid'),
+        'Positive': PatternFill(start_color='9ACD32', end_color='9ACD32', fill_type='solid')
+    }
+
+    # Loop through each row of the input DataFrame and write data to the worksheet
+    for row_num, row_data in df.iterrows():
+        # Calculate the highest score and corresponding sentiment for this row
+        sentiment_cols = ['Negative', 'Neutral', 'Positive']
+        scores = [row_data[col] for col in sentiment_cols]
+        max_score = max(scores)
+        max_index = scores.index(max_score)
+        sentiment = sentiment_cols[max_index]
+
+        # Write the data to the worksheet
+        for col_num, col_data in enumerate(row_data, 1):
+            cell = ws.cell(row=row_num + 2, column=col_num)
+            cell.value = col_data
+            if col_num in [1, 2, 3]:
+                if col_data == max_score:
+                    cell.fill = fill_dict[sentiment]
+            if col_num == 4:
+                fill = fill_dict[sentiment]
+                font_color = WHITE if fill.start_color.rgb == 'BDBDBD' else Color('000000')
+                cell.fill = fill
+                cell.font = Font(color=font_color)
+                if col_data == max_score:
+                    cell.fill = fill_dict[sentiment]
+
+    # Save the workbook
+    excel_file_path = 'result.xlsx'
+    wb.save(excel_file_path)
+
+    return excel_file_path
+
+
+inputs = [
+    gr.inputs.Textbox(label="Nhập Văn Bản"),
+    gr.inputs.File(label="Chọn Tệp Bạn Muốn Phân Tích")
+]
+
+outputs = [
+    gr.outputs.File(label="Kết Quả Phân Tích Excel"),
+    gr.outputs.Image(type="filepath", label="Thông Số Phân Tích")
+]
+
+interface = gr.Interface(
+    fn=analyze_text,
+    inputs=inputs,
+    outputs=outputs,
+    title="Phân Tích Cảm xúc thông qua Hội Thoại",
+    allow_flagging="never"  # Disable flag button
+)
+
+if __name__ == "__main__":
+    interface.launch(share=True)