Update app.py

app.py

@@ -1,21 +1,58 @@
 import os
 import gradio as gr
+import pandas as pd
 from datasets import load_dataset
+from transformers import MarianMTModel, MarianTokenizer
 
-def load_and_show():
+MODEL_NAME = "Helsinki-NLP/opus-mt-fr-en"
+
+tokenizer = MarianTokenizer.from_pretrained(MODEL_NAME)
+model = MarianMTModel.from_pretrained(MODEL_NAME)
+
+def translate_text(text):
+    text = str(text)[:800]
+    inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=512)
+    outputs = model.generate(**inputs, max_length=512)
+    return tokenizer.decode(outputs[0], skip_special_tokens=True)
+
+def translate_rows(n):
     dataset = load_dataset(
         "tcrouzet/journal-large",
-        split="train[:1]",
+        split=f"train[:{int(n)}]",
         token=os.environ["may"]
     )
-    return str({
-        "columns": dataset.column_names,
-        "first_row": dataset[0]
-    })
+
+    rows = []
+    for row in dataset:
+        text_fr = row["combined"]
+
+        try:
+            text_en = translate_text(text_fr)
+        except Exception as e:
+            text_en = f"TRANSLATION_ERROR: {e}"
+
+        rows.append({
+            "id": row["id"],
+            "title": row["Title"],
+            "subtitle": row["Subtitle"],
+            "date": row["Date"],
+            "location": row["Location"],
+            "tags": row["Tags"],
+            "author": row["Author"],
+            "text_fr": text_fr,
+            "text_en": text_en
+        })
+
+    output_file = "/tmp/translated_sample.csv"
+    pd.DataFrame(rows).to_csv(output_file, index=False)
+    return output_file
 
 with gr.Blocks() as demo:
-    gr.Markdown("# Dataset Debug")
-    output = gr.Textbox()
-    demo.load(fn=load_and_show, inputs=None, outputs=output)
+    gr.Markdown("# French Journal Dataset Translator")
+    n = gr.Number(value=10, precision=0, label="Rows to translate")
+    btn = gr.Button("Translate")
+    file_output = gr.File(label="Download translated CSV")
+
+    btn.click(fn=translate_rows, inputs=n, outputs=file_output)
 
 demo.launch()