Create app.py

app.py

@@ -0,0 +1,40 @@
+from transformers import Trainer, TrainingArguments, DeepSeekForQuestionAnswering, DeepSeekTokenizer
+from datasets import load_dataset
+
+# Load the DeepSeek model and tokenizer
+model = DeepSeekForQuestionAnswering.from_pretrained("DeepSeek/DeepSeek-v3")
+tokenizer = DeepSeekTokenizer.from_pretrained("DeepSeek/DeepSeek-v3")
+
+# Load dataset
+dataset = load_dataset("json", data_files={"train": "your_dataset_train.json", "test": "your_dataset_test.json"})
+
+# Tokenize the dataset
+def tokenize_function(examples):
+    return tokenizer(examples['question'], examples['document'], truncation=True, padding=True)
+
+tokenized_datasets = dataset.map(tokenize_function, batched=True)
+
+# Set up the training arguments
+training_args = TrainingArguments(
+    output_dir='./results', 
+    evaluation_strategy="epoch",
+    learning_rate=2e-5,
+    per_device_train_batch_size=16,
+    per_device_eval_batch_size=16,
+    num_train_epochs=3,
+    weight_decay=0.01
+)
+
+# Initialize Trainer
+trainer = Trainer(
+    model=model,
+    args=training_args,
+    train_dataset=tokenized_datasets['train'],
+    eval_dataset=tokenized_datasets['test']
+)
+
+# Start the fine-tuning
+trainer.train()
+
+# Save the model after fine-tuning
+model.save_pretrained('./fine_tuned_deepseek')