Update README.md

README.md

@@ -78,48 +78,6 @@ import pandas as pd
 df = pd.read_parquet("path/to/dataset.parquet")
 ```
 
-### Example: Training a Tokenizer
-
-```python
-from transformers import BertTokenizerFast
-
-# Initialize a new tokenizer from the dataset text
-tokenizer = BertTokenizerFast.from_pretrained("bert-base-multilingual-uncased")
-tokenizer = tokenizer.train_new_from_iterator(dataset["train"]["title"], vocab_size=30000)
-
-# Save the tokenizer
-tokenizer.save_pretrained("./cantonese-tokenizer")
-```
-
-### Example: Fine-tuning a Model
-
-```python
-from transformers import BertForMaskedLM, Trainer, TrainingArguments
-
-# Initialize model
-model = BertForMaskedLM.from_pretrained("bert-base-multilingual-uncased")
-
-# Define training arguments
-training_args = TrainingArguments(
-    output_dir="./cantonese-bert",
-    overwrite_output_dir=True,
-    num_train_epochs=3,
-    per_device_train_batch_size=16,
-    save_steps=10_000,
-    save_total_limit=2,
-)
-
-# Initialize trainer
-trainer = Trainer(
-    model=model,
-    args=training_args,
-    train_dataset=dataset["train"],
-)
-
-# Train model
-trainer.train()
-```
-
 ## Data Format
 
 Each row in the dataset contains: