iput_3rd_DL_theme_2_level_1

English to French Translator (Fine-tuned)

This model is a fine-tuned version of facebook/mbart-large-50-many-to-many-mmt for English to French translation.

This was uploaded as a university assignment and has low accuracy.

Downloading is not recommended.

Usage

from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

# Load model and tokenizer
tokenizer = AutoTokenizer.from_pretrained("GudeGudeEgg/iput_3rd_DL_theme_2_level1", token='your_token')
model = AutoModelForSeq2SeqLM.from_pretrained("GudeGudeEgg/iput_3rd_DL_theme_2_level1", token='your_token')

# Translate
def translate_en_to_fr(text):
    inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True)
    generated_tokens = model.generate(**inputs, max_length=128)
    return tokenizer.decode(generated_tokens[0], skip_special_tokens=True)

# Example
result = translate_en_to_fr("The cat is sleeping")
print(result)  # Should output French translation

Model Details

• Base model: facebook/mbart-large-50-many-to-many-mmt
• Fine-tuned on English-French translation pairs
• Training data: 29000 sentence pairs
• Supports batch translation
• Optimized for conversational text

Training

The model was fine-tuned for 3 epochs using the HuggingFace Trainer with:

• Learning rate: 5e-5
• Batch size: 8
• Max length: 128 tokens