math_english_to_latex

This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.8028

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 4
• eval_batch_size: 20
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss
2.3664	1.0	10	1.6544
1.1415	2.0	20	0.9639
0.8763	3.0	30	0.9328
0.7064	4.0	40	0.8390
0.6468	5.0	50	0.8028

CONVERSION_PROMPT = 'LCT\n'  # LaTeX conversion task

CONVERSION_TOKEN = 'LaTeX:'

loaded_model = GPT2LMHeadModel.from_pretrained('Andyrasika/math_english_to_latex')
latex_generator = pipeline('text-generation', model=loaded_model, tokenizer=tokenizer)

text_sample = 'r of x is sum from 0 to x of x squared'
conversion_text_sample = f'{CONVERSION_PROMPT}English: {text_sample}\n{CONVERSION_TOKEN}'

print(latex_generator(
    conversion_text_sample, num_beams=5, early_stopping=True, temperature=0.7,
    max_length=len(tokenizer.encode(conversion_text_sample)) + 20
)[0]['generated_text'])

Framework versions

• Transformers 4.33.3
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3