---
language:
- en
- it
license: mit
tags:
- pytorch
- nlp
- machine-translation
pipeline_tag: translation
datasets:
- Helsinki-NLP/europarl
---

<h1 align="center">tfs-mt<br>
Transformer from scratch for Machine Translation</h1>

<div align="center">
    <a href="https://img.shields.io/github/v/release/Giovo17/tfs-mt" alt="Release">
        <img src="https://img.shields.io/github/v/release/Giovo17/tfs-mt"/>
    </a>
    <a href="https://github.com/Giovo17/tfs-mt/actions/workflows/main.yml?query=branch%3Amain" alt="Build status">
      <img src="https://img.shields.io/github/actions/workflow/status/Giovo17/tfs-mt/main.yml?branch=main"/>
    </a>
    <a href="https://github.com/Giovo17/tfs-mt/blob/main/LICENSE" alt="License">
        <img src="https://img.shields.io/github/license/Giovo17/tfs-mt"/>
    </a>
    <br>
    <a href="https://github.com/Giovo17/tfs-mt">
        🏠 Homepage
    </a>
    •
    <a href="https://giovo17.github.io/tfs-mt">
        📖 Documentation
    </a>
    •
    <a href="https://huggingface.co/spaces/giovo17/tfs-mt-demo">
        🎬 Demo
    </a>
    •
    <a href="https://pypi.org/project/tfs-mt">
        📦 PyPi
    </a>

</div>

---

This project implements the Transformer architecture from scratch considering Machine Translation as the usecase. It's mainly intended as an educational resource and a functional implementation of the architecture and the training/inference logic.

Here you can find the weights of the trained `small` size Transformer and the pretrained tokenizers.

## Quick Start

```bash
pip install tfs-mt
```

```python
import torch

from tfs_mt.architecture import build_model
from tfs_mt.data_utils import WordTokenizer
from tfs_mt.decoding_utils import greedy_decoding

base_url = "https://huggingface.co/giovo17/tfs-mt/resolve/main/"
src_tokenizer = WordTokenizer.from_pretrained(base_url + "src_tokenizer_word.json")
tgt_tokenizer = WordTokenizer.from_pretrained(base_url + "tgt_tokenizer_word.json")

model = build_model(
    config="https://huggingface.co/giovo17/tfs-mt/resolve/main/config-lock.yaml",
    from_pretrained=True,
    model_path="https://huggingface.co/giovo17/tfs-mt/resolve/main/model.safetensors",
)

device = "cuda:0" if torch.cuda.is_available() else "cpu"
model.to(device)
model.eval()

input_tokens, input_mask = src_tokenizer.encode("Hi, how are you?")

output = greedy_decoding(model, tgt_tokenizer, input_tokens, input_mask)[0]
print(output)
```

## Model Architecture

**Model Size**: `small`

- **Encoder Layers**: 6
- **Decoder Layers**: 6
- **Model Dimension**: 100
- **Attention Heads**: 6
- **FFN Dimension**: 400
- **Normalization Type**: postnorm
- **Dropout**: 0.1
- **Pretrained Embeddings**: GloVe
- **Positional Embeddings**: sinusoidal
- **GloVe Version**: glove.2024.wikigiga.100d

### Tokenizer

- **Type**: word
- **Max Sequence Length**: 131
- **Max Vocabulary Size**: 70000
- **Minimum Frequency**: 2


## Dataset

- **Task**: machine-translation
- **Dataset ID**: `Helsinki-NLP/europarl`
- **Dataset Name**: `en-it`
- **Source Language**: en
- **Target Language**: it
- **Train Split**: 0.95

## Full training configuration

<details>
<summary>Click to expand complete config-lock.yaml</summary>

```yaml
seed: 42
log_every_iters: 1000
save_every_iters: 10000
eval_every_iters: 10000
update_pbar_every_iters: 100
time_limit_sec: -1
checkpoints_retain_n: 5
model_base_name: tfs_mt
model_parameters:
  dropout: 0.1
model_configs:
  pretrained_word_embeddings: GloVe
  positional_embeddings: sinusoidal
  nano:
    num_encoder_layers: 4
    num_decoder_layers: 4
    d_model: 50
    num_heads: 4
    d_ff: 200
    norm_type: postnorm
    glove_version: glove.2024.wikigiga.50d
    glove_filename: wiki_giga_2024_50_MFT20_vectors_seed_123_alpha_0.75_eta_0.075_combined
  small:
    num_encoder_layers: 6
    num_decoder_layers: 6
    d_model: 100
    num_heads: 6
    d_ff: 400
    norm_type: postnorm
    glove_version: glove.2024.wikigiga.100d
    glove_filename: wiki_giga_2024_100_MFT20_vectors_seed_2024_alpha_0.75_eta_0.05.050_combined
  base:
    num_encoder_layers: 8
    num_decoder_layers: 8
    d_model: 300
    num_heads: 8
    d_ff: 800
    norm_type: postnorm
    glove_version: glove.2024.wikigiga.300d
    glove_filename: wiki_giga_2024_300_MFT20_vectors_seed_2024_alpha_0.75_eta_0.05_combined
  original:
    num_encoder_layers: 6
    num_decoder_layers: 6
    d_model: 512
    num_heads: 8
    d_ff: 2048
    norm_type: postnorm
training_hp:
  num_epochs: 5
  distributed_training: false
  use_amp: true
  amp_dtype: bfloat16
  torch_compile_mode: max-autotune
  loss:
    type: KLdiv-labelsmoothing
    label_smoothing: 0.1
  optimizer:
    type: AdamW
    weight_decay: 0.0001
    beta1: 0.9
    beta2: 0.999
    eps: 1.0e-08
  lr_scheduler:
    type: original
    min_lr: 0.0003
    max_lr: 0.001
    warmup_iters: 25000
    stable_iters_prop: 0.7
  max_gradient_norm: 5.0
  early_stopping:
    enabled: false
    patience: 40000
    min_delta: 1.0e-05
tokenizer:
  type: word
  sos_token: <s>
  eos_token: </s>
  pad_token: <PAD>
  unk_token: <UNK>
  max_seq_len: 131
  max_vocab_size: 70000
  vocab_min_freq: 2
dataset:
  dataset_task: machine-translation
  dataset_id: Helsinki-NLP/europarl
  dataset_name: en-it
  train_split: 0.95
  src_lang: en
  tgt_lang: it
  max_len: -1
train_dataloader:
  batch_size: 64
  num_workers: 8
  shuffle: true
  drop_last: true
  prefetch_factor: 2
  pad_all_to_max_len: true
test_dataloader:
  batch_size: 128
  num_workers: 8
  shuffle: false
  drop_last: false
  prefetch_factor: 2
  pad_all_to_max_len: true
backend: none
chosen_model_size: small
model_name: tfs_mt_small_251104-1748
exec_mode: dev
src_tokenizer_vocab_size: 70000
tgt_tokenizer_vocab_size: 70000
num_train_iters_per_epoch: 28889
num_test_iters_per_epoch: 761
```

</details>


## Citation

If you use `tfs-mt` in your research or project, please cite:

```bibtex
@software{Spadaro_tfs-mt,
author = {Spadaro, Giovanni},
license = {MIT},
title = {{tfs-mt}},
url = {https://github.com/Giovo17/tfs-mt}
}
```