Edit model card

iva_mt_wslot-m2m100_418M-en-sv

This model is a fine-tuned version of facebook/m2m100_418M on the iva_mt_wslot dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0107
  • Bleu: 71.0808
  • Gen Len: 19.7647

Model description

More information needed

How to use

First please make sure to install pip install transformers. First download model:

from transformers import M2M100ForConditionalGeneration, M2M100Tokenizer
import torch

def translate(input_text, lang):
    input_ids = tokenizer(input_text, return_tensors="pt")
    generated_tokens = model.generate(**input_ids, forced_bos_token_id=tokenizer.get_lang_id(lang))
    return tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)

model_name = "cartesinus/iva_mt_wslot-m2m100_418M-0.1.0-en-sv"
tokenizer = M2M100Tokenizer.from_pretrained(model_name, src_lang="en", tgt_lang="sv")
model = M2M100ForConditionalGeneration.from_pretrained(model_name)

Then you can translate either plain text like this:

print(translate("set the temperature on my thermostat", "sv"))

or you can translate with slot annotations that will be restored in tgt language:

print(translate("wake me up at <a>nine am<a> on <b>friday<b>", "sv"))

Limitations of translation with slot transfer:

  1. Annotated words must be placed between semi-xml tags like this "this is <a>example<a>"
  2. There is no closing tag for example "<\a>" in above example - this is done on purpose to omit problems with backslash escape
  3. If the sentence consists of more than one slot then simply use the next alphabet letter. For example "this is <a>example<a> with more than <b>one<b> slot"
  4. Please do not add space before the first or last annotated word because this particular model was trained this way and it most probably will lower its results

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 2e-05
  • train_batch_size: 4
  • eval_batch_size: 4
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 7
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Bleu Gen Len
0.0151 1.0 1885 0.0120 67.2332 19.3956
0.0095 2.0 3770 0.0105 69.8147 19.675
0.0065 3.0 5655 0.0104 70.239 19.8404
0.0049 4.0 7540 0.0104 70.3673 19.7154
0.0038 5.0 9425 0.0105 70.1632 19.7743
0.0026 6.0 11310 0.0105 70.7959 19.7809
0.0021 7.0 13195 0.0107 71.0808 19.7647

Framework versions

  • Transformers 4.28.1
  • Pytorch 2.0.0+cu118
  • Datasets 2.11.0
  • Tokenizers 0.13.3

Citation

If you use this model, please cite the following:

@article{Sowanski2023SlotLI,
  title={Slot Lost in Translation? Not Anymore: A Machine Translation Model for Virtual Assistants with Type-Independent Slot Transfer},
  author={Marcin Sowanski and Artur Janicki},
  journal={2023 30th International Conference on Systems, Signals and Image Processing (IWSSIP)},
  year={2023},
  pages={1-5}
}
Downloads last month
2
Safetensors
Model size
486M params
Tensor type
F32
·

Dataset used to train cartesinus/iva_mt_wslot-m2m100_418M-en-sv

Evaluation results