add model in HF format

README.md

@@ -9,7 +9,7 @@ license: cc-by-4.0
 
 ## HPLT MT release v1.0
 
-This repository contains the translation model for en-hi trained with OPUS and HPLT data. For usage instructions, evaluation scripts, and inference scripts, please refer to the [HPLT-MT-Models v1.0](https://github.com/hplt-project/HPLT-MT-Models/tree/main/v1.0) GitHub repository.
+This repository contains the translation model for English-Hindi trained with OPUS and HPLT data. The model is available in both Marian and Hugging Face formats.
 
 ### Model Info
 
@@ -18,26 +18,51 @@ This repository contains the translation model for en-hi trained with OPUS and H
 * Dataset: OPUS and HPLT data
 * Model architecture: Transformer-base
 * Tokenizer: SentencePiece (Unigram)
-* Cleaning: We used OpusCleaner with a set of basic rules. Details can be found in the filter files in [Github](https://github.com/hplt-project/HPLT-MT-Models/tree/main/v1.0/data/en-hi/raw/v2)
+* Cleaning: We used [OpusCleaner](https://github.com/hplt-project/OpusCleaner) with a set of basic rules. Details can be found in the filter files [here](https://github.com/hplt-project/HPLT-MT-Models/tree/main/v1.0/data/en-hi/raw/v2).
 
-You can also read our deliverable report [here](https://hplt-project.org/HPLT_D5_1___Translation_models_for_select_language_pairs.pdf) for more details.
+You can check out our [deliverable report](https://hplt-project.org/HPLT_D5_1___Translation_models_for_select_language_pairs.pdf), [GitHub repository](https://github.com/hplt-project/HPLT-MT-Models/tree/main/v1.0), and [website](https://hplt-project.org) for more details.
 
 ### Usage
 
 
-The model has been trained with Marian. To run inference, refer to the [Inference/Decoding/Translation](https://github.com/hplt-project/HPLT-MT-Models/tree/main/v1.0#inferencedecodingtranslation) section of our GitHub repository.
+The model has been trained with [MarianNMT](https://github.com/marian-nmt/marian) and the weights are in the Marian format. We have also converted the model into the Hugging Face format so it is compatible with `transformers`.
 
-The model can be used with the Hugging Face framework if the weights are converted to the Hugging Face format. We might provide this in the future; contributions are also welcome.
+#### Using Marian
+To run inference with MarianNMT, refer to the [Inference/Decoding/Translation](https://github.com/hplt-project/HPLT-MT-Models/tree/main/v1.0#inferencedecodingtranslation) section of our GitHub repository. You will need the model file `model.npz.best-chrf.npz` and the vocabulary file `model.en-hi.spm` from this repository.
+
+#### Using transformers
+We have also converted this model to the Hugging Face format and you can get started with the script below. **Note** that due a [known issue](https://github.com/huggingface/transformers/issues/26216) in weight conversion, the checkpoint cannot work with transformer versions <4.26 or >4.30. We tested and suggest `pip install transformers==4.28`.
+
+```
+from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
+
+tokenizer = AutoTokenizer.from_pretrained("HPLT/translate-en-hi-v1.0-hplt_opus")
+model = AutoModelForSeq2SeqLM.from_pretrained("HPLT/translate-en-hi-v1.0-hplt_opus")
+
+inputs = ["Input goes here.", "Make sure the language is right."]
+batch_tokenized = tokenizer(inputs, return_tensors="pt", padding=True)
+model_output = model.generate(
+    **batch_tokenized, num_beams=6, max_new_tokens=512
+)
+batch_detokenized = tokenizer.batch_decode(
+    model_output,
+    skip_special_tokens=True,
+)
+
+print(batch_detokenized)
+```
 
 ### Benchmarks
 
-| testset                                | BLEU | chrF++ | COMET22 |
+When decoded using Marian, the model has the following test scores.
+
+| Test set                               | BLEU | chrF++ | COMET22 |
 | -------------------------------------- | ---- | ----- | ----- |
-| flores200     | 34.5 | 57.1  | 0.7858  |
-| ntrex | 27.7   | 50.9  | 0.7465  |
+| FLORES200     | 34.5 | 57.1  | 0.7858  |
+| NTREX | 27.7   | 50.9  | 0.7465  |
 
 ### Acknowledgements
 
 This project has received funding from the European Union's Horizon Europe research and innovation programme under grant agreement No 101070350 and from UK Research and Innovation (UKRI) under the UK government's Horizon Europe funding guarantee [grant number 10052546]
 
-Brought to you by researchers from the University of Edinburgh, Charles University in Prague, and the whole HPLT consortium.
+Brought to you by researchers from the University of Edinburgh and Charles University in Prague with support from the whole HPLT consortium.