Translation problem using ctranslate2

I have trained a transformer model and used ctranslate2 to convert the checkpoints in suitable format, but when i try to do the translation i am not getting any translation

Command line: ct2-opennmt-tf-converter --model_path /content/drive/MyDrive/Data/English-Tamil/en-kn/run/avg --output_dir ende_ctranslate2
–src_vocab /content/drive/MyDrive/Data/English-Tamil/en-kn/vocab.SRC
–tgt_vocab /content/drive/MyDrive/Data/English-Tamil/en-kn/vocab.TGT
–model_type Transformer

I dont understand which is creating the error

Please provide more information:

  • the training command line and configuration
  • the tokenization method
  • how you are running the translation behind the web interface.

Also, did you try other inputs?

  1. onmt-main --model_type Transformer --config data.yaml --auto_config train --with_eval
    2.I am using openNMT tokenizer
  2. I am using streamlite interface
  3. Yes it is not translating to the target language

Try running the inference with OpenNMT-tf and see if it is different.

no i am not running into any problems with opennmt-tf

These kind of problems usually arise from mismatches in tokenization, vocabularies and/or input format between training and inference.

1 Like

Hi Amartya,

Could you please try to test CTranslate2 independently. Please make sure you update your version to the latest before converting the model.

Here is a sample code you can use to test your model. Please change the tokenize and detokenize functions as well as your CTranslate2 model path. Test the model with complete sentences rather than single words.

Kindly revise your whole paths; I agree with Panos there might be something to correct.

import ctranslate2

# Replace with your tokenize function and source tokenization model
def tokenize(input_sentences):
    tokens = [input_sentence.split(" ") for input_sentence in input_sentences]
    return tokens

# Replace with your detokenize function and target tokenization model
def detokenize(outputs):
    translation = [" ".join([t for t in output]) for output in outputs]
    return translation

# Modify the path to the CTranslate2 model directory
model_path = "ctranslate2_model"

source_sentences = ["how are you?", "fine, thanks!", "everything is great.", "I am happy to know that."]

translator = ctranslate2.Translator(model_path, "cpu") # "cpu" or "cuda"

outputs = translator.translate_batch(tokenize(source_sentences), beam_size=5)
translation = detokenize(output[0]["tokens"] for output in outputs)

Kind regards,

1 Like