David Ponce

After training a model, a `train_config.yaml` is generated with the parameters used during that training. Under `data -> inputs ` there are some references to the `source_pos` and `target_pos` fields....

I was trying to pretrain a PredictorEstimator model with parallel data as you show in your [documentation](https://unbabel.github.io/OpenKiwi/usage.html#training-and-pretraining), but I could not find any **data** yaml configuration file to use as...

Update predictor.yaml

1

Solves #104

Considering that these models have their own tokenization and BPE models, what is the format of the input files to train a QE model using any of this LM? Should...

Take into account most common n-grams.

6

Take into account only items of mapA that have a value BIGGER than 300. This if statement does the opposite.

Styling agm-marker labels

6

Is there a way to style the labels that appear just on top the agm-marker? In my case, I would like to add a shadow to the text, but this...

I think it would be nice to have a small guide indicating how to collaborate to the project to keep things as "standardized" as possible, and maybe try to explain...

Error when resuming from checkpoint, unable to load model.

3

I was trying to fine-tune the model using two distinct prompting methods. For this, I first trained a model using one corpus. Then, using the automatically saved checkpoints, I tried...

How to see all possible lora_target_modules?

1

I was wondering where you can find more info about the `lora_target_modules`. So far, I've seen that the default ones are `["q_proj",v_proj"]`, could you please give more insight on where...

[QUESTION] Train UnifiedMetric/XCOMET with word level predictions.

1

#### What is your question? I was trying to train a model both at sentence and word level, but was unable to find any examples of it. I believe XCOMET...