Kurt Shuster
Kurt Shuster
A good example would be the `image_chat` dataset, or any models trained on such a task. The message format would look something like [this](https://github.com/facebookresearch/ParlAI/blob/master/parlai/tasks/image_chat/agents.py#L10); more information about that dataset/models are...
You may need to set `--image-mode resnet152` when evaluating
The released model uses a pre-trained text encoder ([see section 4.3 of the paper](https://arxiv.org/abs/1810.10665)), while the model you are training is from scratch.
oh i think you should actually remove and redownload the model. When fine-tuning, the `--model-file` is where you _save_ the model checkpoint; you would wnat to use `--init-model` to indicate...
You are correct about the difference between the two - we generally use `model-file` for inference, and `init-model` for training (so as not to overwrite the original checkpoint). In any...
You are not specifying `--num-test-labels 5` in your train script, and thus are evaluating on a harder test set
we currently don't support a poetry install
Can you please paste your commands as well as full stack trace?
If you add specialized keys beyond `text`, or the ones that BB2 normally uses, they will be ignored by the model. Your best bet is to either prepend the topic...
> Do you think the model will be able to learn in this way...? if you have suitable training data, I would guess it'll learn something. Only one way to...