🦦 Otter, a multi-modal model based on OpenFlamingo (open-sourced version of DeepMind's Flamingo), trained on MIMIC-IT and showcasing improved instruction-following and in-context learning ability.
To simplify the execution of the loading script, there's no requirement to input the paths for all datasets via the command line. Instead, you can facilitate the process by adding the argument --training_data_yaml=pipeline/train/config.yaml.
In the configuration YAML file, structure your entries as follows:
Description
To simplify the execution of the loading script, there's no requirement to input the paths for all datasets via the command line. Instead, you can facilitate the process by adding the argument
--training_data_yaml=pipeline/train/config.yaml
.In the configuration YAML file, structure your entries as follows:
For instance:
No changes are necessary for the previous script; commands such as
mimicit_vt_path
remain compatible.Checklist
Before you open a pull-request, please check if a similar issue already exists or has been closed before.
When you open a pull-request, please be sure to include the following
dataset
(mimic-it download, usage, etc.),demo
(online demo),doc
(readme, wiki, paper, video, etc.),evaluation
(evaluation result, performance of Otter, etc.),model
(model configuration, components, etc.),train
(training configuration, process, code, etc.)Thank you for your contributions!