main_dist.py
is the main file.dat_loader_simple.py
processes the data. In particular, the SPAT/TEMP/SEP part is modular, can be easily extended to a newer dataset.contrastive_sampling.py
as the name suggests creates the contrastive samples for Training. The validation file already contains these indices.mdl_base.py
is the base model, which just defines bunch of functions to be filled in.mdl_conc_single.py
implements concatenation models and losses for SPAT/TEMP. Similarly,mdl_conc_sep.py
implements SEP concatentation model and loss. These are kept modular, so that they can be re-used with newer models withminimalsome effort.mdl_vog.py
contains the main model implementations of baselines and vog.mdl_selector.py
returns the model, loss and evaluation function to be used based on input arguments.eval_vsrl_corr.py
is the top-level evaluation functions for each of SEP/TEMP/SPAT which processes the output of the model and converts them to uniform format for evaluation.eval_fn_corr.py
contains the main logic for evaluating the models._init_stuff.py
initializes paths to be included, typings, as well as yaml float loader (otherwise 1e-4 cannot be read correctly).extended_config.py
has some handy configuration utils.transformer_code.py
has the transformer implementation, also has the relative transformer which uses relative position encoding (RPE).
Some other useful files are under utils