Model Zoo
Here is the list of models currently implemented in MMF:
Model | Key | Datasets | Notes |
---|---|---|---|
BAN | ban | textvqa, vizwiz, vqa2 | BAN support is preliminary and hasn't been properly fine-tuned yet. |
BUTD | butd | coco | paper |
CNN LSTM | cnn_lstm | clevr | |
FUSIONS | concat_bert, late_fusion, concat_bow | hateful_memes | |
LoRRA | lorra | vqa2, textvqa, vizwiz | paper |
LXMERT | lxmert | coco, gqa, visual_genome, vqa2 | paper |
M4C | m4c | ocrvqa, stvqa, textvqa | paper |
M4C Captioner | m4c_captioner | coco, textcaps | paper |
MMBT | mmbt | hateful_memes, coco, mmimdb, okvqa, vqa2 | paper |
MMF Transformer | mmf_transformer | hateful_memes, okvqa, vqa2 | |
Movie MCAN | movie_mcan | vqa2 | paper |
Pythia | pythia | textvqa, vizwiz, vqa2, visual_genome | paper |
Unimodal | unimodal | hateful_memes | |
VilBERT | vilbert | hateful_memes, coco, conceptual_captions, mmimdb, nlvr2, visual_entailment, vizwiz, vqa2 | paper |
ViLT | vilt | coco, vqa2 | paper |
Visual BERT | visual_bert | gqa, hateful_memes, localized_narratives, coco, conceptual_captions, sbu, vqa2, mmimdb, nlvr2, visual_entailment, vizwiz | paper |
We are adding many more new models which will be available soon.