| 123456789101112131415161718192021222324252627282930313233343536373839404142 |
- Collections:
- - Name: Flamingo
- Metadata:
- Architecture:
- - Transformer
- - Gated Cross-Attention Dense
- Paper:
- Title: 'Flamingo: a Visual Language Model for Few-Shot Learning'
- URL: https://arxiv.org/abs/2204.14198
- README: configs/flamingo/README.md
- Models:
- - Name: flamingo_3rdparty-zeroshot_caption
- Metadata:
- FLOPs: null
- Parameters: 8220452880
- In Collection: Flamingo
- Results:
- - Task: Image Caption
- Dataset: COCO
- Metrics:
- CIDER: 65.50 # Report from the official repo
- Weights: https://download.openmmlab.com/mmclassification/v1/flamingo/openflamingo-9b-adapter_20230505-554310c8.pth
- Config: configs/flamingo/flamingo_zeroshot_caption.py
- Converted From:
- Weights: https://huggingface.co/openflamingo/OpenFlamingo-9B
- Code: https://github.com/mlfoundations/open_flamingo
- - Name: flamingo_3rdparty-zeroshot_vqa
- Metadata:
- FLOPs: null
- Parameters: 8220452880
- In Collection: Flamingo
- Results:
- - Task: Visual Question Answering
- Dataset: VQAv2
- Metrics:
- Accuracy: 43.50 # Report from the official repo
- Weights: https://download.openmmlab.com/mmclassification/v1/flamingo/openflamingo-9b-adapter_20230505-554310c8.pth
- Config: configs/flamingo/flamingo_zeroshot_vqa.py
- Converted From:
- Weights: https://huggingface.co/openflamingo/OpenFlamingo-9B
- Code: https://github.com/mlfoundations/open_flamingo
|
