HuggingFaceM4
/

idefics-80b

Text Generation

image-text-to-text

text-generation-inference

Model card Files Files and versions Community

VictorSanh commited on Aug 3, 2023

Commit

d464935

•

1 Parent(s): a2ee636

small wording issue

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -195,7 +195,7 @@ We use the following hyper and training parameters:
 We start from the base IDEFICS models and fine-tune the models by unfreezing all the parameters (vision encoder, language model, cross-attentions). The mixture is composed of following English datasets:
-| Data Source | Data Description                             | Number of unrepeated samples | Sampling ratio |
 |-------------|----------------------------------------------|------------------------------|----------------|
 | [M3IT](https://huggingface.co/datasets/MMInstruction/M3IT)                       | Prompted image-text academic datasets              | 1.5M  | 7.7%  |
 | [LRV-Instruction](https://huggingface.co/datasets/VictorSanh/LrvInstruction)     | Triplets of image/question/answer                  | 155K  | 1.7%  |

 We start from the base IDEFICS models and fine-tune the models by unfreezing all the parameters (vision encoder, language model, cross-attentions). The mixture is composed of following English datasets:
+| Data Source | Data Description                             | Number of Unique Samples | Sampling ratio |
 |-------------|----------------------------------------------|------------------------------|----------------|
 | [M3IT](https://huggingface.co/datasets/MMInstruction/M3IT)                       | Prompted image-text academic datasets              | 1.5M  | 7.7%  |
 | [LRV-Instruction](https://huggingface.co/datasets/VictorSanh/LrvInstruction)     | Triplets of image/question/answer                  | 155K  | 1.7%  |