Update README.md
Browse files
    	
        README.md
    CHANGED
    
    | @@ -18,9 +18,9 @@ dataset_info: | |
| 18 | 
             
                - name: sha256
         | 
| 19 | 
             
                  dtype: string
         | 
| 20 | 
             
                - name: original_height
         | 
| 21 | 
            -
                  dtype:  | 
| 22 | 
             
                - name: original_width
         | 
| 23 | 
            -
                  dtype:  | 
| 24 | 
             
            ---
         | 
| 25 |  | 
| 26 | 
             
            # Dataset Card for DataComp-12M
         | 
| @@ -32,7 +32,7 @@ We distribute the image url-text samples and metadata under a standard Creative | |
| 32 | 
             
            Image-text models trained on DataComp-12M are significantly better than on CC-12M/YFCC-15M as well as DataComp-Small/Medium.
         | 
| 33 |  | 
| 34 | 
             
            DataComp-12M was introduced in [MobileCLIP paper](https://arxiv.org/abs/2311.17049) and along with the reinforced dataset [DataCompDR-12M](https://huggingface.co/datasets/apple/DataCompDR-12M).
         | 
| 35 | 
            -
            The UIDs per shards match between [mlfoundations/DataComp-12M](https://huggingface.co/datasets/mlfoundations/DataComp-12M) and [apple/DataCompDR-12M](https://huggingface.co/datasets/apple/DataCompDR-12M)
         | 
| 36 |  | 
| 37 | 
             
            ## Terms and Conditions
         | 
| 38 |  | 
| @@ -41,7 +41,7 @@ We have terms of service that are similar to those adopted by HuggingFace (https | |
| 41 |  | 
| 42 | 
             
            ## Citation
         | 
| 43 |  | 
| 44 | 
            -
            **[ | 
| 45 | 
             
            Gadre, Samir Yitzhak, et al.
         | 
| 46 | 
             
            ```
         | 
| 47 | 
             
            @article{gadre2024datacomp,
         | 
|  | |
| 18 | 
             
                - name: sha256
         | 
| 19 | 
             
                  dtype: string
         | 
| 20 | 
             
                - name: original_height
         | 
| 21 | 
            +
                  dtype: int32
         | 
| 22 | 
             
                - name: original_width
         | 
| 23 | 
            +
                  dtype: int32
         | 
| 24 | 
             
            ---
         | 
| 25 |  | 
| 26 | 
             
            # Dataset Card for DataComp-12M
         | 
|  | |
| 32 | 
             
            Image-text models trained on DataComp-12M are significantly better than on CC-12M/YFCC-15M as well as DataComp-Small/Medium.
         | 
| 33 |  | 
| 34 | 
             
            DataComp-12M was introduced in [MobileCLIP paper](https://arxiv.org/abs/2311.17049) and along with the reinforced dataset [DataCompDR-12M](https://huggingface.co/datasets/apple/DataCompDR-12M).
         | 
| 35 | 
            +
            The UIDs per shards match between [mlfoundations/DataComp-12M](https://huggingface.co/datasets/mlfoundations/DataComp-12M) and [apple/DataCompDR-12M](https://huggingface.co/datasets/apple/DataCompDR-12M).
         | 
| 36 |  | 
| 37 | 
             
            ## Terms and Conditions
         | 
| 38 |  | 
|  | |
| 41 |  | 
| 42 | 
             
            ## Citation
         | 
| 43 |  | 
| 44 | 
            +
            **[DataComp: In search of the next generation of multimodal datasets](https://arxiv.org/abs/2304.14108).  (NeurIPS 2024)**
         | 
| 45 | 
             
            Gadre, Samir Yitzhak, et al.
         | 
| 46 | 
             
            ```
         | 
| 47 | 
             
            @article{gadre2024datacomp,
         |