Gregor commited on
Commit
a487dc3
1 Parent(s): 7faac8a

Updated image information

Browse files
.gitattributes CHANGED
@@ -70,3 +70,4 @@ vqa/vqav2_vqa_mt.json filter=lfs diff=lfs merge=lfs -text
70
  vqa/vqav2_vqg_mt.json filter=lfs diff=lfs merge=lfs -text
71
  xgqa/xgqa_input_train.json filter=lfs diff=lfs merge=lfs -text
72
  xvnli/xvnli_input_train.json filter=lfs diff=lfs merge=lfs -text
 
 
70
  vqa/vqav2_vqg_mt.json filter=lfs diff=lfs merge=lfs -text
71
  xgqa/xgqa_input_train.json filter=lfs diff=lfs merge=lfs -text
72
  xvnli/xvnli_input_train.json filter=lfs diff=lfs merge=lfs -text
73
+ ccs_synthetic_filtered_large_2273005_raw.json filter=lfs diff=lfs merge=lfs -text
README.md CHANGED
@@ -29,11 +29,20 @@ af, am, ar, az, be, bg, bn, ca, ceb, cs, cy, da, de, el, en, eo, es, et, eu, fa,
29
  Languages are translated proportional to their size in [mC4](https://www.tensorflow.org/datasets/catalog/c4#c4multilingual), i.e., as 6% of examples in mC4 are German, we translate 6% of the data to German.
30
 
31
 
 
 
32
  **Dataset structure:**
33
  - `task_mix_mt.json`: The instruction mix data in the processed, translated, and combined form.
34
  - Folders: The folders contain 1) the separate tasks used to generate the mix
35
  and 2) the files of the tasks used to evaluate the model.
36
 
 
 
 
 
 
 
 
37
 
38
  **License:**
39
  Must comply with license of the original datasets used to create this mix. See https://github.com/gregor-ge/mBLIP/data/README.md for more.
 
29
  Languages are translated proportional to their size in [mC4](https://www.tensorflow.org/datasets/catalog/c4#c4multilingual), i.e., as 6% of examples in mC4 are German, we translate 6% of the data to German.
30
 
31
 
32
+
33
+
34
  **Dataset structure:**
35
  - `task_mix_mt.json`: The instruction mix data in the processed, translated, and combined form.
36
  - Folders: The folders contain 1) the separate tasks used to generate the mix
37
  and 2) the files of the tasks used to evaluate the model.
38
 
39
+ **Images:**
40
+ We do not include any images with this dataset.
41
+ Images from the public datasets (MSCOCO for instruction training, and others for evaluation) can be downloaded
42
+ from the respective websites.
43
+ For the BLIP captions, we provide the URLs and filenames as used by us [here](blip_captions/ccs_synthetic_filtered_large_2273005_raw.json).
44
+ To download them, [our code](https://github.com/gregor-ge/mBLIP/tree/main/data#blip-web-capfilt) can be adapted, for example.
45
+
46
 
47
  **License:**
48
  Must comply with license of the original datasets used to create this mix. See https://github.com/gregor-ge/mBLIP/data/README.md for more.
ccs_synthetic_filtered_large_2273005_raw.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23229e1e500ceb826fcd865f4b2138dda57203b444137dbe4a132c155a78e01a
3
+ size 506886927