add data

UX-Decoder · Jan 22, 2024 · b82dd16 · b82dd16
1 parent 74f1b30
commit b82dd16
Show file tree

Hide file tree

Showing 3 changed files with 35 additions and 5 deletions.
diff --git a/README.md b/README.md
@@ -18,6 +18,7 @@
     - [Install](#install)
     - [LLaVA-Grounding Weights](#llava-grounding-weights)
     - [Demo](#demo)
+    - [Training data](#training-data)
     - [Training](#training)
     - [Citation](#citation)
 
@@ -54,6 +55,35 @@ CUDA_VISIBLE_DEVICES=0 python gradio_demo/LLaVA_G_Demo.py --path_vision_cfg conf
 ```
 
 Please refer to our [Online Demo](https://llava-grounding.deepdataspace.com/) for the more detailed user's guidence.
+### Training data
+```text
+data
+├── flickr30k_entities
+│   ├── train/
+│   ├── val/
+│   ├── annotations
+│          ├──final_flickr_separateGT_train.json
+│          ├──final_flickr_separateGT_val.json
+├── coco
+│   ├── train2014/
+│   ├── train2017/
+│   ├── panoptic_train2017/
+│   ├── panoptic_semseg_train2017/
+│   ├── annotations
+│   │      ├──instances_train2017.json
+│   │      ├──instances_train2017_gvc.json
+│   │      ├──grounded_visual_chat_data.json
+│   │      ├──instances_train2014_filter.json
+│   │      ├──panoptic_train2017_filter.json
+│   │      ├──grounding_train2017.json
+├── llava
+│   ├── annotations
+│          ├── cap600k_brackets_all.json
+│          ├── llava_instruct_150k.json
+│          ├── llava_instruct_150k_visual_prompt.json
+
+
+```
 
 ### Training
 Stage 1

diff --git a/datasets_os/registration/register_coco_instruct_grounding_dataset.py b/datasets_os/registration/register_coco_instruct_grounding_dataset.py
@@ -21,14 +21,14 @@
 
     "coco_instruct_train_v3": (
         "coco/train2014", # image_root
-        "coco/annotations/instances_train2017_instruct_grounding_fix_v3.json", # annot_root
-        "llava/annotations/merged_v3_fixed.json",
+        "coco/annotations/instances_train2017_gvc.json", # annot_root
+        "llava/annotations/grounded_visual_chat_data.json",
     ),
 
     "coco_interactive": (
         "coco/train2014", # image_root
         "coco/annotations/instances_train2014_filter.json", # annot_root
-        "llava/annotations/llava_instruct_150k_instr.json",
+        "llava/annotations/llava_instruct_150k_visual_prompt.json",
     ),
     "coco_interactive_refcoco": (
         "coco/train2017", # image_root

diff --git a/datasets_os/registration/register_flickr_dataset.py b/datasets_os/registration/register_flickr_dataset.py
@@ -16,11 +16,11 @@
 _PREDEFINED_SPLITS = {
 
     "flickr_val": (
-        "goldg/flickr30k_entities/val", # image_root
+        "flickr30k_entities/val", # image_root
         "final_flickr_separateGT_val.json", # # anno_path
     ),
     "flickr_train": (
-        "goldg/flickr30k_entities/train", # image_root
+        "flickr30k_entities/train", # image_root
         "final_flickr_separateGT_train.json", # # anno_path
     ),
 }