Spaces:
Paused
Paused
Merge branch 'main' of https://github.com/dvlab-research/LISA into main
Browse filesFormer-commit-id: fb0cd1ebe589199878152b5b50202a713139d899
README.md
CHANGED
|
@@ -93,6 +93,11 @@ Important keys contained in JSON files:
|
|
| 93 |
|
| 94 |
The elements of the "shapes" exhibit two categories, namely **"target"** and **"ignore"**. The former category is indispensable for evaluation, while the latter category denotes the ambiguous region and hence disregarded during the evaluation process.
|
| 95 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 96 |
Besides, we leveraged GPT-3.5 for rephrasing instructions, so images in the training set may have **more than one instructions (but fewer than six)** in the "text" field. Users can randomly select one instruction as the text query to obtain a better model.
|
| 97 |
|
| 98 |
|
|
|
|
| 93 |
|
| 94 |
The elements of the "shapes" exhibit two categories, namely **"target"** and **"ignore"**. The former category is indispensable for evaluation, while the latter category denotes the ambiguous region and hence disregarded during the evaluation process.
|
| 95 |
|
| 96 |
+
We provide a script that demonstrates how to process the annotations.
|
| 97 |
+
```
|
| 98 |
+
python3 utils/data_proc_demo.py
|
| 99 |
+
```
|
| 100 |
+
|
| 101 |
Besides, we leveraged GPT-3.5 for rephrasing instructions, so images in the training set may have **more than one instructions (but fewer than six)** in the "text" field. Users can randomly select one instruction as the text query to obtain a better model.
|
| 102 |
|
| 103 |
|