Instructions to use FoundationVision/groma-7b-pretrain with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use FoundationVision/groma-7b-pretrain with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("FoundationVision/groma-7b-pretrain", dtype="auto") - Notebooks
- Google Colab
- Kaggle
Update config.json
Browse files- config.json +2 -0
config.json
CHANGED
|
@@ -2,7 +2,9 @@
|
|
| 2 |
"architectures": [
|
| 3 |
"GromaModel"
|
| 4 |
],
|
|
|
|
| 5 |
"box_score_thres": 0.15,
|
|
|
|
| 6 |
"llm_cfg": {
|
| 7 |
"_name_or_path": "vicuna-7b-v1.5",
|
| 8 |
"architectures": [
|
|
|
|
| 2 |
"architectures": [
|
| 3 |
"GromaModel"
|
| 4 |
],
|
| 5 |
+
"nms_thres": 0.6,
|
| 6 |
"box_score_thres": 0.15,
|
| 7 |
+
"max_region_num": 100,
|
| 8 |
"llm_cfg": {
|
| 9 |
"_name_or_path": "vicuna-7b-v1.5",
|
| 10 |
"architectures": [
|