Instructions to use tue-mps/coco_instance_eomt_large_1280 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use tue-mps/coco_instance_eomt_large_1280 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-segmentation", model="tue-mps/coco_instance_eomt_large_1280")# Load model directly from transformers import AutoImageProcessor, EomtForUniversalSegmentation processor = AutoImageProcessor.from_pretrained("tue-mps/coco_instance_eomt_large_1280") model = EomtForUniversalSegmentation.from_pretrained("tue-mps/coco_instance_eomt_large_1280", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Upload EomtForUniversalSegmentation
Browse files- config.json +4 -0
- model.safetensors +3 -0
config.json
CHANGED
|
@@ -1,4 +1,7 @@
|
|
| 1 |
{
|
|
|
|
|
|
|
|
|
|
| 2 |
"attention_dropout": 0.0,
|
| 3 |
"class_weight": 2.0,
|
| 4 |
"dice_weight": 5.0,
|
|
@@ -188,6 +191,7 @@
|
|
| 188 |
"num_upscale_blocks": 2,
|
| 189 |
"oversample_ratio": 3.0,
|
| 190 |
"patch_size": 16,
|
|
|
|
| 191 |
"train_num_points": 12544,
|
| 192 |
"transformers_version": "4.54.0.dev0",
|
| 193 |
"use_swiglu_ffn": false
|
|
|
|
| 1 |
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"EomtForUniversalSegmentation"
|
| 4 |
+
],
|
| 5 |
"attention_dropout": 0.0,
|
| 6 |
"class_weight": 2.0,
|
| 7 |
"dice_weight": 5.0,
|
|
|
|
| 191 |
"num_upscale_blocks": 2,
|
| 192 |
"oversample_ratio": 3.0,
|
| 193 |
"patch_size": 16,
|
| 194 |
+
"torch_dtype": "float32",
|
| 195 |
"train_num_points": 12544,
|
| 196 |
"transformers_version": "4.54.0.dev0",
|
| 197 |
"use_swiglu_ffn": false
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7e405469a7c2e91d064eaad714249febb3cb33c77bab0f90501ac8d66e072b21
|
| 3 |
+
size 1286272632
|