Push model using huggingface_hub.
Browse files- README.md +62 -0
- config.json +15 -0
- model.safetensors +3 -0
README.md
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: segmentation-models-pytorch
|
| 3 |
+
license: mit
|
| 4 |
+
pipeline_tag: image-segmentation
|
| 5 |
+
tags:
|
| 6 |
+
- semantic-segmentation
|
| 7 |
+
- pytorch
|
| 8 |
+
- segmentation-models-pytorch
|
| 9 |
+
languages:
|
| 10 |
+
- python
|
| 11 |
+
---
|
| 12 |
+
# FPN Model Card
|
| 13 |
+
|
| 14 |
+
Table of Contents:
|
| 15 |
+
- [Load trained model](#load-trained-model)
|
| 16 |
+
- [Model init parameters](#model-init-parameters)
|
| 17 |
+
- [Model metrics](#model-metrics)
|
| 18 |
+
- [Dataset](#dataset)
|
| 19 |
+
|
| 20 |
+
## Load trained model
|
| 21 |
+
```python
|
| 22 |
+
import segmentation_models_pytorch as smp
|
| 23 |
+
|
| 24 |
+
model = smp.FPN.from_pretrained("GID-segmentation-FPN_resnet101")
|
| 25 |
+
```
|
| 26 |
+
|
| 27 |
+
## Model init parameters
|
| 28 |
+
```python
|
| 29 |
+
model_init_params = {
|
| 30 |
+
"encoder_name": "resnet101",
|
| 31 |
+
"encoder_depth": 5,
|
| 32 |
+
"encoder_weights": "imagenet",
|
| 33 |
+
"decoder_pyramid_channels": 256,
|
| 34 |
+
"decoder_segmentation_channels": 128,
|
| 35 |
+
"decoder_merge_policy": "add",
|
| 36 |
+
"decoder_dropout": 0.2,
|
| 37 |
+
"in_channels": 4,
|
| 38 |
+
"classes": 1,
|
| 39 |
+
"activation": None,
|
| 40 |
+
"upsampling": 4,
|
| 41 |
+
"aux_params": None
|
| 42 |
+
}
|
| 43 |
+
```
|
| 44 |
+
|
| 45 |
+
## Model metrics
|
| 46 |
+
```json
|
| 47 |
+
[
|
| 48 |
+
{
|
| 49 |
+
"test_per_image_iou": 0.6290831565856934,
|
| 50 |
+
"test_dataset_iou": 0.7467864155769348
|
| 51 |
+
}
|
| 52 |
+
]
|
| 53 |
+
```
|
| 54 |
+
|
| 55 |
+
## Dataset
|
| 56 |
+
Dataset name: GID
|
| 57 |
+
|
| 58 |
+
## More Information
|
| 59 |
+
- Library: https://github.com/qubvel/segmentation_models.pytorch
|
| 60 |
+
- Docs: https://smp.readthedocs.io/en/latest/
|
| 61 |
+
|
| 62 |
+
This model has been pushed to the Hub using the [PytorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin)
|
config.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_model_class": "FPN",
|
| 3 |
+
"activation": null,
|
| 4 |
+
"aux_params": null,
|
| 5 |
+
"classes": 1,
|
| 6 |
+
"decoder_dropout": 0.2,
|
| 7 |
+
"decoder_merge_policy": "add",
|
| 8 |
+
"decoder_pyramid_channels": 256,
|
| 9 |
+
"decoder_segmentation_channels": 128,
|
| 10 |
+
"encoder_depth": 5,
|
| 11 |
+
"encoder_name": "resnet101",
|
| 12 |
+
"encoder_weights": "imagenet",
|
| 13 |
+
"in_channels": 4,
|
| 14 |
+
"upsampling": 4
|
| 15 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a12bf46c1165baff38feefca4b298ac26b9cb9c68e0be881629b261347a586d
|
| 3 |
+
size 180932628
|