Onegafer
/

segformer-v-mesh-0

Image Segmentation

Generated from Trainer

Inference Endpoints

Model card Files Files and versions Community

Onegafer commited on Apr 25, 2023

Commit

8fe5900

·

1 Parent(s): e7fa281

update model card README.md

Files changed (1) hide show

README.md +8 -13

README.md CHANGED Viewed

@@ -16,14 +16,14 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [nvidia/mit-b2](https://huggingface.co/nvidia/mit-b2) on the Onegafer/vehicle_segmentation dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0513
-- Mean Iou: 0.4015
-- Mean Accuracy: 0.8030
-- Overall Accuracy: 0.8030
 - Accuracy No: nan
-- Accuracy Yes: 0.8030
 - Iou No: 0.0
-- Iou Yes: 0.8030
 ## Model description
@@ -48,18 +48,13 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 1
 ### Training results
 | Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Accuracy No | Accuracy Yes | Iou No | Iou Yes |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:-------------:|:----------------:|:-----------:|:------------:|:------:|:-------:|
-| 0.2362        | 0.16  | 20   | 0.2073          | 0.2925   | 0.5849        | 0.5849           | nan         | 0.5849       | 0.0    | 0.5849  |
-| 0.1487        | 0.31  | 40   | 0.1029          | 0.3193   | 0.6386        | 0.6386           | nan         | 0.6386       | 0.0    | 0.6386  |
-| 0.0993        | 0.47  | 60   | 0.0861          | 0.4281   | 0.8562        | 0.8562           | nan         | 0.8562       | 0.0    | 0.8562  |
-| 0.1559        | 0.62  | 80   | 0.0603          | 0.3784   | 0.7568        | 0.7568           | nan         | 0.7568       | 0.0    | 0.7568  |
-| 0.0675        | 0.78  | 100  | 0.0536          | 0.3889   | 0.7777        | 0.7777           | nan         | 0.7777       | 0.0    | 0.7777  |
-| 0.0826        | 0.94  | 120  | 0.0513          | 0.4015   | 0.8030        | 0.8030           | nan         | 0.8030       | 0.0    | 0.8030  |
 ### Framework versions

 This model is a fine-tuned version of [nvidia/mit-b2](https://huggingface.co/nvidia/mit-b2) on the Onegafer/vehicle_segmentation dataset.
 It achieves the following results on the evaluation set:
+- Loss: 0.2948
+- Mean Iou: 0.2111
+- Mean Accuracy: 0.4222
+- Overall Accuracy: 0.4222
 - Accuracy No: nan
+- Accuracy Yes: 0.4222
 - Iou No: 0.0
+- Iou Yes: 0.4222
 ## Model description
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
+- num_epochs: 0.2
 ### Training results
 | Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Accuracy No | Accuracy Yes | Iou No | Iou Yes |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|:-------------:|:----------------:|:-----------:|:------------:|:------:|:-------:|
+| 0.3775        | 0.16  | 20   | 0.2948          | 0.2111   | 0.4222        | 0.4222           | nan         | 0.4222       | 0.0    | 0.4222  |
 ### Framework versions