Instructions to use NbAiLab/autocrop-tekst with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use NbAiLab/autocrop-tekst with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-segmentation", model="NbAiLab/autocrop-tekst")# Load model directly from transformers import AutoImageProcessor, SegformerForSemanticSegmentation processor = AutoImageProcessor.from_pretrained("NbAiLab/autocrop-tekst") model = SegformerForSemanticSegmentation.from_pretrained("NbAiLab/autocrop-tekst") - Notebooks
- Google Colab
- Kaggle
End of training
Browse files- README.md +9 -7
- all_results.json +19 -0
- eval_results.json +14 -0
- train_results.json +8 -0
- trainer_state.json +0 -0
README.md
CHANGED
|
@@ -3,6 +3,8 @@ library_name: transformers
|
|
| 3 |
license: other
|
| 4 |
base_model: nvidia/mit-b0
|
| 5 |
tags:
|
|
|
|
|
|
|
| 6 |
- generated_from_trainer
|
| 7 |
datasets:
|
| 8 |
- generator
|
|
@@ -16,16 +18,16 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
# autocrop-tekst
|
| 18 |
|
| 19 |
-
This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on the
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
-
- Mean Iou: 0.
|
| 23 |
-
- Mean Accuracy: 0.
|
| 24 |
-
- Overall Accuracy: 0.
|
| 25 |
- Accuracy Background: nan
|
| 26 |
-
- Accuracy Crop: 0.
|
| 27 |
- Iou Background: 0.0
|
| 28 |
-
- Iou Crop: 0.
|
| 29 |
|
| 30 |
## Model description
|
| 31 |
|
|
|
|
| 3 |
license: other
|
| 4 |
base_model: nvidia/mit-b0
|
| 5 |
tags:
|
| 6 |
+
- image-segmentation
|
| 7 |
+
- vision
|
| 8 |
- generated_from_trainer
|
| 9 |
datasets:
|
| 10 |
- generator
|
|
|
|
| 18 |
|
| 19 |
# autocrop-tekst
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on the /mnt/disk1/autocrop-data/datasets/tekst dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 0.0197
|
| 24 |
+
- Mean Iou: 0.4970
|
| 25 |
+
- Mean Accuracy: 0.9939
|
| 26 |
+
- Overall Accuracy: 0.9939
|
| 27 |
- Accuracy Background: nan
|
| 28 |
+
- Accuracy Crop: 0.9939
|
| 29 |
- Iou Background: 0.0
|
| 30 |
+
- Iou Crop: 0.9939
|
| 31 |
|
| 32 |
## Model description
|
| 33 |
|
all_results.json
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_accuracy_background": NaN,
|
| 4 |
+
"eval_accuracy_crop": 0.9939011345509936,
|
| 5 |
+
"eval_iou_background": 0.0,
|
| 6 |
+
"eval_iou_crop": 0.9939011345509936,
|
| 7 |
+
"eval_loss": 0.019662605598568916,
|
| 8 |
+
"eval_mean_accuracy": 0.9939011345509936,
|
| 9 |
+
"eval_mean_iou": 0.4969505672754968,
|
| 10 |
+
"eval_overall_accuracy": 0.9939011345509936,
|
| 11 |
+
"eval_runtime": 35.7077,
|
| 12 |
+
"eval_samples_per_second": 24.673,
|
| 13 |
+
"eval_steps_per_second": 3.109,
|
| 14 |
+
"total_flos": 4.3705957092950016e+18,
|
| 15 |
+
"train_loss": 0.02810766745263185,
|
| 16 |
+
"train_runtime": 4914.0636,
|
| 17 |
+
"train_samples_per_second": 50.742,
|
| 18 |
+
"train_steps_per_second": 6.349
|
| 19 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_accuracy_background": NaN,
|
| 4 |
+
"eval_accuracy_crop": 0.9939011345509936,
|
| 5 |
+
"eval_iou_background": 0.0,
|
| 6 |
+
"eval_iou_crop": 0.9939011345509936,
|
| 7 |
+
"eval_loss": 0.019662605598568916,
|
| 8 |
+
"eval_mean_accuracy": 0.9939011345509936,
|
| 9 |
+
"eval_mean_iou": 0.4969505672754968,
|
| 10 |
+
"eval_overall_accuracy": 0.9939011345509936,
|
| 11 |
+
"eval_runtime": 35.7077,
|
| 12 |
+
"eval_samples_per_second": 24.673,
|
| 13 |
+
"eval_steps_per_second": 3.109
|
| 14 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"total_flos": 4.3705957092950016e+18,
|
| 4 |
+
"train_loss": 0.02810766745263185,
|
| 5 |
+
"train_runtime": 4914.0636,
|
| 6 |
+
"train_samples_per_second": 50.742,
|
| 7 |
+
"train_steps_per_second": 6.349
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|