32health/non-ada-classification-dinov3

README.md

@@ -0,0 +1,78 @@
+---
+library_name: transformers
+license: other
+base_model: facebook/dinov3-vitb16-pretrain-lvd1689m
+tags:
+- generated_from_trainer
+metrics:
+- precision
+- recall
+- f1
+model-index:
+- name: output_dinov3_baseline
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# output_dinov3_baseline
+
+This model is a fine-tuned version of [facebook/dinov3-vitb16-pretrain-lvd1689m](https://huggingface.co/facebook/dinov3-vitb16-pretrain-lvd1689m) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.0540
+- Precision: 0.9657
+- Recall: 0.9748
+- F1: 0.9702
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-06
+- train_batch_size: 4
+- eval_batch_size: 4
+- seed: 42
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 32
+- optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 0.1
+- num_epochs: 10
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     |
+|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|
+| 0.4394        | 1.0   | 145  | 0.4057          | 0.6110    | 0.5578 | 0.5832 |
+| 0.2090        | 2.0   | 290  | 0.2168          | 0.8060    | 0.8992 | 0.8500 |
+| 0.1433        | 3.0   | 435  | 0.1430          | 0.9130    | 0.9265 | 0.9197 |
+| 0.1036        | 4.0   | 580  | 0.1033          | 0.9238    | 0.9674 | 0.9451 |
+| 0.0768        | 5.0   | 725  | 0.0816          | 0.9542    | 0.9632 | 0.9587 |
+| 0.0623        | 6.0   | 870  | 0.0682          | 0.9585    | 0.9706 | 0.9645 |
+| 0.0553        | 7.0   | 1015 | 0.0606          | 0.9606    | 0.9727 | 0.9666 |
+| 0.0490        | 8.0   | 1160 | 0.0576          | 0.9596    | 0.9727 | 0.9661 |
+| 0.0480        | 9.0   | 1305 | 0.0554          | 0.9656    | 0.9737 | 0.9697 |
+| 0.0482        | 10.0  | 1450 | 0.0540          | 0.9657    | 0.9748 | 0.9702 |
+
+
+### Framework versions
+
+- Transformers 5.0.0
+- Pytorch 2.10.0+cu128
+- Datasets 4.5.0
+- Tokenizers 0.22.2

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8f1668f55a8fe2861888d36ee4794fff2c896059db088a446b42fa3b903a2db
+size 345051000

preprocessor_config.json

@@ -0,0 +1,24 @@
+{
+  "data_format": "channels_first",
+  "default_to_square": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.485,
+    0.456,
+    0.406
+  ],
+  "image_processor_type": "DINOv3ViTImageProcessorFast",
+  "image_std": [
+    0.229,
+    0.224,
+    0.225
+  ],
+  "resample": 2,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "height": 224,
+    "width": 224
+  }
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93906609772f2b1afb026d8c1ddf7d7629227115361e3a69b9d45a25dca008c9
+size 5201