Upload folder using huggingface_hub

README.md

@@ -1,26 +1,44 @@
 ---
+license: cc-by-4.0
 tags:
-- ml-intern
+  - medical-imaging
+  - ultrasound
+  - thyroid
+  - classification
+  - efficientnet
+datasets:
+  - Johnyquest7/TN5000-thyroid-nodule-classification
 ---
 
-# Johnyquest7/Thyroid_EfficientNetV2
+# Thyroid Nodule Classification – EfficientNetV2-S
 
-<!-- ml-intern-provenance -->
-## Generated by ML Intern
+This model was trained on the TN5000 thyroid ultrasound dataset for binary classification of thyroid nodules (Benign vs Malignant).
 
-This model repository was generated by [ML Intern](https://github.com/huggingface/ml-intern), an agent for machine learning research and development on the Hugging Face Hub.
+## Training Configuration
 
-- Try ML Intern: https://smolagents-ml-intern.hf.space
-- Source code: https://github.com/huggingface/ml-intern
+| Parameter | Value |
+|-----------|-------|
+| Backbone | tf_efficientnetv2_s.in1k (timm, ImageNet pretrained) |
+| Input size | 384×384 |
+| Batch size | 64 |
+| Epochs | 32 (early stopped) |
+| LR head | 0.0001 |
+| LR backbone | 1e-05 |
+| Weight decay | 0.0001 |
+| Warmup | 3 epochs |
+| Scheduler | Cosine annealing |
+| EMA decay | 0.999 |
 
-## Usage
+## Test Set Performance
 
-```python
-from transformers import AutoModelForCausalLM, AutoTokenizer
+| Metric | Value | 95% CI |
+|--------|-------|--------|
+| Sensitivity | 0.9603 | [0.9435, 0.9733] |
+| Specificity | 0.0706 | [0.0431, 0.1081] |
+| PPV | 0.7374 | [0.7082, 0.7651] |
+| NPV | 0.3958 | [0.2577, 0.5473] |
+| AUC-ROC | 0.5663 | [0.5247, 0.6004] |
 
-model_id = "Johnyquest7/Thyroid_EfficientNetV2"
-tokenizer = AutoTokenizer.from_pretrained(model_id)
-model = AutoModelForCausalLM.from_pretrained(model_id)
-```
+## Citation
 
-For non-causal architectures, replace `AutoModelForCausalLM` with the appropriate `AutoModel` class.
+Yu, Xiaoxian et al. "TN5000: An Ultrasound Image Dataset for Thyroid Nodule Detection and Classification." Scientific Data (Nature), 2025.

best_model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2666f3a219c72cffc672b5b15f89209577ff5aa27e49d908b5032d363231ada
+size 324288274

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83909dfd687d03f24c77d2360b92ef2b155eb16cafce7f079953dc6bf7a28e3f
+size 81611074

results.json

@@ -0,0 +1,46 @@
+{
+  "best_epoch": 22,
+  "best_val_sensitivity": 0.9706666666666667,
+  "test_metrics": {
+    "sensitivity": 0.960328317373461,
+    "sensitivity_ci": [
+      0.9435196217687885,
+      0.9732730947107845
+    ],
+    "specificity": 0.07063197026022305,
+    "specificity_ci": [
+      0.043057945192480684,
+      0.1081048777072488
+    ],
+    "ppv": 0.7373949579831933,
+    "ppv_ci": [
+      0.7082097216248917,
+      0.7651033561693786
+    ],
+    "npv": 0.3958333333333333,
+    "npv_ci": [
+      0.2576989728464088,
+      0.5473041990893203
+    ],
+    "auc": 0.5663372982978961,
+    "auc_ci": [
+      0.5246876211230521,
+      0.6004399582613559
+    ],
+    "tp": 702,
+    "tn": 19,
+    "fp": 250,
+    "fn": 29,
+    "threshold": 0.5
+  },
+  "config": {
+    "backbone": "tf_efficientnetv2_s.in1k",
+    "img_size": 384,
+    "batch_size": 64,
+    "epochs_trained": 32,
+    "lr_head": 0.0001,
+    "lr_backbone": 1e-05,
+    "weight_decay": 0.0001,
+    "ema_decay": 0.999
+  }
+}