Upload folder using huggingface_hub

README.md

@@ -1,20 +1,25 @@
 ---
 license: cc-by-4.0
 tags:
-- medical-imaging
-- ultrasound
-- thyroid
-- classification
-- efficientnet
-- ml-intern
+  - medical-imaging
+  - ultrasound
+  - thyroid
+  - classification
+  - efficientnet
 datasets:
-- Johnyquest7/TN5000-thyroid-nodule-classification
+  - Johnyquest7/TN5000-thyroid-nodule-classification
 ---
 
-# Thyroid Nodule Classification – EfficientNetV2-S (AUC-Optimized)
+# Thyroid Nodule Classification – EfficientNetV2-S (AUC-Optimized v2)
 
 This model was trained on the TN5000 thyroid ultrasound dataset for binary classification of thyroid nodules (Benign vs Malignant), **optimized for AUC-ROC**.
 
+## Key Training Decisions
+
+- **No class weighting in BCE loss**: AUC is a ranking metric; class weighting distorts probability calibration and hurts discriminative power.
+- **WeightedRandomSampler**: Used to ensure balanced batches during training.
+- **EMA decay 0.999**: For stable validation/test predictions.
+
 ## Training Configuration
 
 | Parameter | Value |
@@ -29,6 +34,7 @@ This model was trained on the TN5000 thyroid ultrasound dataset for binary class
 | Warmup | 3 epochs |
 | Scheduler | Cosine annealing |
 | EMA decay | 0.999 |
+| Loss | BCE (no pos_weight) |
 | Optimization metric | AUC-ROC |
 
 ## Test Set Performance
@@ -36,31 +42,11 @@ This model was trained on the TN5000 thyroid ultrasound dataset for binary class
 | Metric | Value | 95% CI |
 |--------|-------|--------|
 | Sensitivity | 0.9111 | [0.8881, 0.9307] |
-| Specificity | 0.1078 | [0.0734, 0.1511] |
-| PPV | 0.7351 | [0.7051, 0.7636] |
-| NPV | 0.3085 | [0.2173, 0.4122] |
-| AUC-ROC | 0.5201 | [0.4779, 0.5554] |
+| Specificity | 0.1115 | [0.0765, 0.1554] |
+| PPV | 0.7359 | [0.7059, 0.7644] |
+| NPV | 0.3158 | [0.2242, 0.4192] |
+| AUC-ROC | 0.5195 | [0.4772, 0.5549] |
 
 ## Citation
 
 Yu, Xiaoxian et al. "TN5000: An Ultrasound Image Dataset for Thyroid Nodule Detection and Classification." Scientific Data (Nature), 2025.
-
-<!-- ml-intern-provenance -->
-## Generated by ML Intern
-
-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.
-
-- Try ML Intern: https://smolagents-ml-intern.hf.space
-- Source code: https://github.com/huggingface/ml-intern
-
-## Usage
-
-```python
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-model_id = 'Johnyquest7/Thyroid_EfficientNetV2'
-tokenizer = AutoTokenizer.from_pretrained(model_id)
-model = AutoModelForCausalLM.from_pretrained(model_id)
-```
-
-For non-causal architectures, replace `AutoModelForCausalLM` with the appropriate `AutoModel` class.

best_model.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5002cd1ce3073ba44ca4e841529042b5b3d640038ae079067be36629ca7044c1
+oid sha256:15482835c21c210cee07fa1abcfa57738e34d45140074311f91b883e55fd7c98
 size 324288274

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec00324575fa05cffc06e62b9825294126b56359de21e1737c6f75c4df665713
+oid sha256:2ff41a1f37ee9f32c2d78c0371fd7cc230ec8187344f4d72c8c7c692d9ce5d65
 size 81611074

results.json

@@ -1,35 +1,35 @@
 {
   "best_epoch": 1,
-  "best_val_auc": 0.482752,
+  "best_val_auc": 0.4818346666666667,
   "test_metrics": {
     "sensitivity": 0.9110807113543091,
     "sensitivity_ci": [
       0.8880679545923451,
       0.930704677143796
     ],
-    "specificity": 0.10780669144981413,
+    "specificity": 0.11152416356877323,
     "specificity_ci": [
-      0.0733982286883636,
-      0.15114930580486097
+      0.07652180042423258,
+      0.155368432564055
     ],
-    "ppv": 0.7350993377483444,
+    "ppv": 0.7359116022099448,
     "ppv_ci": [
-      0.7050788339309577,
-      0.7635823335148073
+      0.7058993124376329,
+      0.7643793122110114
     ],
-    "npv": 0.30851063829787234,
+    "npv": 0.3157894736842105,
     "npv_ci": [
-      0.2172931496088876,
-      0.41218276665696707
+      0.22421979057887004,
+      0.41920736370136785
     ],
-    "auc": 0.5200774007190843,
+    "auc": 0.5194773163004287,
     "auc_ci": [
-      0.4778848439216083,
-      0.5553833568557005
+      0.4771984664530449,
+      0.5548698473128044
     ],
     "tp": 666,
-    "tn": 29,
-    "fp": 240,
+    "tn": 30,
+    "fp": 239,
     "fn": 65,
     "threshold": 0.5
   },
@@ -42,6 +42,7 @@
     "lr_backbone": 1e-05,
     "weight_decay": 0.0001,
     "ema_decay": 0.999,
-    "optimization_metric": "auc"
+    "optimization_metric": "auc",
+    "pos_weight": 1.0
   }
 }