Add EfficientNet-B2 defect classifier: weights, model card, eval artifacts

LICENSE

@@ -0,0 +1,10 @@
+These model weights were fine-tuned on sample imagery provided by Intel Corporation for the Semiconductor
+Solutions Challenge 2026 (Problem A: Small-Sample Learning for Defect Classification). The underlying
+EfficientNet-B2 backbone (torchvision, ImageNet-pretrained) is used under its original license.
+
+This checkpoint is shared for educational and research purposes as a challenge submission artifact. It is not
+an official Intel product, and no rights to the underlying Intel-provided dataset are granted beyond what the
+challenge organizers permit. Contact the repository owner before any commercial use.
+
+Intel and the Intel logo are trademarks of Intel Corporation or its subsidiaries in the U.S. and/or other
+countries.

README.md

@@ -0,0 +1,165 @@
+---
+license: other
+license_name: intel-challenge-dataset
+license_link: LICENSE
+tags:
+  - pytorch
+  - image-classification
+  - efficientnet
+  - defect-detection
+  - defect-classification
+  - semiconductor
+  - wafer-inspection
+  - manufacturing
+  - few-shot-learning
+  - small-sample-learning
+  - computer-vision
+library_name: pytorch
+pipeline_tag: image-classification
+metrics:
+  - accuracy
+  - f1
+model-index:
+  - name: defect-vision-efficientnet-b2
+    results:
+      - task:
+          type: image-classification
+        metrics:
+          - type: accuracy
+            value: 0.9556
+            name: Test accuracy
+          - type: accuracy
+            value: 0.975
+            name: Best validation accuracy
+---
+
+# Defect Vision: EfficientNet-B2 Semiconductor Defect Classifier
+
+Fine-tuned EfficientNet-B2 for **small-sample wafer defect classification**, built for the **Intel
+Semiconductor Solutions Challenge 2026, Problem A: Small-Sample Learning for Defect Classification**.
+
+Classifies gray-scale wafer/die images into **8 defect classes + "no defect"** (9-way), trained on a
+class-balanced, heavily-augmented small dataset rather than large-scale labeled data. The challenge's core
+constraint is that production defect data is scarce and imbalanced.
+
+- **Code, FastAPI service, React demo UI, training notebook:** https://github.com/Sehastrajit-S/defect-vision
+- **Backbone:** `torchvision.models.efficientnet_b2` (ImageNet-pretrained), custom classifier head
+- **Params:** ~9.2M
+- **Input:** 260×260 RGB (gray-scale images converted to 3-channel), ImageNet normalization
+
+## Results
+
+| Metric | Target (challenge brief) | Achieved |
+|---|---|---|
+| Overall classification accuracy | ~85% | **95.6%** (test, 360 held-out images) |
+| Best validation accuracy | n/a | **97.5%** |
+| Inference latency | ~1s/image | ~40–500ms/image (GPU), ~0.1–1s (CPU) |
+
+<details>
+<summary>Full per-class classification report (test set)</summary>
+
+```text
+Test Loss : 0.6153  |  Test Accuracy : 0.9556
+
+              precision    recall  f1-score   support
+
+     defect1     0.9773    0.9556    0.9663        45
+     defect2     0.9375    1.0000    0.9677        45
+     defect3     1.0000    1.0000    1.0000        45
+     defect4     1.0000    1.0000    1.0000        45
+     defect5     0.9130    0.9333    0.9231        45
+     defect8     0.8837    0.8444    0.8636        45
+     defect9     0.9556    0.9556    0.9556        45
+    defect10     0.9773    0.9556    0.9663        45
+    new_good     0.0000    0.0000    0.0000         0
+
+    accuracy                         0.9556       360
+   macro avg     0.8494    0.8494    0.8492       360
+weighted avg     0.9555    0.9556    0.9553       360
+```
+
+`new_good` (no defect) has zero held-out samples in this dataset revision. The 9th output neuron is reserved
+for future "no defect found" imagery without requiring re-architecture.
+
+</details>
+
+![Confusion matrix](confusion_matrix_test.png)
+![Training curves](training_curves.png)
+
+## Handling class imbalance with few samples
+
+- **Class-balanced dataset construction**: equal train/val/test counts per class (210/45/45) via augmentation,
+  instead of naive minority oversampling or loss reweighting, so the model never learns a majority-class prior.
+- **Aggressive augmentation**: random crop, flips, rotation, perspective warp, and color jitter multiply the
+  small per-class sample count without duplicating exact pixels.
+- **Label smoothing (0.1)** on cross-entropy keeps the model from over-committing on visually similar defect
+  types.
+- **OneCycleLR + early stopping** (patience 7) for fast, stable convergence on limited data. This checkpoint
+  converged and early-stopped at epoch 16.
+
+## Usage
+
+```python
+import torch
+import torch.nn as nn
+from torchvision import models, transforms
+from PIL import Image
+from huggingface_hub import hf_hub_download
+
+CLASSES = ["defect1", "defect2", "defect3", "defect4", "defect5",
+           "defect8", "defect9", "defect10", "new_good"]
+
+def build_model(num_classes: int) -> nn.Module:
+    model = models.efficientnet_b2(weights=None)
+    in_f = model.classifier[1].in_features
+    model.classifier = nn.Sequential(
+        nn.Dropout(p=0.4),
+        nn.Linear(in_f, 512),
+        nn.SiLU(inplace=True),
+        nn.Dropout(p=0.3),
+        nn.Linear(512, num_classes),
+    )
+    return model
+
+weights_path = hf_hub_download(repo_id="Sehastrajit/defect-vision-efficientnet-b2", filename="best_model.pth")
+model = build_model(len(CLASSES))
+ckpt = torch.load(weights_path, map_location="cpu", weights_only=False)
+model.load_state_dict(ckpt["model_state"])
+model.eval()
+
+transform = transforms.Compose([
+    transforms.Resize((260, 260)),
+    transforms.ToTensor(),
+    transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225]),
+])
+
+img = Image.open("wafer_sample.png").convert("RGB")
+x = transform(img).unsqueeze(0)
+with torch.no_grad():
+    probs = torch.softmax(model(x), dim=1)[0]
+
+pred = CLASSES[probs.argmax().item()]
+print(pred, probs.max().item())
+```
+
+## Training setup
+
+| | |
+|---|---|
+| GPU | NVIDIA RTX 3060 12GB (fp16 AMP) |
+| Optimizer | AdamW, lr 2e-4, weight decay 1e-4 |
+| Schedule | OneCycleLR, cosine anneal |
+| Batch | 64 × 2 grad-accum steps (effective 128) |
+| Split | 70% train / 15% val / 15% test |
+| Epochs | early-stopped at 16 (patience 7) |
+
+Full training script: [`h1.ipynb`](https://github.com/Sehastrajit-S/defect-vision/blob/main/src/app/h1.ipynb) in
+the main repo.
+
+## Intended use & limitations
+
+Built as a challenge submission demonstrating small-sample defect classification technique, not validated for
+production fab deployment. Trained on Intel-provided sample imagery for the Semiconductor Solutions Challenge
+2026; `new_good` has no held-out evaluation samples in this dataset revision. Intel and the Intel logo are
+trademarks of Intel Corporation or its subsidiaries. This is an independent student project, not an Intel
+product.

best_model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc5ac651cf9e661ca55942a795d1563d2758a0c0c3cbe8d50dc28319a3643856
+size 34163260

classification_report.txt

@@ -0,0 +1,17 @@
+Test Loss : 0.6153  |  Test Accuracy : 0.9556
+
+              precision    recall  f1-score   support
+
+     defect1     0.9773    0.9556    0.9663        45
+     defect2     0.9375    1.0000    0.9677        45
+     defect3     1.0000    1.0000    1.0000        45
+     defect4     1.0000    1.0000    1.0000        45
+     defect5     0.9130    0.9333    0.9231        45
+     defect8     0.8837    0.8444    0.8636        45
+     defect9     0.9556    0.9556    0.9556        45
+    defect10     0.9773    0.9556    0.9663        45
+    new_good     0.0000    0.0000    0.0000         0
+
+    accuracy                         0.9556       360
+   macro avg     0.8494    0.8494    0.8492       360
+weighted avg     0.9555    0.9556    0.9553       360

split_summary.json

@@ -0,0 +1,32 @@
+{
+  "train": {
+    "defect1": 210,
+    "defect2": 210,
+    "defect3": 210,
+    "defect4": 210,
+    "defect5": 210,
+    "defect8": 210,
+    "defect9": 210,
+    "defect10": 210
+  },
+  "val": {
+    "defect1": 45,
+    "defect2": 45,
+    "defect3": 45,
+    "defect4": 45,
+    "defect5": 45,
+    "defect8": 45,
+    "defect9": 45,
+    "defect10": 45
+  },
+  "test": {
+    "defect1": 45,
+    "defect2": 45,
+    "defect3": 45,
+    "defect4": 45,
+    "defect5": 45,
+    "defect8": 45,
+    "defect9": 45,
+    "defect10": 45
+  }
+}

training_history.json

@@ -0,0 +1,74 @@
+{
+  "train_loss": [
+    2.1933312461489725,
+    2.1553719974699472,
+    2.0721990971338182,
+    1.8812163693564279,
+    1.4302027429853166,
+    0.9784145218985422,
+    0.7884583825156802,
+    0.6902853545688448,
+    0.6232542821339199,
+    0.5995042261623201,
+    0.5878782681056431,
+    0.5768878437223889,
+    0.5563245608693077,
+    0.5441939581008185,
+    0.5425735553105672,
+    0.542737436862219
+  ],
+  "train_acc": [
+    0.1619047619047619,
+    0.2702380952380952,
+    0.44821428571428573,
+    0.655952380952381,
+    0.7428571428571429,
+    0.8154761904761905,
+    0.881547619047619,
+    0.9172619047619047,
+    0.9547619047619048,
+    0.9625,
+    0.9678571428571429,
+    0.9648809523809524,
+    0.9791666666666666,
+    0.9851190476190477,
+    0.9845238095238096,
+    0.9833333333333333
+  ],
+  "val_loss": [
+    2.176660336388482,
+    2.123828093210856,
+    2.002728467517429,
+    1.6516859107547337,
+    1.0096800830629138,
+    0.7460225237740411,
+    0.6595639175838894,
+    0.6276597248183357,
+    0.5659777522087097,
+    0.574584882789188,
+    0.5686059143808153,
+    0.5664335568745931,
+    0.5855459094047546,
+    0.595595407485962,
+    0.5847034321890937,
+    0.5646144628524781
+  ],
+  "val_acc": [
+    0.2361111111111111,
+    0.5,
+    0.7277777777777777,
+    0.7972222222222223,
+    0.8277777777777777,
+    0.9138888888888889,
+    0.9305555555555556,
+    0.9416666666666667,
+    0.975,
+    0.9638888888888889,
+    0.9611111111111111,
+    0.9694444444444444,
+    0.9611111111111111,
+    0.95,
+    0.9527777777777777,
+    0.9611111111111111
+  ]
+}