Publish FactoryVision AI model package (public review)

README.md

@@ -9,46 +9,71 @@ tags:
 - casting-defect-detection
 - education
 ---
-# FactoryVision AI — Custom CNN for Casting Inspection
+# 🏭 FactoryVision AI
+## Casting Inspection · Custom CNN
 
-**FactoryVision AI — Credits and Attribution**
+> **Educational prototype** · Binary image classification · **Human-reviewed inspection only**
 
-**Original Developer & Model Author:** Theivanth Rajamohan  
-**Technical Review, Code Contributions & Model Publication Support:** [Swapin Vidya](https://huggingface.co/Swapinvidya)  
-**Project Contributor:** [Rajah Shanmugasundaram](https://huggingface.co/rajahshan)  
-**Publishing Organisation:** TechSphere Consultancy Services Private Limited  
-**Hugging Face:** https://huggingface.co/TechsphereAI  
-**Official Website:** https://techsphereconsultancy.com/
+| Model | Task | Framework | Weights |
+|---|---|---|---|
+| FactoryVision AI Custom CNN | Casting defect classification | PyTorch | `model.pth` |
 
-**Project Provenance:** FactoryVision AI was originally developed by Theivanth Rajamohan. [Swapin Vidya](https://huggingface.co/Swapinvidya) contributed technical review, detection-metrics improvements, testing, model publication preparation and documentation. The models are being prepared for publication under TechSphere Consultancy Services Private Limited's Hugging Face namespace, with the original developer's permission.
+## 📊 Held-out test results
 
-**Original GitHub Repository:** https://github.com/Theivanth/Factory-Vision-AI  
-**Merged contribution by [Swapin Vidya](https://huggingface.co/Swapinvidya):** https://github.com/Theivanth/Factory-Vision-AI/pull/1
+| Metric | Reported result |
+|---|---:|
+| Accuracy | **0.9860** |
+| Macro precision | **0.9849** |
+| Macro recall | **0.9849** |
+| Macro F1 | **0.9849** |
+| Mean inference latency | **0.608 ms** |
 
-## Purpose and limitations
-Educational two-class classification of casting images (`def_front` = defective; `ok_front` = acceptable). For **human-reviewed prototype inspection only**, not an industrial release decision or certified quality system.
+**Confusion matrix** — rows are true labels, columns are predictions; class order: `def_front`, `ok_front`.
 
-## Architecture and files
-- Custom PyTorch CNN trained from scratch; see the source repository's `src/quality.py` for architecture and preprocessing
-- Weights: `model.pth` (archive `checkpoints/casting_cnn.pth`)
-- Exact ordered labels in `labels.json`
+```text
+[[448, 5],
+ [  5, 257]]
+```
+
+Metrics and latency are reported from the supplied Kaggle output, not independently reproduced. Latency is environment/measurement specific and is not a target CPU benchmark. The model is not validated on new factory deployments.
+
+## 🎯 Purpose and limitations
+
+Educational two-class classification of casting images (`def_front` = defective; `ok_front` = acceptable). For **human-reviewed prototype inspection only**; not an industrial release decision or certified quality system.
+
+## 🧠 Model details
+
+- Custom PyTorch CNN trained from scratch; see `src/quality.py` in the source repository for architecture and preprocessing
+- Packaged weights: `model.pth` (archive path `checkpoints/casting_cnn.pth`)
+- Exact ordered labels: `labels.json`
 - Use the project's original loader and preprocessing rather than assuming a generic Hugging Face Transformer loader can use the `.pth` file.
 
-## Training dataset
-Real-Life Industrial Dataset of Casting Product by Ravirajsinh Dabhi (https://www.kaggle.com/datasets/ravirajsinh45/real-life-industrial-dataset-of-casting-product). The source README identifies its Kaggle licence as **CC BY-NC-ND 4.0**. It includes non-commercial and no-derivatives restrictions. The rights to publish dataset-trained checkpoint weights require separate verification; model-developer consent does not settle third-party licensing. No input images are redistributed here.
+## 🗂️ Training data
+
+Real-Life Industrial Dataset of Casting Product by Ravirajsinh Dabhi ([Kaggle dataset](https://www.kaggle.com/datasets/ravirajsinh45/real-life-industrial-dataset-of-casting-product)). The source README identifies its Kaggle licence as **CC BY-NC-ND 4.0**, including non-commercial and no-derivatives restrictions. Rights to publish dataset-trained checkpoint weights require separate verification; model-developer consent does not settle third-party licensing. No input images are redistributed here.
 
-## Held-out test performance (from supplied Kaggle output)
-- Accuracy: **0.9860**
-- Macro precision: **0.9849**
-- Macro recall: **0.9849**
-- Macro F1: **0.9849**
-- Test confusion matrix (rows true, columns predicted; class order `def_front`, `ok_front`): `[[448, 5], [5, 257]]`
-- Mean inference latency reported by Kaggle: **0.608 ms** (environment/measurement specific; not a target CPU benchmark)
+## 🔧 Inference and dependencies
 
-The metrics are reported, not independently reproduced. The model is not validated on new factory deployments.
+The checkpoint is intended to be loaded using the matching architecture from [`src/quality.py`](https://github.com/Theivanth/Factory-Vision-AI) in the source repository. Do **not** call arbitrary `torch.load` on untrusted checkpoints. The project has known image-size, normalisation, and class-order requirements; check the current source for the exact implementation before inference.
 
-## Inference and dependencies
-The checkpoint is intended to be loaded using the matching architecture from `src/quality.py` in https://github.com/Theivanth/Factory-Vision-AI . Do **not** call arbitrary `torch.load` on untrusted checkpoints. The project has known image-size, normalisation, and class-order requirements; check the current source for exact implementation before inference.
+## ⚖️ Licensing and distribution
 
-## Licensing and distribution
 The source project has an MIT licence for **its code**, not necessarily the checkpoint. **Keep this repository private until the casting dataset's non-commercial/no-derivatives conditions and trained-weight publication rights have been resolved.** No model-weight licence is asserted in this card.
+
+---
+
+## 🤝 Credits and attribution
+
+| Role | Contributor |
+|---|---|
+| Original Developer & Model Author | Theivanth Rajamohan |
+| Technical Reviewer, Code Contributor & Publication Support | [Swapin Vidya](https://huggingface.co/Swapinvidya) |
+| Project Contributor | [Rajah Shanmugasundaram](https://huggingface.co/rajahshan) |
+| Publishing Organisation | TechSphere Consultancy Services Private Limited |
+
+**Project provenance:** FactoryVision AI was originally developed by Theivanth Rajamohan. [Swapin Vidya](https://huggingface.co/Swapinvidya) contributed technical review, detection-metrics improvements, testing, model publication preparation and documentation. The models are being prepared for publication under TechSphere Consultancy Services Private Limited's Hugging Face namespace, with the original developer's permission.
+
+- 🤗 [TechSphereAI on Hugging Face](https://huggingface.co/TechsphereAI)
+- 🌐 [Official website](https://techsphereconsultancy.com/)
+- 💻 [Original GitHub project](https://github.com/Theivanth/Factory-Vision-AI)
+- 🔀 [Swapin Vidya's merged contribution](https://github.com/Theivanth/Factory-Vision-AI/pull/1)