Model Card for grahamaco/beans-classifier

Model Details

Model Description

Bean leaf disease classifier trained with self-improving Classifier Lab pipeline.

• Developed by: Graham Anderson (Embry OS)
• Funded by [optional]: [More Information Needed]
• Shared by [optional]: [More Information Needed]
• Model type: [More Information Needed]
• Language(s) (NLP): en
• License: apache-2.0
• Finetuned from model [optional]: [More Information Needed]

Model Sources [optional]

• Repository: https://github.com/grahamaco/embry-os
• Paper [optional]: [More Information Needed]
• Demo [optional]: [More Information Needed]

Uses

Direct Use

[More Information Needed]

Downstream Use [optional]

[More Information Needed]

Out-of-Scope Use

[More Information Needed]

Bias, Risks, and Limitations

[More Information Needed]

Recommendations

Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.

How to Get Started with the Model

Use the code below to get started with the model.

[More Information Needed]

Training Details

Training Data

[More Information Needed]

Training Procedure

Preprocessing [optional]

[More Information Needed]

Training Hyperparameters

• Training regime: [More Information Needed]

Speeds, Sizes, Times [optional]

[More Information Needed]

Evaluation

Testing Data, Factors & Metrics

Testing Data

[More Information Needed]

Factors

[More Information Needed]

Metrics

[More Information Needed]

Results

[More Information Needed]

Summary

Model Examination [optional]

[More Information Needed]

Environmental Impact

Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).

• Hardware Type: [More Information Needed]
• Hours used: [More Information Needed]
• Cloud Provider: [More Information Needed]
• Compute Region: [More Information Needed]
• Carbon Emitted: [More Information Needed]

Technical Specifications [optional]

Model Architecture and Objective

[More Information Needed]

Compute Infrastructure

[More Information Needed]

Hardware

[More Information Needed]

Software

[More Information Needed]

Citation [optional]

BibTeX:

[More Information Needed]

APA:

[More Information Needed]

Glossary [optional]

[More Information Needed]

More Information [optional]

[More Information Needed]

Model Card Authors [optional]

[More Information Needed]

Model Card Contact

[More Information Needed]

Results (Verified on Held-Out Test Set)

128 test images, never seen during training or validation.

Metric	Value
Macro F1	0.9218
Accuracy	0.9219
Holdout Gate	PASSED >= 0.90

Per-Class Metrics

Class	Precision	Recall	F1	Support
angular_leaf_spot	0.92	0.84	0.88	43.0
bean_rust	0.87	0.95	0.91	43.0
healthy	0.98	0.98	0.98	42.0

Confusion Matrix

	angular_leaf_spot	bean_rust	healthy
angular_leaf_spot	36	6	1
bean_rust	2	41	0
healthy	1	0	41

Self-Improvement Loop

The classifier was trained iteratively until the holdout gate (F1 >= 0.90) was met. Each round adjusts hyperparameters and augmentation strategy based on prior failures.

Round	Epochs	LR	Augment	Val F1	Test F1	Gate
1	10	0.0002	1	0.9328	0.8762	FAILED
2	15	0.0001	2	0.9240	0.9218	PASSED

Winning configuration: Round 2

Architecture

• Backbone: efficientnet_b0 (pretrained on ImageNet)
• Framework: PyTorch + timm
• Classes: angular_leaf_spot, bean_rust, healthy
• Image size: 224x224

Training Pipeline (Classifier Lab)

1. Research -- Identified EfficientNet B0 as candidate
2. Data -- Beans dataset from HuggingFace (1034 train, 133 val, 128 test)
3. Self-improvement loop -- Round 1 failed gate (F1 0.876), adjusted LR and augmentation, Round 2 passed (F1 0.922)
4. Evaluate -- Held-out test set (128 images, never seen during training)
5. Promote -- Gate passed, pushed to HuggingFace with model checkpoint

License

Apache 2.0