AvoCahDoe/invoice-layout-yolo11m

Invoice Layout Detection — YOLO11m

Fine-tuned YOLO11m for 8-class invoice document layout region detection (metadata, vendor/customer blocks, table, line items, summary, payment, column headers).

Part of the Hub collection Invoice Layout Extraction.

Property	Value
Architecture	YOLO11m (20.1M params, 68.0 GFLOPs)
Base weights	ultralytics/yolo11m
Input size	1024 px
Classes	8 layout regions
Training epochs	100 (best @ epoch 91)
Optimizer	AdamW
Dataset	AvoCahDoe/invoice-annotated-bbox
Demo Space	AvoCahDoe/invoice-layout-yolov8n-demo
Train / val / test	372 / 14 / 7 pages

Classes

ID	Name
0	invoice_metadata
1	vendor_block
2	customer_block
3	table_block
4	line_item
5	summary_block
6	payment_block
7	Column

Metrics

Test split (held-out, 7 images)

Metric	Value
mAP50	0.8418
mAP50-95	0.4926
Precision	0.9033
Recall	0.8466

Validation (best epoch 91)

Metric	Best	Final (epoch 100)
mAP50	0.8493	0.8780
mAP50-95	0.4951	0.4917
Precision	0.8280	0.8682
Recall	0.7991	0.7980

Training time: ~180.4 min on RTX 4070 Laptop GPU.

Training configuration

Parameter	Value
Batch	6
Image size	1024
Patience	25
LR (cosine)	0.01 → 0.01
Mosaic	1.0
Mixup	0.15
Copy-paste	0.1
Horizontal flip	0.0

Training results

Learning curves

Precision–Recall

Confusion matrices

Raw	Normalized

Validation predictions

Ground truth	Model predictions

Training batches (augmented)

Batch 0	Batch 1	Batch 2

Label distribution

Usage

Ultralytics (recommended)

from ultralytics import YOLO
from huggingface_hub import hf_hub_download

weights = hf_hub_download("AvoCahDoe/invoice-layout-yolo11m", "weights/best.pt")
model = YOLO(weights)
results = model.predict("invoice_page.png", imgsz=1024, conf=0.25)
results[0].show()

Load from Hub by repo id

from ultralytics import YOLO

model = YOLO("hf://AvoCahDoe/invoice-layout-yolo11m/weights/best.pt")
results = model.predict("invoice_page.png")

Training reproduction

python scripts/train_comparison.py --models yolo11m

Repository layout

weights/best.pt          # Best checkpoint (use this)
weights/last.pt          # Last epoch checkpoint
config/                  # Training configuration
metrics/                 # Per-epoch and summary metrics
assets/                  # Plots and visualizations

Model comparison (test split)

All six architectures fine-tuned on the same invoice layout dataset.

Rank	Model	test mAP50	test mAP50-95	Precision	Recall	Hub
1	yolov8n	0.9600	0.7167	0.9502	0.9592	YOLOv8n
2	yolov8s	0.9006	0.6739	0.9419	0.8677	YOLOv8s
3	yolov8x	0.8757	0.6224	0.9144	0.8735	YOLOv8x
4	yolo11x	0.8738	0.6127	0.9484	0.8530	YOLO11x
5	yolo11m ← this model	0.8418	0.4926	0.9033	0.8466	YOLO11m
6	yolov8m	0.8388	0.5289	0.9077	0.8580	YOLOv8m

License

Apache 2.0