pictograph/pg-coco-resnet50

View on Pictograph · Pictograph Research · Classification · resnet50 · FP32 · Custom License

Overview

This is a classification model built on resnet50, trained on the COCO-CLS-5k dataset. It expects 448x448 input and runs in FP32 precision. The ONNX weights are in this repo; run them with the Pictograph SDK (below) or open the model on Pictograph to test it in-browser, call the hosted API, or deploy it as an always-on endpoint.

Performance

Headline: 72.6% Accuracy.

Metric	Value
Accuracy	72.6%
F1	58.7%
Precision	60.7%
Recall	59.5%

Classes

Class index matches the model's output order.

#	Class
0	skateboarding
1	skiing
2	bathroom
3	bird
4	frisbee
5	giraffe
6	soccer
7	teddy bear
8	elephant
9	vintage photo
10	bear
11	zebra
12	tennis
13	train
14	baseball
15	cat
16	dog
17	airplane
18	kitchen
19	kitesurfing
20	kite
21	boat
22	beach
23	surfing
24	horse
25	vase
26	livestock
27	bedroom
28	fruit
29	living room
30	market
31	motorcycle
32	portrait
33	clock
34	street
35	fire hydrant
36	bus
37	luggage
38	bench
39	pizza
40	office
41	electronics
42	child
43	video games
44	dining
45	formal wear
46	donut
47	dessert
48	meal
49	vegetables
50	cake
51	hot dog
52	street sign
53	stop sign
54	other

Training

Setting	Value
Base dataset	COCO-CLS-5k
Dataset size	5,000 images
Input size	448x448
Epochs	40
Batch size	32
Precision	FP32
Version	1.0.0

Use this model

The weights ship as ONNX (pg-coco-resnet50.onnx) with a config.json. The Pictograph SDK applies the model's exact pre/post-processing for you:

Pictograph SDK

pip install "pictograph[inference]"

from huggingface_hub import hf_hub_download
from pictograph import load_model, ClassificationModel

onnx = hf_hub_download("pictograph/pg-coco-resnet50", "pg-coco-resnet50.onnx")
config = hf_hub_download("pictograph/pg-coco-resnet50", "config.json")

model: ClassificationModel = load_model(onnx, config, task="classification")
result = model.predict("photo.jpg")   # ClassificationResult
print("top:", result.top.name, round(result.top.confidence, 3))
for c in result.classes:               # ranked, highest first
    print(f"  {c.name} {c.confidence:.3f}")

Example output (illustrative; your exact values depend on the image):

top: skateboarding 0.972
  skateboarding 0.972
  skiing 0.021

Raw ONNX Runtime (no SDK)

from huggingface_hub import hf_hub_download
import onnxruntime as ort

onnx = hf_hub_download("pictograph/pg-coco-resnet50", "pg-coco-resnet50.onnx")
sess = ort.InferenceSession(onnx, providers=["CPUExecutionProvider"])
inp = sess.get_inputs()[0]
print(inp.name, inp.shape)   # NCHW float32; input size + normalization in config.json
# outputs = sess.run(None, {inp.name: preprocessed_batch})
# Decoding is architecture-specific; the Pictograph SDK (above) does it for you.

• Test it in-browser, call the hosted API, or deploy an endpoint: open the model on Pictograph.
• SDK + local inference guide: docs.

License

Non-commercial use only. Trained on COCO-CLS-5k, whose images are COCO's: the COCO Consortium does not own their copyright; each image comes from Flickr under its own license, recorded per image in COCO's annotation files, and most of them (68.9% of train2017) are Creative Commons NonCommercial. COCO's annotations are CC BY 4.0, the COCO Consortium. Cite Lin et al., Microsoft COCO: Common Objects in Context (ECCV 2014).

---

Trained and published with Pictograph - annotate, train, and deploy from one API.