MNIST RNN Classifier

This repository contains a validation-selected MNIST RNN digit classifier trained with dlab.

Architecture

Results

3-seed confirmation and test audit for the selected RNN recipe:

metric	value
validation accuracy	97.9167% ± 0.0491 pp
validation loss	0.07222 ± 0.00309
test accuracy	97.9733% ± 0.1370 pp
test loss	0.06857 ± 0.00544

Representative checkpoint selected by validation accuracy:

metric	value
seed	`9001`
selected validation accuracy	97.9667%
selected validation loss	0.06796
test accuracy	97.7800%
test loss	0.07622

The ONNX model was exported from the validation-selected checkpoint. Test metrics were produced after the recipe was selected and were logged in W&B test-audit run vdt5duxq.

Model Details

Dataset: MNIST
Architecture: RNN sequence classifier
Sequence axis: rows
Pooling: mean
Hidden width: 512
Recurrent layers: 2
Bidirectional: false
Dropout: 0
Optimizer: AdamW
Learning rate: 0.003
Weight decay: 0.0001
Scheduler: constant
Label smoothing: 0
Batch size: 256
Training augmentation: false
Checkpoint selection: max validation accuracy
Source W&B run: t8u8llqo

Input / Output

Use model.onnx for code-independent inference.

Input name: images
Input shape: [batch, 1, 28, 28]
Input dtype: float32
Output name: logits
Output shape: [batch, 10]

Preprocessing:

Convert image to grayscale.
Resize to 28 x 28.
Scale pixel values to [0, 1].
Normalize with mean 0.1307 and standard deviation 0.3081.
Arrange the tensor as channels-first [batch, 1, 28, 28].

Usage

Install the runtime dependencies:

pip install huggingface_hub onnxruntime pillow numpy

Run inference with the ONNX model:

import numpy as np
import onnxruntime as ort
from huggingface_hub import hf_hub_download
from PIL import Image

LABELS = {
    0: "0",
    1: "1",
    2: "2",
    3: "3",
    4: "4",
    5: "5",
    6: "6",
    7: "7",
    8: "8",
    9: "9",
}

model_path = hf_hub_download(
    repo_id="tsilva/mnist-classifier-rnn",
    filename="model.onnx",
)

image = Image.open("example.png").convert("L").resize((28, 28))
x = np.asarray(image, dtype=np.float32) / 255.0
x = (x - 0.1307) / 0.3081
x = x[None, None, :, :].astype(np.float32)

session = ort.InferenceSession(model_path, providers=["CPUExecutionProvider"])
logits = session.run(["logits"], {"images": x})[0]
prediction = int(logits.argmax(axis=1)[0])

print(prediction, LABELS[prediction])

Labels

MNIST labels:

id	label
0	0
1	1
2	2
3	3
4	4
5	5
6	6
7	7
8	8
9	9

Files

model.onnx: ONNX export of the validation-selected checkpoint. Prefer this file for portable inference.
model.ckpt: PyTorch Lightning checkpoint for the same model. This is code-dependent and mainly useful for PyTorch-based inspection or continued experimentation.
config.yaml: resolved Hydra training config.
metrics.csv: training metrics from the uploaded checkpoint run.
metadata.json: compact metadata for inference and provenance.

Limitations

This RNN model treats each MNIST image as a short sequence rather than using convolutional inductive bias. It is intended for normalized 28 x 28 grayscale MNIST-style images; remaining errors are expected to concentrate in ambiguous handwritten digits and distribution shifts outside that input format.

Downloads last month: 6

tsilva
/

mnist-classifier-rnn