Skip to content
Other

HMH-1-Cell

Cell classifier and detector

2 min read

A two-stage pipeline based on onnxruntime for microscopic image analysis, combining object detection and classification across 20 categories. The system detects cells using a YOLO11s-seg model and then classifies them into cell lines, blood cells, immune cells, and malaria samples. The solution runs on standard CPU hardware without requiring heavy PyTorch installations. This project will be useful for researchers, bioinformaticians, and developers of cell culture analysis applications.

Key features

  • Two-stage architecture: YOLO11s-seg for localization and ResNet18 for precise cell type identification.
  • Support for 20 distinct classes, including cell lines, blood elements, and malaria samples.
  • Fully autonomous CPU operation powered by the onnxruntime library.
  • Open data structure with expansion and customization capabilities via configuration files.

What you can do

  • Integrate the pipeline into local scripts for automated microscopic image processing.
  • Run a complete processing cycle on full-size images with detection, cropping, and classification.
  • Use pre-trained weights and classes to build specialized diagnostic prototypes.

Example

import json, numpy as np, onnxruntime as ort
from PIL import Image

cls = ort.InferenceSession("model.onnx", providers=["CPUExecutionProvider"])
meta = json.load(open("classes.json", encoding="utf-8"))
classes = meta["classes"]

im = Image.open("sample_cell.jpg").convert("RGB").resize((128, 128))
x = np.asarray(im, dtype=np.float32) / 255.0
x = x.transpose(2, 0, 1)[None]
logits = cls.run(None, {cls.get_inputs()[0].name: x})[0][0]
print(classes[int(np.argmax(logits))]["name_cn"])
Read moreDetails
• Classifier accuracy (top-1): 94.04% on a test set of 29,668 images.
• Detector (LIVECell): recall 0.813, precision 0.781, F1 0.797.
• Trained on LIVECell, BloodMNIST, Deepcell, and NIH Malaria datasets.
• License: CC BY-NC 4.0 (non-commercial use only).
• Limitation: optimized for phase-contrast and brightfield microscopy.
Discuss Telegram 1

Will it run?

Runs in BF16

Fits without quantisation.

  • BF16weights 0.0 GB · ready files0.5 GB
  • FP8 / INT8recommendedweights 0.0 GB0.5 GB
  • 4-bitweights 0.0 GB0.5 GB

Tip: go 8-bit.FP8 / INT8 is within about a percent of full precision in quality, takes half the memory and runs faster on RTX 40/50 and H100. Full precision only pays off for fine-tuning.

What to get for this model

Massed ComputeOpen Massed Compute

Your own GPU: FP8 / INT8 — from RTX 2060 · BF16 — from RTX 2060

Massed Compute hourly prices, November 2026. The discounted price applies with the coupon at checkout.

Estimate: weights plus working memory. Long videos, high resolutions and long contexts need more.

Run it on Massed Compute

The script downloads the model to the rented machine. How to run it is in the model card.

  1. 1. Rent a machine with a fitting GPU

    Open Massed Compute
  2. 2. Paste the script into the machine's terminal

    Connect over SSH (the Massed Compute panel shows the command) or open a terminal on the machine's desktop.

    HF_TOKEN / CIVITAI_TOKENThe script starts with token lines: paste yours between the quotes. The Hugging Face token is for gated models, the Civitai one for Civitai models.

    # fengraoyaoshi/HMH-1-Cell-20class
    mkdir -p ~/models
    curl -LsSf https://astral.sh/uv/install.sh | sh
    source "$HOME/.local/bin/env"
    uv tool install "huggingface_hub[cli]"
    hf download fengraoyaoshi/HMH-1-Cell-20class --local-dir ~/models/HMH-1-Cell-20class
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to HMH-1-Cell-20class
  3. 3. Use it

    The model is in ~/models. How to run it is in the model card (link above).

Stop the machine in the panel when you are done: it bills by the minute while it runs.

How to speed it up

Picked for this model type. Choose your hardware above to see what fits you exactly.

  • FlashAttention 44.0.0b40from November 18, 2026new release

    fastest exact attention on B200

    Kernels for Blackwell (beta). Exact attention without quantisation — fine for everything, LLMs included.

    How to enable: pip install flash-attn-4; vLLM and SGLang pick it up themselves.

    Earlier generations (2)
    • FlashAttention 3 — The Hopper (H100/H200) version with FP8.
    • FlashAttention 2 — The standard for RTX 30/40/50 and A100: exact attention without the memory cost of long contexts.
  • torch.compile2.15.0from October 28, 2026new release

    10–30% for free

    Compiles the model for your GPU. Slower first run, faster afterwards.

    How to enable: model = torch.compile(model); in ComfyUI the TorchCompileModel node.

Model family

New posts about versions of HMH-1-Cell-20class (quants, finetunes, LoRAs) will show up under the bell.

No known relatives of this model yet. All derivatives on Hugging Face

About the model

fengraoyaoshi/HMH-1-Cell-20class
Author
fengraoyaoshi
Category
Other
Released
Posted

On NeyroLearn

Hugging Face

01

Ask AI about this model

Answered by Gemini 4 · the chat is kept only in this tab

Comments