Image Classification
ultralytics
LiteRT
Bengali
English
yolo26-cls
agriculture
bangladesh
crop-disease
yolo26
on-device
rice
chashibhai
Instructions to use Shaq2/chashibhai-rice-disease-cls with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ultralytics
How to use Shaq2/chashibhai-rice-disease-cls with ultralytics:
# Couldn't find a valid YOLO version tag. # Replace XX with the correct version. from ultralytics import YOLOvXX model = YOLOvXX.from_pretrained("Shaq2/chashibhai-rice-disease-cls") source = 'http://images.cocodataset.org/val2017/000000039769.jpg' model.predict(source=source, save=True) - Notebooks
- Google Colab
- Kaggle
Publish rice YOLO26-cls TFLite + labels + source weights with model card
Browse files- CITATION.cff +13 -0
- LICENSE +21 -0
- NOTICE.md +8 -0
- README.md +142 -0
- best.pt +3 -0
- config.json +35 -0
- labels.json +28 -0
- model.tflite +3 -0
CITATION.cff
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
cff-version: 1.2.0
|
| 2 |
+
title: "ChashiBhAI Rice Disease Classifier"
|
| 3 |
+
message: If you use this model, please cite it.
|
| 4 |
+
type: software
|
| 5 |
+
authors:
|
| 6 |
+
- family-names: Ahmed
|
| 7 |
+
given-names: Shakil
|
| 8 |
+
alias: Shaq2
|
| 9 |
+
repository-code: https://github.com/MRSHAKILS/AI-Powered-Smart-Agriculture-Advisory-Platform-for-Bangladesh
|
| 10 |
+
url: https://huggingface.co/Shaq2/chashibhai-rice-disease-cls
|
| 11 |
+
license: MIT
|
| 12 |
+
version: "0.1.0"
|
| 13 |
+
date-released: "2026-08-16"
|
LICENSE
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
MIT License
|
| 2 |
+
|
| 3 |
+
Copyright (c) 2026 Shakil Ahmed (ChashiBhAI)
|
| 4 |
+
|
| 5 |
+
Permission is hereby granted, free of charge, to any person obtaining a copy
|
| 6 |
+
of this software and associated documentation files (the "Software"), to deal
|
| 7 |
+
in the Software without restriction, including without limitation the rights
|
| 8 |
+
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
| 9 |
+
copies of the Software, and to permit persons to whom the Software is
|
| 10 |
+
furnished to do so, subject to the following conditions:
|
| 11 |
+
|
| 12 |
+
The above copyright notice and this permission notice shall be included in all
|
| 13 |
+
copies or substantial portions of the Software.
|
| 14 |
+
|
| 15 |
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
| 16 |
+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
| 17 |
+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
| 18 |
+
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
| 19 |
+
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
| 20 |
+
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
|
| 21 |
+
SOFTWARE.
|
NOTICE.md
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Attribution
|
| 2 |
+
|
| 3 |
+
On-device crop disease classifier for ChashiBhAI.
|
| 4 |
+
Author: Shakil Ahmed (Hugging Face: Shaq2).
|
| 5 |
+
|
| 6 |
+
Not claimed here:
|
| 7 |
+
- KrishokChat dataset / LLM / RAG — RaiyanKhaan/KrishokChat-Advisory-System, arXiv:2606.29243
|
| 8 |
+
- Ultralytics YOLO
|
README.md
ADDED
|
@@ -0,0 +1,142 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
library_name: ultralytics
|
| 4 |
+
pipeline_tag: image-classification
|
| 5 |
+
tags:
|
| 6 |
+
- agriculture
|
| 7 |
+
- bangladesh
|
| 8 |
+
- crop-disease
|
| 9 |
+
- tflite
|
| 10 |
+
- yolo26
|
| 11 |
+
- on-device
|
| 12 |
+
- rice
|
| 13 |
+
- chashibhai
|
| 14 |
+
language:
|
| 15 |
+
- bn
|
| 16 |
+
- en
|
| 17 |
+
base_model_relation: quantized
|
| 18 |
+
---
|
| 19 |
+
|
| 20 |
+
# ChashiBhAI Rice Disease Classifier
|
| 21 |
+
|
| 22 |
+
| | |
|
| 23 |
+
|---|---|
|
| 24 |
+
| **Author** | [Shaq2](https://huggingface.co/Shaq2) (Shakil Ahmed) |
|
| 25 |
+
| **Crop** | rice (ধান) |
|
| 26 |
+
| **Task** | Image classification (leaf disease) |
|
| 27 |
+
| **Architecture** | YOLO26-cls → TFLite FP16 |
|
| 28 |
+
| **Input** | `[1, 640, 640, 3]` NHWC RGB `/255.0` |
|
| 29 |
+
| **Output** | `[1, 8]` softmax probabilities (`nms: false`) |
|
| 30 |
+
| **Status** | `production-demo` |
|
| 31 |
+
| **App** | ChashiBhAI (Expo / React Native, on-device diagnosis) |
|
| 32 |
+
| **Code** | [GitHub](https://github.com/MRSHAKILS/AI-Powered-Smart-Agriculture-Advisory-Platform-for-Bangladesh) |
|
| 33 |
+
| **Collection** | [ChashiBhAI on-device classifiers](https://huggingface.co/collections/Shaq2/chashibhai-on-device-disease-classifiers) |
|
| 34 |
+
|
| 35 |
+
Disease ID in ChashiBhAI **always** runs on-device. Gemini / KrishokChat generate advisory **text only** and never see the photo.
|
| 36 |
+
|
| 37 |
+
## Files
|
| 38 |
+
|
| 39 |
+
| File | Role |
|
| 40 |
+
|---|---|
|
| 41 |
+
| `model.tflite` | On-device graph used by the Android app (~3.12 MB) |
|
| 42 |
+
| `labels.json` | Canonical class names **and** preprocess contract |
|
| 43 |
+
| `best.pt` | Ultralytics source weights (export parent) |
|
| 44 |
+
|
| 45 |
+
## Classes (8)
|
| 46 |
+
|
| 47 |
+
| Label | English | Bangla |
|
| 48 |
+
|---|---|---|
|
| 49 |
+
| `Rice__Bacterial_Leaf_Blight` | Bacterial Leaf Blight | ব্যাকটেরিয়াজনিত পাতা পোড়া |
|
| 50 |
+
| `Rice__Brown_Spot` | Brown Spot | বাদামী দাগ |
|
| 51 |
+
| `Rice__Healthy_Leaf` | Healthy Leaf | সুস্থ পাতা |
|
| 52 |
+
| `Rice__Leaf_Blast` | Leaf Blast | ব্লাস্ট |
|
| 53 |
+
| `Rice__Leaf_Scald` | Leaf Scald | পাতা পোড়া |
|
| 54 |
+
| `Rice__Narrow_Brown_Leaf_Spot` | Narrow Brown Leaf Spot | সরু বাদামী দাগ |
|
| 55 |
+
| `Rice__Rice_Hispa` | Rice Hispa | হিসপা |
|
| 56 |
+
| `Rice__Sheath_Blight` | Sheath Blight | শেথ ব্লাইট |
|
| 57 |
+
|
| 58 |
+
## Preprocessing (variant C) — required
|
| 59 |
+
|
| 60 |
+
Do **not** letterbox. Letterbox disagreed with the `.pt` on non-square photos.
|
| 61 |
+
|
| 62 |
+
1. Resize the **shortest side** to `imgsz` = **640**, keep aspect ratio.
|
| 63 |
+
2. **Centre-crop** to `640×640`.
|
| 64 |
+
3. RGB, NHWC, `float32 / 255.0`.
|
| 65 |
+
|
| 66 |
+
`labels.json` is the source of truth (`preprocess: centercrop`).
|
| 67 |
+
|
| 68 |
+
## Measured export checks
|
| 69 |
+
|
| 70 |
+
| Check | Result |
|
| 71 |
+
|---|---|
|
| 72 |
+
| Preprocess verified vs `.pt` | yes (variant C, 100% top-1 vs `.pt` on the rice hold-out used for export) |
|
| 73 |
+
| FP16 vs FP32 top-1 agreement | 1.0 |
|
| 74 |
+
| FP16 vs FP32 max softmax diff | 0.000372171 |
|
| 75 |
+
| Bundled in APK | yes (rice) |
|
| 76 |
+
|
| 77 |
+
## Intended use
|
| 78 |
+
|
| 79 |
+
- On-device diagnosis in ChashiBhAI for Bangladeshi farmers (Bangla-first UI).
|
| 80 |
+
- Research reproduction of the mobile export.
|
| 81 |
+
|
| 82 |
+
**Out of scope:** detection / bounding boxes, crop auto-routing, chemical dosage (handled by a separate advisory stack with a refuse gate).
|
| 83 |
+
|
| 84 |
+
## Limitations
|
| 85 |
+
|
| 86 |
+
Demo-grade on-device classifier. Not a plant-pathologist substitute. Retake if confidence is low or the leaf is not centred.
|
| 87 |
+
|
| 88 |
+
## Load (Python)
|
| 89 |
+
|
| 90 |
+
```python
|
| 91 |
+
import json
|
| 92 |
+
from pathlib import Path
|
| 93 |
+
import numpy as np
|
| 94 |
+
from PIL import Image
|
| 95 |
+
import tensorflow as tf
|
| 96 |
+
|
| 97 |
+
def preprocess_centercrop(path: str, imgsz: int = 640) -> np.ndarray:
|
| 98 |
+
im = Image.open(path).convert("RGB")
|
| 99 |
+
w, h = im.size
|
| 100 |
+
scale = imgsz / min(w, h)
|
| 101 |
+
nw, nh = int(round(w * scale)), int(round(h * scale))
|
| 102 |
+
im = im.resize((nw, nh), Image.BILINEAR)
|
| 103 |
+
left, top = (nw - imgsz) // 2, (nh - imgsz) // 2
|
| 104 |
+
im = im.crop((left, top, left + imgsz, top + imgsz))
|
| 105 |
+
return (np.asarray(im, dtype=np.float32) / 255.0)[None, ...]
|
| 106 |
+
|
| 107 |
+
labels = json.loads(Path("labels.json").read_text(encoding="utf-8"))
|
| 108 |
+
it = tf.lite.Interpreter(model_path="model.tflite")
|
| 109 |
+
it.allocate_tensors()
|
| 110 |
+
inp, out = it.get_input_details()[0], it.get_output_details()[0]
|
| 111 |
+
it.set_tensor(inp["index"], preprocess_centercrop("leaf.jpg", labels["imgsz"]))
|
| 112 |
+
it.invoke()
|
| 113 |
+
p = it.get_tensor(out["index"])[0]
|
| 114 |
+
i = int(p.argmax())
|
| 115 |
+
print(labels["names"][i], float(p[i]))
|
| 116 |
+
```
|
| 117 |
+
|
| 118 |
+
## Related models (same author)
|
| 119 |
+
|
| 120 |
+
- [Rice](https://huggingface.co/Shaq2/chashibhai-rice-disease-cls)
|
| 121 |
+
- [Brassica](https://huggingface.co/Shaq2/chashibhai-brassica-disease-cls)
|
| 122 |
+
- [Corn](https://huggingface.co/Shaq2/chashibhai-corn-disease-cls)
|
| 123 |
+
- Suite index: [chashibhai-disease-classifiers](https://huggingface.co/Shaq2/chashibhai-disease-classifiers)
|
| 124 |
+
|
| 125 |
+
## Credit (not this model)
|
| 126 |
+
|
| 127 |
+
KrishokChat Bengali advisory LLM / RAG is **not** this classifier. See [RaiyanKhaan/KrishokChat-Advisory-System](https://huggingface.co/RaiyanKhaan/KrishokChat-Advisory-System) and arXiv:2606.29243 (Reza & Shahid).
|
| 128 |
+
|
| 129 |
+
## Citation
|
| 130 |
+
|
| 131 |
+
```bibtex
|
| 132 |
+
@software{ahmed2026chashibhai_rice_cls,
|
| 133 |
+
author = {Ahmed, Shakil},
|
| 134 |
+
title = {ChashiBhAI Rice Disease Classifier},
|
| 135 |
+
year = {2026},
|
| 136 |
+
url = {https://huggingface.co/Shaq2/chashibhai-rice-disease-cls}
|
| 137 |
+
}
|
| 138 |
+
```
|
| 139 |
+
|
| 140 |
+
## License
|
| 141 |
+
|
| 142 |
+
MIT (see `LICENSE`). Ultralytics remains under its own license. This pack redistributes **weights**, not training images.
|
best.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f25243959a3d7dd0851f4d2e53e30890515cadd5e318871ffe04c0413865d61a
|
| 3 |
+
size 3205243
|
config.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"YOLO26Cls"
|
| 4 |
+
],
|
| 5 |
+
"model_type": "yolo26-cls",
|
| 6 |
+
"library_name": "ultralytics",
|
| 7 |
+
"crop": "rice",
|
| 8 |
+
"task": "image-classification",
|
| 9 |
+
"imgsz": 640,
|
| 10 |
+
"num_classes": 8,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "Rice__Bacterial_Leaf_Blight",
|
| 13 |
+
"1": "Rice__Brown_Spot",
|
| 14 |
+
"2": "Rice__Healthy_Leaf",
|
| 15 |
+
"3": "Rice__Leaf_Blast",
|
| 16 |
+
"4": "Rice__Leaf_Scald",
|
| 17 |
+
"5": "Rice__Narrow_Brown_Leaf_Spot",
|
| 18 |
+
"6": "Rice__Rice_Hispa",
|
| 19 |
+
"7": "Rice__Sheath_Blight"
|
| 20 |
+
},
|
| 21 |
+
"label2id": {
|
| 22 |
+
"Rice__Bacterial_Leaf_Blight": 0,
|
| 23 |
+
"Rice__Brown_Spot": 1,
|
| 24 |
+
"Rice__Healthy_Leaf": 2,
|
| 25 |
+
"Rice__Leaf_Blast": 3,
|
| 26 |
+
"Rice__Leaf_Scald": 4,
|
| 27 |
+
"Rice__Narrow_Brown_Leaf_Spot": 5,
|
| 28 |
+
"Rice__Rice_Hispa": 6,
|
| 29 |
+
"Rice__Sheath_Blight": 7
|
| 30 |
+
},
|
| 31 |
+
"preprocess": "centercrop",
|
| 32 |
+
"softmaxed": true,
|
| 33 |
+
"nms": false,
|
| 34 |
+
"status": "production-demo"
|
| 35 |
+
}
|
labels.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"crop": "rice-disease",
|
| 3 |
+
"task": "classify",
|
| 4 |
+
"imgsz": 640,
|
| 5 |
+
"inputShape": [
|
| 6 |
+
1,
|
| 7 |
+
640,
|
| 8 |
+
640,
|
| 9 |
+
3
|
| 10 |
+
],
|
| 11 |
+
"outputShape": [
|
| 12 |
+
1,
|
| 13 |
+
8
|
| 14 |
+
],
|
| 15 |
+
"softmaxed": true,
|
| 16 |
+
"preprocess": "centercrop",
|
| 17 |
+
"preprocessVerified": true,
|
| 18 |
+
"names": [
|
| 19 |
+
"Rice__Bacterial_Leaf_Blight",
|
| 20 |
+
"Rice__Brown_Spot",
|
| 21 |
+
"Rice__Healthy_Leaf",
|
| 22 |
+
"Rice__Leaf_Blast",
|
| 23 |
+
"Rice__Leaf_Scald",
|
| 24 |
+
"Rice__Narrow_Brown_Leaf_Spot",
|
| 25 |
+
"Rice__Rice_Hispa",
|
| 26 |
+
"Rice__Sheath_Blight"
|
| 27 |
+
]
|
| 28 |
+
}
|
model.tflite
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f9b5e2de9e23c49ba3ccc3a54645d24a4aab991897a7f18aa599c212957ad478
|
| 3 |
+
size 3121879
|