# 🌱 LittleLearn – Touch the Big World with Little Steps
LittleLearn is an experimental and original machine learning framework built from scratch — inspired by the simplicity of Keras and the flexibility of PyTorch, yet designed with its own architecture, philosophy, and gradient engine.
## 🧠 What Makes LittleLearn Different?
- 🔧 Not a wrapper – LittleLearn is not built on top of TensorFlow, PyTorch, or other major ML libraries.
- 💡 Fully original layers, modules, and autodiff engine (GradientReflector).
- 🧩 Customizable down to the node level: build models from high-level APIs or go low-level for complete control.
- 🛠️ Features unique like:
- Node-level gradient clipping
- Inline graph tracing
- Custom attention mechanisms (e.g., Multi-Head Attention from scratch)
- 🤯 Designed for both research experimentation and deep learning education.
## ⚙️ Core Philosophy
Touch the Big World with Little Steps.
Whether you want rapid prototyping or total model control — LittleLearn gives you both.
LittleLearn provides multiple levels of abstraction:
| Usage Style | Tools Available |
|--------------------------|-------------------------------------------|
| 💬 One-liner models | `AutoBuildModel`, `AutoTransformers` (soon) |
| ⚙️ Modular models | `Sequential`, `ModelByNode` (soon) |
| 🔬 Low-level experiment | Layers, Loss, Optimizer manual calls |
| 🧠 Custom gradients | `GradientReflector` engine backend |
## 📦 Ecosystem Features
- ✅ Deep learning modules: Dense, LSTM, attention mechanisms, and more
- 🧮 Classical ML components (in progress)
- 🤖 Automated tools like AutoBuildModel
- 🔄 Custom training loops with full backend access
- 🧠 All powered by the GradientReflector engine — providing automatic differentiation with transparency and tweakability
## 🔧 Installation
```bash
pip install littlelearn
```
🚀 Quick Example :
```bash
import LittleLearn as ll
x_train = 'your datasets'
y_train = 'your target'
model = ll.DeepLearning.Model.AutoBuildModel(type='mlp-binaryclassification',level='balance')
model.fit(x_train,y_train.reshape(-1,1),epochs=10,verbose=1)
```
📌 Disclaimer
While inspired by well-known frameworks, LittleLearn is built entirely from scratch with its own mechanics.
It is suitable for:
- 🔬 Experimental research
- 🏗️ Framework building
- 📚 Educational purposes
- 🔧 Custom low-level operations
This is an alpha-stage project — expect bugs, sharp edges, and lots of potential.
👤 Author
Candra Alpin Gunawan
📧 hinamatsuriairin@gmail.com
🌐 GitHub https://github.com/Airinchan818/LittleLearn
Raw data
{
"_id": null,
"home_page": "https://github.com/Airinchan818/LittleLearn",
"name": "littlelearn",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.7",
"maintainer_email": null,
"keywords": "deep learning, artificial intellengence, engine grad, ai, framework, api, machine learning",
"author": "Candra Alpin Gunawan",
"author_email": "hinamatsuriairin@gmail.com",
"download_url": "https://files.pythonhosted.org/packages/44/f6/81f85313a055754e53a22dba3b549a83bf5a7f0b73d5ca9af6ddd857cf66/littlelearn-0.1.3.tar.gz",
"platform": null,
"description": "# \ud83c\udf31 LittleLearn \u2013 Touch the Big World with Little Steps\r\n\r\nLittleLearn is an experimental and original machine learning framework built from scratch \u2014 inspired by the simplicity of Keras and the flexibility of PyTorch, yet designed with its own architecture, philosophy, and gradient engine.\r\n\r\n## \ud83e\udde0 What Makes LittleLearn Different?\r\n- \ud83d\udd27 Not a wrapper \u2013 LittleLearn is not built on top of TensorFlow, PyTorch, or other major ML libraries.\r\n\r\n- \ud83d\udca1 Fully original layers, modules, and autodiff engine (GradientReflector).\r\n\r\n- \ud83e\udde9 Customizable down to the node level: build models from high-level APIs or go low-level for complete control.\r\n\r\n- \ud83d\udee0\ufe0f Features unique like:\r\n\r\n- Node-level gradient clipping\r\n\r\n- Inline graph tracing\r\n\r\n- Custom attention mechanisms (e.g., Multi-Head Attention from scratch)\r\n\r\n- \ud83e\udd2f Designed for both research experimentation and deep learning education.\r\n\r\n## \u2699\ufe0f Core Philosophy\r\nTouch the Big World with Little Steps.\r\nWhether you want rapid prototyping or total model control \u2014 LittleLearn gives you both.\r\n\r\nLittleLearn provides multiple levels of abstraction:\r\n\r\n| Usage Style | Tools Available |\r\n|--------------------------|-------------------------------------------|\r\n| \ud83d\udcac One-liner models | `AutoBuildModel`, `AutoTransformers` (soon) |\r\n| \u2699\ufe0f Modular models | `Sequential`, `ModelByNode` (soon) |\r\n| \ud83d\udd2c Low-level experiment | Layers, Loss, Optimizer manual calls |\r\n| \ud83e\udde0 Custom gradients | `GradientReflector` engine backend |\r\n\r\n\r\n## \ud83d\udce6 Ecosystem Features\r\n- \u2705 Deep learning modules: Dense, LSTM, attention mechanisms, and more\r\n\r\n- \ud83e\uddee Classical ML components (in progress)\r\n\r\n- \ud83e\udd16 Automated tools like AutoBuildModel\r\n\r\n- \ud83d\udd04 Custom training loops with full backend access\r\n\r\n- \ud83e\udde0 All powered by the GradientReflector engine \u2014 providing automatic differentiation with transparency and tweakability\r\n\r\n## \ud83d\udd27 Installation\r\n\r\n```bash\r\n pip install littlelearn\r\n```\r\n\r\n\ud83d\ude80 Quick Example : \r\n```bash\r\nimport LittleLearn as ll \r\n\r\nx_train = 'your datasets'\r\ny_train = 'your target'\r\n\r\nmodel = ll.DeepLearning.Model.AutoBuildModel(type='mlp-binaryclassification',level='balance')\r\nmodel.fit(x_train,y_train.reshape(-1,1),epochs=10,verbose=1)\r\n```\r\n\ud83d\udccc Disclaimer\r\nWhile inspired by well-known frameworks, LittleLearn is built entirely from scratch with its own mechanics.\r\nIt is suitable for:\r\n\r\n- \ud83d\udd2c Experimental research\r\n\r\n- \ud83c\udfd7\ufe0f Framework building\r\n\r\n- \ud83d\udcda Educational purposes\r\n\r\n- \ud83d\udd27 Custom low-level operations\r\n\r\nThis is an alpha-stage project \u2014 expect bugs, sharp edges, and lots of potential.\r\n\r\n\ud83d\udc64 Author\r\nCandra Alpin Gunawan\r\n\ud83d\udce7 hinamatsuriairin@gmail.com\r\n\ud83c\udf10 GitHub https://github.com/Airinchan818/LittleLearn\r\n",
"bugtrack_url": null,
"license": null,
"summary": "machine learning ecosystem",
"version": "0.1.3",
"project_urls": {
"Homepage": "https://github.com/Airinchan818/LittleLearn"
},
"split_keywords": [
"deep learning",
" artificial intellengence",
" engine grad",
" ai",
" framework",
" api",
" machine learning"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "ce411bcfba685a382d2fc13ecc1ee8a91ba6296a3d77018380797aa68b6eb725",
"md5": "86d362ecd48f38f227def2ae39f785c9",
"sha256": "b375904444916152572f5d223f5c1c8b84d28c9feb5f967ce360e82e8598407c"
},
"downloads": -1,
"filename": "littlelearn-0.1.3-py3-none-any.whl",
"has_sig": false,
"md5_digest": "86d362ecd48f38f227def2ae39f785c9",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.7",
"size": 62661,
"upload_time": "2025-07-21T15:53:18",
"upload_time_iso_8601": "2025-07-21T15:53:18.327425Z",
"url": "https://files.pythonhosted.org/packages/ce/41/1bcfba685a382d2fc13ecc1ee8a91ba6296a3d77018380797aa68b6eb725/littlelearn-0.1.3-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "44f681f85313a055754e53a22dba3b549a83bf5a7f0b73d5ca9af6ddd857cf66",
"md5": "3e4335b5b8b0c4198248b7d39422457e",
"sha256": "7dbe954c8b1e41f3c5ce9042db432bdba9041ac13e911628f77c9d161d496034"
},
"downloads": -1,
"filename": "littlelearn-0.1.3.tar.gz",
"has_sig": false,
"md5_digest": "3e4335b5b8b0c4198248b7d39422457e",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.7",
"size": 57798,
"upload_time": "2025-07-21T15:53:20",
"upload_time_iso_8601": "2025-07-21T15:53:20.428569Z",
"url": "https://files.pythonhosted.org/packages/44/f6/81f85313a055754e53a22dba3b549a83bf5a7f0b73d5ca9af6ddd857cf66/littlelearn-0.1.3.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-07-21 15:53:20",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "Airinchan818",
"github_project": "LittleLearn",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"requirements": [
{
"name": "numpy",
"specs": [
[
">=",
"1.26.4"
]
]
},
{
"name": "numba",
"specs": [
[
">=",
"0.61.2"
]
]
},
{
"name": "networkx",
"specs": [
[
">=",
"3.2.1"
]
]
},
{
"name": "matplotlib",
"specs": [
[
">=",
"3.8.3"
]
]
}
],
"lcname": "littlelearn"
}
JSON
