Merge pull request #266 from SKaiNET-developers/fxi/0.5-io-reader

Fxi/0.5 io reader

Author: michalharakal

CHANGELOG.md

@@ -1,5 +1,22 @@
 # Changelog
 
+## [0.5.1] - 2025-12-26
+
+### Added
+- Common I/O abstraction with `ModelReader` and `TensorInfo` in `skainet-io-core` for unified model loading.
+- Efficient memory handling with non-copying `slice` views in `MemoryChunk`.
+- Unified `skainet-tensor-tools` CLI combining ONNX and GGUF utilities.
+- `OnnxStatsCli` tool for analyzing ONNX model parameters and structure.
+
+### Changed
+- Migrated project to `SKaiNET-developers` organization; updated repository URLs and deployment configurations.
+- Standardized artifact naming in documentation (e.g., `SKaiNET-lang-core`).
+- Improved `GGUFReader` with better alignment parsing and tensor data handling.
+- Optimized test infrastructure: increased heap size to 8GB for large model tests and added `ReadmeSnippetsTest` for documentation verification.
+
+### Removed
+- Legacy standalone applications and tools: `skainet-KGPChat`, `skainet-mnist`, and separate ONNX/GGUF tool modules.
+
 ## [0.5.0] - 2025-12-06
 
 ### Added

LICENCE

@@ -1,6 +1,7 @@
 MIT License
 
-Copyright (c) 2023-2025 sk-ai-net contributors
+Copyright (c) 2023-2025 Michal Harakal
+Copyright (c) 2025 SKaiNET-developers contributors
 
 Permission is hereby granted, free of charge, to any person obtaining a copy
 of this software and associated documentation files (the "Software"), to deal

README.md

@@ -2,17 +2,103 @@
 [![Maven Central](https://img.shields.io/maven-central/v/sk.ainet.core/skainet-lang-core.svg)](https://central.sonatype.com/artifact/sk.ainet.core/skainet-lang-core)
 [![DeepWiki](https://img.shields.io/badge/DeepWiki-View%20Docs-blue?logo=readthedocs&logoColor=white)](https://deepwiki.com/sk-ai-net/SKaiNET)
 
-# SKaiNET
+<img src="docs//SKaiNET-logo.png" alt="SKaiNET logo" width="150">
 
 **SKaiNET** is an open-source deep learning framework written in Kotlin Multiplatform, designed with developers in mind to enable the creation modern AI powered applications with ease.
 
+## Key features at a glance
+
+### SKaiNET is Data
+
+- Data loaders (MNIST, JSON datasets, image helpers)
+- Type‑safe tensors across JVM/JS/Native
+
+```kotlin
+// data loaders
+val ds = MNIST.load(train = true)
+val (x, y) = ds.nextBatch(64)
+
+// Type-safe tensor creation via tensor DSL
+val ctx = DefaultNeuralNetworkExecutionContext()
+val mask = data<FP32, Float>(ctx) {
+    tensor{
+        shape(3, 3) {
+            from(
+                1f, 0f, 0f,
+                1f, 1f, 0f,
+                1f, 1f, 1f,
+            )
+        }
+    }
+}
+
+val t = tensor<FP32, Float>(ctx, FP32::class) {
+    tensor {
+        shape(2, 3) {
+            from(
+                0f, 1f, 2f,
+                10f, 11f, 12f
+            )
+        }
+    }
+}
+println("shape=${t.shape} first=${t.data[0,0]}")
+```
+
+### SKaiNET is Language
+
+- Kotlin DSLs for Data, Neural Nets, and Pipelines
+
+```kotlin
+// Neural network DSL
+val model = nn {
+    input(28 * 28)
+    dense(out = 128)
+    relu()
+    dense(out = 10)
+}
+```
+
+### SKaiNET is Tools
+
+- Kotlin Notebook support Explorer and Notebook-friendly APIs
+
+```kotlin
+// Works smoothly in Kotlin Notebooks
+display(model.summary())
+println(ds.describe())
+```
+
+### SKaiNET is Compiler
+
+- MLIR/StableHLO based lowering (modules provided in `SKaiNET-compile-*`)
+
+```kotlin
+// Illustrative: export graph to JSON/StableHLO IR
+val ir = Compile.toStableHlo(model)
+println(ir.pretty())
+```
+
+### SKaiNET is for Developers
+
+- Clean APIs, growing docs, Maven Central artifacts
+- Get productive in minutes with minimal deps
+
+```kotlin
+dependencies {
+    implementation("sk.ainet.core:SKaiNET-lang-core:0.5.1")
+    implementation("sk.ainet.core:SKaiNET-backend-cpu:0.5.1")
+}
+// Ready to build & run in ~8 minutes
+```
+
 ## Use it
 
 - From Kotlin code in apps, libraries, CLIs
 - In Kotlin Notebooks for quick exploration
 - With sample projects to learn patterns
 
-See also CHANGELOG for what’s new in 0.5.0.
+See also CHANGELOG for what’s new in 0.5.1.
 
 ## Quick start
 
@@ -27,15 +113,15 @@ dependencyResolutionManagement {
 
 dependencies {
     // minimal dependency with simple CPU backend
-    implementation("sk.ainet.core:skainet-lang-core:0.5.0")
-    implementation("sk.ainet.core:skainet-backend-cpu:0.5.0")
-    
+    implementation("sk.ainet.core:SKaiNET-lang-core:0.5.1")
+    implementation("sk.ainet.core:SKaiNET-backend-cpu:0.5.1")
+
     // simple model zoo
-    implementation("sk.ainet.core:skainet-lang-models:0.5.0")
-    
+    implementation("sk.ainet.core:SKaiNET-lang-models:0.5.1")
+
     // Optional I/O (e.g., GGUF loader, JSON)
-    implementation("sk.ainet.core:skainet-io-core:0.5.0")
-    implementation("sk.ainet.core:skainet-io-gguf:0.5.0")
+    implementation("sk.ainet.core:SKaiNET-io-core:0.5.1")
+    implementation("sk.ainet.core:SKaiNET-io-gguf:0.5.1")
 }
 ```
 
@@ -44,15 +130,15 @@ Maven:
 ```xml
 <dependency>
   <groupId>sk.ainet.core</groupId>
-  <artifactId>skainet-lang-core</artifactId>
-  <version>0.5.0</version>
+  <artifactId>SKaiNET-lang-core</artifactId>
+  <version>0.5.1</version>
 </dependency>
 ```
 
 ## Samples and notebooks
 
-- Sample app: https://github.com/sk-ai-net/skainet-samples/tree/feature/MNIST/SinusApproximator
-- Kotlin Notebook: https://github.com/sk-ai-net/skainet-notebook
+- Sample app: https://github.com/sk-ai-net/SKaiNET-samples/tree/feature/MNIST/SinusApproximator
+- Kotlin Notebook: https://github.com/sk-ai-net/SKaiNET-notebook
 
 ## Architecture
 
@@ -142,8 +228,8 @@ Notes and limitations:
 - You can customize initializers for the mixing weights, basis, and bias via the DSL block.
 
 See source for details:
-- skainet-lang/skainet-kan/src/commonMain/kotlin/sk/ainet/lang/kan/KanDsl.kt
-- skainet-lang/skainet-kan/src/commonMain/kotlin/sk/ainet/lang/kan/KanLayer.kt
+- SKaiNET-lang/SKaiNET-kan/src/commonMain/kotlin/sk/ainet/lang/kan/KanDsl.kt
+- SKaiNET-lang/SKaiNET-kan/src/commonMain/kotlin/sk/ainet/lang/kan/KanLayer.kt
 
 ## License
 

build.gradle.kts

@@ -29,6 +29,10 @@ subprojects {
             jvmToolchain(21)
         }
     }
+
+    tasks.withType<Test>().configureEach {
+        maxHeapSize = "8192m"
+    }
 }
 
 kover {

docs/SKaiNET-logo.png

@@ -1,20 +1,20 @@
 GROUP=sk.ainet.core
-VERSION_NAME=0.5.0
+VERSION_NAME=0.5.1
 
 POM_DESCRIPTION=SKaiNET
 
-POM_URL=https://github.com/sk-ai-net/skainet/
-POM_SCM_URL=https://github.com/sk-ai-net/skainet
-POM_SCM_CONNECTION=scm:git:git@github.com:sk-ai-net/skainet.git
-POM_SCM_DEV_CONNECTION=scm:git:ssh@github.com:sk-ai-net/skainet.git
+POM_URL=https://github.com/SKaiNET-developers/skainet/
+POM_SCM_URL=https://github.com/SKaiNET-developers/skainet
+POM_SCM_CONNECTION=scm:git:git@github.com:SKaiNET-developers/skainet.git
+POM_SCM_DEV_CONNECTION=scm:git:ssh@github.com:SKaiNET-developers/skainet.git
 
 POM_LICENCE_NAME=MIT License
 POM_LICENCE_URL=https://mit-license.org/
 POM_LICENCE_DIST=repo
 
-POM_DEVELOPER_ID=sk-ai-net
+POM_DEVELOPER_ID=SKaiNET-developers
 POM_DEVELOPER_NAME=SKaiNET development team
-POM_DEVELOPER_URL=https://github.com/sk-ai-net/
+POM_DEVELOPER_URL=https://github.com/SKaiNET-developers/
 
 mavenCentralPublishing=true
 mavenCentralAutomaticPublishing=true

gradle.properties

@@ -57,6 +57,5 @@ include("skainet-models:skainet-model-yolo")
 
 
 // ====== APPS
-//include("skainet-apps:skainet-KotlinGPT")
-include("skainet-apps:skainet-onnx-tools")
-include("skainet-apps:skainet-onnx-detect")
+//include("skainet-apps:skainet-KGPChat")
+include("skainet-apps:skainet-tensor-tools")