File size: 2,876 Bytes
f203641 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 |
# OktoEngine Examples
Complete working examples demonstrating OktoEngine capabilities.
---
## Table of Contents
1. [Basic Training](#basic-training)
2. [LoRA Fine-tuning](#lora-fine-tuning)
3. [Chatbot Training](#chatbot-training)
4. [Multi-format Export](#multi-format-export)
---
## Basic Training
**Location:** [`basic-training/`](./basic-training/)
Minimal working example for training a simple model.
**Files:**
- `scripts/train.okt` - Training configuration
- `dataset/train.jsonl` - Sample training data
- `dataset/val.jsonl` - Sample validation data
**Usage:**
```bash
cd basic-training
okto validate
okto train
```
---
## LoRA Fine-tuning
**Location:** [`lora-training/`](./lora-training/)
Example of efficient LoRA fine-tuning for large models.
**Files:**
- `scripts/train.okt` - LoRA configuration
- `dataset/train.jsonl` - Training data
**Usage:**
```bash
cd lora-training
okto validate
okto train
```
---
## Chatbot Training
**Location:** [`chatbot/`](./chatbot/)
Complete example for training a conversational AI model.
**Files:**
- `scripts/train.okt` - Chatbot configuration
- `dataset/train.jsonl` - Conversation data
- `dataset/val.jsonl` - Validation conversations
**Usage:**
```bash
cd chatbot
okto validate
okto train
okto eval
```
---
## Multi-format Export
**Location:** [`multi-export/`](./multi-export/)
Example showing how to export models to multiple formats.
**Files:**
- `scripts/train.okt` - Configuration with multiple export formats
**Usage:**
```bash
cd multi-export
okto train
okto export --format okm,onnx,gguf
```
---
## Running Examples
1. **Navigate to example directory:**
```bash
cd examples/basic-training
```
2. **Validate configuration:**
```bash
okto validate
```
3. **Train the model:**
```bash
okto train
```
4. **Check results:**
```bash
ls runs/
ls export/
```
---
## Customizing Examples
All examples can be customized:
1. **Edit `scripts/train.okt`** - Modify training parameters
2. **Replace `dataset/*.jsonl`** - Use your own data
3. **Adjust `MODEL.base`** - Use different base models
4. **Modify `EXPORT.format`** - Change export formats
---
## Example Output
**Training output:**
```
🐙 OktoEngine v0.1
📄 Reading: "scripts/train.okt"
📊 Environment Check:
✔ Runtime: Python 3.14.0
✔ GPU: NVIDIA GeForce RTX 4070
✔ RAM: 63GB (40GB available)
🚀 Starting training pipeline...
Epoch 1/5: 100%|████████████| 500/500 [02:15<00:00, 3.70it/s]
Loss: 2.345 → 1.892
✅ Training completed successfully!
📁 Output: runs/MyModel/
```
---
**Need help?** Check the [Getting Started Guide](../docs/GETTING_STARTED.md) or [FAQ](../docs/FAQ.md).
|