aprender 0.30.0

Next-generation ML framework in pure Rust — `cargo install aprender` for the `apr` CLI
Documentation 
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

<!-- PCU: getting-started-first-training | contract: contracts/apr-page-getting-started-first-training-v1.yaml -->
<!-- Example: cargo run -p aprender-core --example none -->
<!-- Status: enforced -->

# First Training

Fine-tune a model with LoRA:

```bash
# Fine-tune with LoRA adapter
apr finetune model.gguf \
  --adapter lora \
  --rank 64 \
  --data train.jsonl \
  --output adapter.safetensors

# Apply the adapter
apr merge model.gguf adapter.safetensors -o fine-tuned.gguf
```

## Training Pipeline

For full training workflows:

```bash
# Plan (dry-run, shows what will happen)
apr train plan config.yaml

# Execute
apr train apply config.yaml

# Monitor in real-time
apr monitor
```

## Knowledge Distillation

```bash
apr distill \
  --teacher large-model.gguf \
  --student small-model.gguf \
  --data train.jsonl \
  -o distilled.gguf
```