santhoshj/kv-ai
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

docs: update README and forge documentation

Browse Source 
- README: Fixed backend command, added GPU compatibility reference
- forge.md: Fixed train CLI (--output-dir), added GPU troubleshooting
- Added reference to GPU compatibility guide for RTX 50-series
This commit is contained in:
Santhosh Janardhanan
2026-04-13 17:23:58 -04:00
parent 05401e48dc
commit 1d0ea4f2cf
2 changed files with 19 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
README.md
View File

@@ -39,6 +39,7 @@ A fully local, privacy-first AI companion trained on your Obsidian vault. Combin
- Node.js 18+ (for UI) - Node.js 18+ (for UI)
- Ollama running locally - Ollama running locally
- RTX 5070 or equivalent (12GB+ VRAM for fine-tuning) - RTX 5070 or equivalent (12GB+ VRAM for fine-tuning)
- See [GPU Compatibility Guide](docs/gpu-compatibility.md) for RTX 50-series setup
### Installation ### Installation
@@ -76,7 +77,7 @@ See [docs/config.md](docs/config.md) for full configuration reference.
**Terminal 1 - Backend:** **Terminal 1 - Backend:**
```bash ```bash
python -m uvicorn companion.api:app --host 0.0.0.0 --port 7373 python -m companion.api
``` ```
**Terminal 2 - Frontend:** **Terminal 2 - Frontend:**
@@ -139,8 +140,9 @@ python -m companion.forge.cli reload ~/.companion/training/final
| `companion.config` | Configuration management | [docs/config.md](docs/config.md) | | `companion.config` | Configuration management | [docs/config.md](docs/config.md) |
| `companion.rag` | RAG engine (chunk, embed, search) | [docs/rag.md](docs/rag.md) | | `companion.rag` | RAG engine (chunk, embed, search) | [docs/rag.md](docs/rag.md) |
| `companion.forge` | Fine-tuning pipeline | [docs/forge.md](docs/forge.md) | | `companion.forge` | Fine-tuning pipeline | [docs/forge.md](docs/forge.md) |
| `companion.api` | FastAPI backend | [docs/api.md](docs/api.md) | | `companion.api` | FastAPI backend | This README |
| `ui/` | React frontend | [docs/ui.md](docs/ui.md) | | `ui/` | React frontend | [docs/ui.md](docs/ui.md) |
| **GPU Setup** | RTX 50-series compatibility | [docs/gpu-compatibility.md](docs/gpu-compatibility.md) |
## Project Structure ## Project Structure

23
docs/forge.md
View File

@@ -193,22 +193,24 @@ new_path = reload_model(
```bash ```bash
# Extract training data # Extract training data
companion.forge.cli extract [--output PATH] python -m companion.forge.cli extract [--output PATH]
# Train model # Train model
companion.forge.cli train \ python -m companion.forge.train \
[--data PATH] \ --data PATH \
[--output PATH] \ --output-dir PATH \
[--epochs N] \ --epochs N \
[--lr FLOAT] --lr FLOAT
# Check model status # Check model status
companion.forge.cli status python -m companion.forge.cli status
# Reload model # Reload model
companion.forge.cli reload MODEL_PATH [--no-backup] python -m companion.forge.cli reload MODEL_PATH [--no-backup]
``` ```
**Note:** Use `--output-dir` (or `--output`) to specify the training output directory.
## Training Tips ## Training Tips
**Dataset Size:** **Dataset Size:**
@@ -243,6 +245,11 @@ Use `gradient_accumulation_steps` to increase effective batch size.
## Troubleshooting ## Troubleshooting
**GPU Not Detected / CUDA Not Available**
- See [GPU Compatibility Guide](gpu-compatibility.md)
- Common issue on RTX 50-series: Install CUDA-enabled PyTorch: `pip install torch==2.5.1+cu121 torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121`
- Verify: `python -c "import torch; print(torch.cuda.is_available())"`
**CUDA Out of Memory** **CUDA Out of Memory**
- Reduce `lora_rank` to 8 - Reduce `lora_rank` to 8
- Reduce `batch_size` to 2 - Reduce `batch_size` to 2