1
Fork 0

refactor: set wandb entity and simplify READMEs

This commit is contained in:
Tibo De Peuter 2026-04-01 10:26:39 +02:00
parent 3e992a3377
commit 7e7c5bf27c
7 changed files with 21 additions and 218 deletions

View file

@ -2,91 +2,22 @@
This directory contains configuration files for training experiments.
## Quick Start
## Usage
### 1. Choose a Template
Use `--config` with `src/train.py` to run an experiment:
**For Development/Testing:**
```bash
cp configs/dev_test.yaml configs/my_dev.yaml
python src/train.py --config configs/default_ppo.yaml
```
**For Production Training:**
You can overriding settings via CLI:
```bash
cp configs/production_training.yaml configs/my_experiment.yaml
python src/train.py --config configs/default_ppo.yaml --learning-rate 0.001
```
### 2. Configure Your Settings
## Available Configurations
Edit your config file and **set your wandb entity**:
```yaml
# ⚠️ IMPORTANT: Set this to your WandB username or team name
wandb_entity: "your-wandb-username"
track: true # Enable WandB logging
```
### 3. Run Training
**Using config file:**
```bash
python src/train.py --config configs/my_experiment.yaml
```
**Override specific parameters:**
```bash
python src/train.py --config configs/my_experiment.yaml --learning-rate 0.001 --num-envs 32
```
**Pure CLI (no config file):**
```bash
python src/train.py --track --wandb-entity your-username --total-timesteps 1000000
```
## Features
### 📊 WandB Integration
- Real-time metrics logging
- Model checkpoints as artifacts
- Run comparison and collaboration
### 🔧 Flexible Configuration
- YAML files for reproducible experiments
- CLI overrides for quick adjustments
- Team collaboration without code changes
## Configuration Templates
### `dev_test.yaml`
- Fast iteration for development
- Short runs (100K timesteps)
- Frequent checkpoints
- Small environment count
### `production_training.yaml`
- Full-scale training (50M timesteps)
- Optimized hyperparameters
- Production-ready settings
### `default_ppo.yaml`
- Baseline configuration template
- Balanced settings for most use cases
## Team Collaboration
Each team member should create their own config file:
```yaml
# configs/alice_experiment.yaml
exp_name: "alice_locomotion_v2"
track: true
wandb_project_name: "PPO-Modularity"
wandb_entity: "alice-research" # Alice's WandB username
total_timesteps: 20000000
# ... other settings
```
This allows everyone to:
- Use their own WandB account
- Run different experiments simultaneously
- Share configurations via version control
- Avoid conflicts in run names
- `default_ppo.yaml`: Baseline config.
- `dev_test.yaml`: Fast iteration for development.
- `production_training.yaml`: Full-scale training.
- `personal_template.yaml`: Template for team members to customize.

View file

@ -15,7 +15,7 @@ seed: 1
# Tracking settings
track: false # Set to true to enable WandB logging
wandb_project_name: "PPO-Modularity"
wandb_entity: null # Set to your WandB username or team name
wandb_entity: "SEL3-2026-Groep-4" # Set to your WandB username or team name
# Model saving
save_model: true

View file

@ -9,7 +9,7 @@ seed: 123
# Tracking settings - IMPORTANT: Set your own wandb_entity!
track: true
wandb_project_name: "PPO-Modularity-Dev"
wandb_entity: null # ⚠️ SET THIS TO YOUR WANDB USERNAME OR TEAM
wandb_entity: "SEL3-2026-Groep-4" # ⚠️ SET THIS TO YOUR WANDB USERNAME OR TEAM
# Model saving
save_model: true

View file

@ -9,7 +9,7 @@ seed: 42
# WandB settings - ⚠️ IMPORTANT: Set your credentials!
track: true # Enable WandB tracking
wandb_project_name: "PPO-Modularity"
wandb_entity: "YOUR_WANDB_USERNAME" # ⚠️ CHANGE THIS to your WandB username/team
wandb_entity: "SEL3-2026-Groep-4" # ⚠️ CHANGE THIS to your WandB username/team
# Quick experiment settings (modify as needed)
total_timesteps: 500000 # 500K for quick results

View file

@ -10,7 +10,7 @@ seed: 42
# Tracking settings - IMPORTANT: Set your own wandb_entity!
track: true
wandb_project_name: "PPO-Modularity"
wandb_entity: null # ⚠️ SET THIS TO YOUR WANDB USERNAME OR TEAM
wandb_entity: "SEL3-2026-Groep-4" # ⚠️ SET THIS TO YOUR WANDB USERNAME OR TEAM
# Model saving
save_model: true

View file

@ -26,7 +26,7 @@ class PPOArgs:
wandb_project_name: str = "PPO-Modularity"
# the entity (team) of wandb's project
wandb_entity: str | None = None
wandb_entity: str | None = "SEL3-2026-Groep-4"
# whether to capture videos of the agent performances (check out `videos` folder)
capture_video: bool = False

View file

@ -1,145 +1,17 @@
# Experiment Logger
A lightweight, standalone logging framework for machine learning experiments with multi-backend support.
A lightweight logging framework supporting Weights & Biases, local JSON, and stdout.
## Features
- **Multi-backend logging**: Simultaneously log to WandB, local disk (JSON), and stdout
- **Data preservation**: All metrics saved locally, even if WandB is unavailable
- **Checkpoint management**: Save model checkpoints with metadata
- **WandB integration**: Optional artifact upload for model versioning
- **Graceful degradation**: Works without WandB installed
- **Simple API**: Minimal configuration required
## Installation
This package is included in the project. To use it in your code:
```python
from experiment_logger import UnifiedLogger
```
## Quick Start
## Usage
```python
from experiment_logger import UnifiedLogger
# Initialize logger
logger = UnifiedLogger(
run_name="my_experiment",
config={"learning_rate": 0.001, "batch_size": 32},
project_name="MyProject",
entity="my-wandb-username", # Optional
use_wandb=True, # Set to False to disable WandB
)
logger = UnifiedLogger(run_name="my_experiment", config={"lr": 0.001})
# Log metrics
for step in range(100):
logger.log({
"loss": 1.0 / (step + 1),
"accuracy": step * 0.01,
}, step=step)
# Save checkpoint
logger.save_checkpoint(
params=model_params,
step=100,
metadata={"epoch": 1, "val_acc": 0.95},
)
# Save final model
logger.save_final_model(
params=final_params,
metadata={"final_accuracy": 0.98},
)
# Finalize (flushes remaining metrics)
logger.log({"loss": 0.5}, step=1)
logger.save_checkpoint(params=model_params, step=1)
logger.finish()
```
## Context Manager
Use as a context manager for automatic cleanup:
```python
with UnifiedLogger(run_name="my_exp", config={}) as logger:
logger.log({"metric": 1.0})
# Automatically calls finish() on exit
```
## Configuration
### Constructor Parameters
- `run_name` (str): Unique name for this run
- `config` (dict): Configuration dictionary with hyperparameters
- `project_name` (str): WandB project name (default: "PPO-Modularity")
- `entity` (str, optional): WandB entity (team/user name)
- `base_dir` (str): Base directory for local storage (default: "runs")
- `use_wandb` (bool): Enable WandB logging (default: True)
- `save_code` (bool): Save code to WandB (default: True)
### Directory Structure
```
runs/
└── my_experiment/
├── config.json # Saved configuration
├── metrics/
│ └── metrics.jsonl # Line-delimited JSON metrics
├── checkpoints/
│ ├── checkpoint_step_100.flax
│ └── checkpoint_step_100_metadata.json
└── final_model.flax
```
## API Reference
### `log(metrics, step=None, commit=True)`
Log metrics to all backends.
**Parameters:**
- `metrics` (dict): Dictionary of metric name -> value
- `step` (int, optional): Global step counter (auto-incremented if None)
- `commit` (bool): Whether to commit to WandB immediately
### `save_checkpoint(params, step, prefix="checkpoint", metadata=None)`
Save model checkpoint to disk and optionally to WandB.
**Parameters:**
- `params`: Model parameters (Flax params or any serializable object)
- `step` (int): Current training step
- `prefix` (str): Prefix for checkpoint filename
- `metadata` (dict, optional): Additional metadata to save
### `save_final_model(params, metadata=None)`
Save the final trained model.
**Parameters:**
- `params`: Model parameters
- `metadata` (dict, optional): Metadata about the final model
### `finish()`
Finalize logging and cleanup. Flushes remaining metrics to disk.
## Usage in Projects
This logger is designed to be:
- **Project-agnostic**: Use in any ML project, not just this one
- **Framework-agnostic**: Works with JAX, PyTorch, TensorFlow, etc.
- **Minimal dependencies**: Only requires `wandb` (optional), `flax` (for serialization), and `numpy`
## Design Philosophy
1. **Never lose data**: All metrics saved locally, regardless of WandB availability
2. **Simple API**: Minimal boilerplate, easy to integrate
3. **Fail gracefully**: Missing WandB shouldn't break experiments
4. **Reproducibility**: Save full configuration with every run
## License
Part of the 2026SEL3-project-BrittleStar repository.
Logs and checkoints are saved in the `runs/` directory. If `track=True` (or `use_wandb=True`), everything is additionally synced to Weights & Biases.