diff --git a/examples/2_evaluate_pretrained_policy.py b/examples/2_evaluate_pretrained_policy.py
index 46409041..bb73167b 100644
--- a/examples/2_evaluate_pretrained_policy.py
+++ b/examples/2_evaluate_pretrained_policy.py
@@ -1 +1,39 @@
-# TODO
+"""
+This scripts demonstrates how to evaluate a pretrained policy from the HuggingFace Hub or from your local
+training outputs directory. In the latter case, you might want to run examples/3_train_policy.py first.
+"""
+
+from pathlib import Path
+
+from huggingface_hub import snapshot_download
+
+from lerobot.common.utils import init_hydra_config
+from lerobot.scripts.eval import eval
+
+# Get a pretrained policy from the hub.
+hub_id = "lerobot/diffusion_policy_pusht_image"
+folder = Path(snapshot_download(hub_id))
+# OR uncomment the following to evaluate a policy from the local outputs/train folder.
+folder = Path("outputs/train/example_pusht_diffusion")
+
+config_path = folder / "config.yaml"
+weights_path = folder / "model.pt"
+stats_path = folder / "stats.pth"  # normalization stats
+
+# Override some config parameters to do with evaluation.
+overrides = [
+    f"policy.pretrained_model_path={weights_path}",
+    "eval_episodes=10",
+    "rollout_batch_size=10",
+    "device=cuda",
+]
+
+# Create a Hydra config.
+cfg = init_hydra_config(config_path, overrides)
+
+# Evaluate the policy and save the outputs including metrics and videos.
+eval(
+    cfg,
+    out_dir=f"outputs/eval/example_{cfg.env.name}_{cfg.policy.name}",
+    stats_path=stats_path,
+)
diff --git a/examples/3_train_policy.py b/examples/3_train_policy.py
index 46409041..01a4cf76 100644
--- a/examples/3_train_policy.py
+++ b/examples/3_train_policy.py
@@ -1 +1,55 @@
-# TODO
+"""This scripts demonstrates how to train Diffusion Policy on the PushT environment.
+
+Once you have trained a model with this script, you can try to evaluate it on
+examples/2_evaluate_pretrained_policy.py
+"""
+
+import os
+from pathlib import Path
+
+import torch
+from omegaconf import OmegaConf
+from tqdm import trange
+
+from lerobot.common.datasets.factory import make_offline_buffer
+from lerobot.common.policies.diffusion.policy import DiffusionPolicy
+from lerobot.common.utils import init_hydra_config
+
+output_directory = Path("outputs/train/example_pusht_diffusion")
+os.makedirs(output_directory, exist_ok=True)
+
+overrides = [
+    "env=pusht",
+    "policy=diffusion",
+    # Adjust as you prefer. 5000 steps are needed to get something worth evaluating.
+    "offline_steps=5000",
+    "log_freq=250",
+    "device=cuda",
+]
+
+cfg = init_hydra_config("lerobot/configs/default.yaml", overrides)
+
+policy = DiffusionPolicy(
+    cfg=cfg.policy,
+    cfg_device=cfg.device,
+    cfg_noise_scheduler=cfg.noise_scheduler,
+    cfg_rgb_model=cfg.rgb_model,
+    cfg_obs_encoder=cfg.obs_encoder,
+    cfg_optimizer=cfg.optimizer,
+    cfg_ema=cfg.ema,
+    n_action_steps=cfg.n_action_steps + cfg.n_latency_steps,
+    **cfg.policy,
+)
+policy.train()
+
+offline_buffer = make_offline_buffer(cfg)
+
+for offline_step in trange(cfg.offline_steps):
+    train_info = policy.update(offline_buffer, offline_step)
+    if offline_step % cfg.log_freq == 0:
+        print(train_info)
+
+# Save the policy, configuration, and normalization stats for later use.
+policy.save(output_directory / "model.pt")
+OmegaConf.save(cfg, output_directory / "config.yaml")
+torch.save(offline_buffer.transform[-1].stats, output_directory / "stats.pth")
diff --git a/lerobot/common/datasets/abstract.py b/lerobot/common/datasets/abstract.py
index a81de49b..c05d25c0 100644
--- a/lerobot/common/datasets/abstract.py
+++ b/lerobot/common/datasets/abstract.py
@@ -59,6 +59,8 @@ class AbstractDataset(TensorDictReplayBuffer):
         collate_fn: Callable | None = None,
         writer: Writer | None = None,
         transform: "torchrl.envs.Transform" = None,
+        # Don't actually load any data. This is a stand-in solution to get the transforms.
+        dummy: bool = False,
     ):
         assert (
             self.available_datasets is not None
@@ -77,7 +79,7 @@ class AbstractDataset(TensorDictReplayBuffer):
                 f"The version of the dataset ({self.version}) is not enforced when root is provided ({self.root})."
             )
 
-        storage = self._download_or_load_dataset()
+        storage = self._download_or_load_dataset() if not dummy else None
 
         super().__init__(
             storage=storage,
diff --git a/lerobot/common/datasets/aloha.py b/lerobot/common/datasets/aloha.py
index 031c2cd3..83d1581a 100644
--- a/lerobot/common/datasets/aloha.py
+++ b/lerobot/common/datasets/aloha.py
@@ -97,6 +97,7 @@ class AlohaDataset(AbstractDataset):
         collate_fn: Callable | None = None,
         writer: Writer | None = None,
         transform: "torchrl.envs.Transform" = None,
+        dummy: bool = False,
     ):
         super().__init__(
             dataset_id,
@@ -110,6 +111,7 @@ class AlohaDataset(AbstractDataset):
             collate_fn=collate_fn,
             writer=writer,
             transform=transform,
+            dummy=dummy,
         )
 
     @property
diff --git a/lerobot/common/datasets/factory.py b/lerobot/common/datasets/factory.py
index 4212e023..276dc761 100644
--- a/lerobot/common/datasets/factory.py
+++ b/lerobot/common/datasets/factory.py
@@ -21,7 +21,12 @@ def make_offline_buffer(
     overwrite_batch_size=None,
     overwrite_prefetch=None,
     stats_path=None,
+    # Don't actually load any data. This is a stand-in solution to get the transforms.
+    dummy=False,
 ):
+    if dummy and normalize and stats_path is None:
+        raise ValueError("`stats_path` is required if `dummy` and `normalize` are True.")
+
     if cfg.policy.balanced_sampling:
         assert cfg.online_steps > 0
         batch_size = None
@@ -93,6 +98,7 @@ def make_offline_buffer(
         root=DATA_DIR,
         pin_memory=pin_memory,
         prefetch=prefetch if isinstance(prefetch, int) else None,
+        dummy=dummy,
     )
 
     if cfg.policy.name == "tdmpc":
diff --git a/lerobot/common/datasets/pusht.py b/lerobot/common/datasets/pusht.py
index 624fb140..d167f3ea 100644
--- a/lerobot/common/datasets/pusht.py
+++ b/lerobot/common/datasets/pusht.py
@@ -100,6 +100,7 @@ class PushtDataset(AbstractDataset):
         collate_fn: Callable | None = None,
         writer: Writer | None = None,
         transform: "torchrl.envs.Transform" = None,
+        dummy: bool = False,
     ):
         super().__init__(
             dataset_id,
@@ -113,6 +114,7 @@ class PushtDataset(AbstractDataset):
             collate_fn=collate_fn,
             writer=writer,
             transform=transform,
+            dummy=dummy,
         )
 
     def _download_and_preproc_obsolete(self):
diff --git a/lerobot/common/datasets/simxarm.py b/lerobot/common/datasets/simxarm.py
index dc30e69e..06931d3f 100644
--- a/lerobot/common/datasets/simxarm.py
+++ b/lerobot/common/datasets/simxarm.py
@@ -51,6 +51,7 @@ class SimxarmDataset(AbstractDataset):
         collate_fn: Callable | None = None,
         writer: Writer | None = None,
         transform: "torchrl.envs.Transform" = None,
+        dummy: bool = False,
     ):
         super().__init__(
             dataset_id,
@@ -64,6 +65,7 @@ class SimxarmDataset(AbstractDataset):
             collate_fn=collate_fn,
             writer=writer,
             transform=transform,
+            dummy=dummy,
         )
 
     def _download_and_preproc_obsolete(self):
diff --git a/lerobot/common/utils.py b/lerobot/common/utils.py
index 2af1d966..86383cdc 100644
--- a/lerobot/common/utils.py
+++ b/lerobot/common/utils.py
@@ -1,9 +1,13 @@
 import logging
+import os.path as osp
 import random
 from datetime import datetime
+from pathlib import Path
 
+import hydra
 import numpy as np
 import torch
+from omegaconf import DictConfig
 
 
 def get_safe_torch_device(cfg_device: str, log: bool = False) -> torch.device:
@@ -63,3 +67,29 @@ def format_big_number(num):
         num /= divisor
 
     return num
+
+
+def _relative_path_between(path1: Path, path2: Path) -> Path:
+    """Returns path1 relative to path2."""
+    path1 = path1.absolute()
+    path2 = path2.absolute()
+    try:
+        return path1.relative_to(path2)
+    except ValueError:  # most likely because path1 is not a subpath of path2
+        common_parts = Path(osp.commonpath([path1, path2])).parts
+        return Path(
+            "/".join([".."] * (len(path2.parts) - len(common_parts)) + list(path1.parts[len(common_parts) :]))
+        )
+
+
+def init_hydra_config(config_path: str, overrides: list[str] | None = None) -> DictConfig:
+    """Initialize a Hydra config given only the path to the relevant config file.
+
+    For config resolution, it is assumed that the config file's parent is the Hydra config dir.
+    """
+    # Hydra needs a path relative to this file.
+    hydra.initialize(
+        str(_relative_path_between(Path(config_path).absolute().parent, Path(__file__).absolute().parent))
+    )
+    cfg = hydra.compose(Path(config_path).stem, overrides)
+    return cfg
diff --git a/lerobot/scripts/eval.py b/lerobot/scripts/eval.py
index 1de0bb0e..72517504 100644
--- a/lerobot/scripts/eval.py
+++ b/lerobot/scripts/eval.py
@@ -30,14 +30,12 @@ python lerobot/scripts/eval.py --hub-id HUB/ID --revision v1.0 eval_episodes=10
 import argparse
 import json
 import logging
-import os.path as osp
 import threading
 import time
 from datetime import datetime as dt
 from pathlib import Path
 
 import einops
-import hydra
 import imageio
 import numpy as np
 import torch
@@ -52,7 +50,7 @@ from lerobot.common.envs.factory import make_env
 from lerobot.common.logger import log_output_dir
 from lerobot.common.policies.abstract import AbstractPolicy
 from lerobot.common.policies.factory import make_policy
-from lerobot.common.utils import get_safe_torch_device, init_logging, set_global_seed
+from lerobot.common.utils import get_safe_torch_device, init_hydra_config, init_logging, set_global_seed
 
 
 def write_video(video_path, stacked_frames, fps):
@@ -195,7 +193,8 @@ def eval(cfg: dict, out_dir=None, stats_path=None):
     log_output_dir(out_dir)
 
     logging.info("Making transforms.")
-    offline_buffer = make_offline_buffer(cfg, stats_path=stats_path)
+    # TODO(alexander-soare): Completely decouple datasets from evaluation.
+    offline_buffer = make_offline_buffer(cfg, stats_path=stats_path, dummy=stats_path is not None)
 
     logging.info("Making environment.")
     env = make_env(cfg, transform=offline_buffer.transform)
@@ -229,19 +228,6 @@ def eval(cfg: dict, out_dir=None, stats_path=None):
     logging.info("End of eval")
 
 
-def _relative_path_between(path1: Path, path2: Path) -> Path:
-    """Returns path1 relative to path2."""
-    path1 = path1.absolute()
-    path2 = path2.absolute()
-    try:
-        return path1.relative_to(path2)
-    except ValueError:  # most likely because path1 is not a subpath of path2
-        common_parts = Path(osp.commonpath([path1, path2])).parts
-        return Path(
-            "/".join([".."] * (len(path2.parts) - len(common_parts)) + list(path1.parts[len(common_parts) :]))
-        )
-
-
 if __name__ == "__main__":
     parser = argparse.ArgumentParser(
         description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter
@@ -259,19 +245,14 @@ if __name__ == "__main__":
 
     if args.config is not None:
         # Note: For the config_path, Hydra wants a path relative to this script file.
-        hydra.initialize(
-            config_path=str(
-                _relative_path_between(Path(args.config).absolute().parent, Path(__file__).parent)
-            )
-        )
-        cfg = hydra.compose(Path(args.config).stem, args.overrides)
+        cfg = init_hydra_config(args.config, args.overrides)
         # TODO(alexander-soare): Save and load stats in trained model directory.
         stats_path = None
     elif args.hub_id is not None:
         folder = Path(snapshot_download(args.hub_id, revision="v1.0"))
-        cfg = hydra.initialize(config_path=str(_relative_path_between(folder, Path(__file__).parent)))
-        cfg = hydra.compose("config", args.overrides)
-        cfg.policy.pretrained_model_path = folder / "model.pt"
+        cfg = init_hydra_config(
+            folder / "config.yaml", [f"policy.pretrained_model_path={folder / 'model.pt'}", *args.overrides]
+        )
         stats_path = folder / "stats.pth"
 
     eval(
diff --git a/tests/test_examples.py b/tests/test_examples.py
index 6c21eb4c..9da7a663 100644
--- a/tests/test_examples.py
+++ b/tests/test_examples.py
@@ -1,19 +1,56 @@
-import pytest
 from pathlib import Path
 
-@pytest.mark.parametrize(
-    "path",
-    [
-        "examples/1_visualize_dataset.py",
-        "examples/2_evaluate_pretrained_policy.py",
-        "examples/3_train_policy.py",
-    ],
-)
-def test_example(path):
 
-    with open(path, 'r') as file:
+def _find_and_replace(text: str, finds: list[str], replaces: list[str]) -> str:
+    for f, r in zip(finds, replaces):
+        assert f in text
+        text = text.replace(f, r)
+    return text
+
+
+def test_example_1():
+    path = "examples/1_visualize_dataset.py"
+
+    with open(path, "r") as file:
         file_contents = file.read()
     exec(file_contents)
 
-    if path == "examples/1_visualize_dataset.py":
-        assert Path("outputs/visualize_dataset/example/episode_0.mp4").exists()
+    assert Path("outputs/visualize_dataset/example/episode_0.mp4").exists()
+
+
+def test_examples_3_and_2():
+    """
+    Train a model with example 3, check the outputs.
+    Evaluate the trained model with example 2, check the outputs.
+    """
+
+    path = "examples/3_train_policy.py"
+
+    with open(path, "r") as file:
+        file_contents = file.read()
+
+    # Do less steps and use CPU.
+    file_contents = _find_and_replace(
+        file_contents,
+        ['"offline_steps=5000"', '"device=cuda"'],
+        ['"offline_steps=1"', '"device=cpu"'],
+    )
+
+    exec(file_contents)
+
+    for file_name in ["model.pt", "stats.pth", "config.yaml"]:
+        assert Path(f"outputs/train/example_pusht_diffusion/{file_name}").exists()
+
+    path = "examples/2_evaluate_pretrained_policy.py"
+
+    with open(path, "r") as file:
+        file_contents = file.read()
+
+    # Do less evals and use CPU.
+    file_contents = _find_and_replace(
+        file_contents,
+        ['"eval_episodes=10"', '"rollout_batch_size=10"', '"device=cuda"'],
+        ['"eval_episodes=1"', '"rollout_batch_size=1"','"device=cpu"'],
+    )
+
+    assert Path(f"outputs/train/example_pusht_diffusion").exists()
\ No newline at end of file