diff --git a/examples/simple_train.py b/examples/simple_train.py
new file mode 100644
index 0000000..a4bf502
--- /dev/null
+++ b/examples/simple_train.py
@@ -0,0 +1,41 @@
+# %%
+import torch
+from cellmap_data import CellMapDataset, CellMapDataLoader
+from torchvision.models import resnet18
+
+# %%
+# Define the dataset files to use
+dataset_dict = {
+    "train": {"raw": "train_data.zarr/raw", "gt": "train_data.zarr/gt", "weight": 1.0},
+    "val": {"raw": "val_data.zarr/raw", "gt": "val_data.zarr/gt"},
+    "test": {"raw": "test_data.zarr/raw", "gt": "test_data.zarr/gt"},
+}
+
+# %%
+# Create the dataset and dataloader
+dataset = CellMapDataset(dataset_dict)
+dataloader = CellMapDataLoader(dataset, batch_size=4, shuffle=True, num_workers=0)
+
+# %%
+# Create the network
+model = resnet18(num_classes=2)
+
+# %%
+# Define the loss function and optimizer
+loss = torch.nn.MSELoss()
+optimizer = torch.optim.Adam(model.parameters(), lr=0.001)
+
+# %%
+# Train the network
+for epoch in range(10):
+    for i, data in enumerate(dataloader):
+        inputs, targets = data
+        optimizer.zero_grad()
+        outputs = model(inputs)
+        loss_value = loss(outputs, targets)
+        loss_value.backward()
+        optimizer.step()
+        print(f"Epoch {epoch}, Batch {i}, Loss {loss_value.item()}")
+# %%
+# Save the trained model
+torch.save(model.state_dict(), "model.pth")
diff --git a/pyproject.toml b/pyproject.toml
index 0e7f05d..9259fd0 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -41,6 +41,7 @@ test = ["pytest>=6.0", "pytest-cov"]
 dev = [
     "black",
     "ipython",
+    "jupyter",
     "mypy",
     "pdbpp",
     "pre-commit",
diff --git a/src/cellmap_data/__init__.py b/src/cellmap_data/__init__.py
index 7419624..ec9266f 100644
--- a/src/cellmap_data/__init__.py
+++ b/src/cellmap_data/__init__.py
@@ -14,3 +14,6 @@
 
 __author__ = "Jeff Rhoades"
 __email__ = "rhoadesj@hhmi.org"
+
+from .dataset import CellMapDataset
+from .dataloader import CellMapDataLoader
diff --git a/src/cellmap_data/dataloader.py b/src/cellmap_data/dataloader.py
new file mode 100644
index 0000000..194a2c7
--- /dev/null
+++ b/src/cellmap_data/dataloader.py
@@ -0,0 +1,6 @@
+from torch.utils.data import DataLoader
+from cellmap_data.load import transforms
+
+
+class CellMapDataLoader(DataLoader):
+    def __init__(self, dataset, batch_size=1, shuffle=False, num_workers=0): ...
diff --git a/src/cellmap_data/dataset.py b/src/cellmap_data/dataset.py
index f75404e..432683f 100644
--- a/src/cellmap_data/dataset.py
+++ b/src/cellmap_data/dataset.py
@@ -1,24 +1,27 @@
 # %%
 import csv
+from typing import Dict, Optional
 from torch.utils.data import Dataset
 import tensorstore as tswift
+from fibsem_tools import read, read_xarray
 
 
 # %%
 class CellMapDataset(Dataset):
-    def __init__(self, ...):
-        ...
+    def __init__(
+        self, dataset_dict: Optional[Dict[str, Dict[str, str | float]]] = None
+    ):
+        self.dataset_dict = dataset_dict
+        self.construct()
 
-    def __len__(self):
-        ...
+    def __len__(self): ...
 
-    def __getitem__(self, idx):
-        ...
+    def __getitem__(self, idx): ...
 
     def from_csv(self, csv_path):
         # Load file data from csv file
         dataset_dict = {}
-        with open(csv_path, 'r') as f:
+        with open(csv_path, "r") as f:
             reader = csv.reader(f)
             for row in reader:
                 if row[0] not in dataset_dict:
@@ -29,7 +32,11 @@ def from_csv(self, csv_path):
                     dataset_dict[row[0]]["weight"] = row[3]
                 else:
                     dataset_dict[row[0]]["weight"] = 1.0
-        
+
         self.dataset_dict = dataset_dict
-        
+        self.construct()
+
+    def construct(self): ...
+
+
 # %%