diff --git a/deep/src/main/java/smile/deep/Model.java b/deep/src/main/java/smile/deep/Model.java
index 196517563..34f752104 100644
--- a/deep/src/main/java/smile/deep/Model.java
+++ b/deep/src/main/java/smile/deep/Model.java
@@ -127,6 +127,7 @@ public void train(int epochs, Optimizer optimizer, Loss loss, Dataset train) {
      * @param train the training data.
      * @param eval optional evaluation data.
      * @param checkpoint optional checkpoint file path.
+     * @param metrics the evaluation metrics.
      */
     public void train(int epochs, Optimizer optimizer, Loss loss, Dataset train, Dataset eval, String checkpoint, Metric... metrics) {
         train(); // training mode
@@ -184,6 +185,7 @@ public void train(int epochs, Optimizer optimizer, Loss loss, Dataset train, Dat
     /**
      * Evaluates the model accuracy on a test dataset.
      * @param dataset the test dataset.
+     * @param metrics the evaluation metrics.
      * @return the accuracy.
      */
     public Map<String, Double> eval(Dataset dataset, Metric... metrics) {
diff --git a/deep/src/main/java/smile/deep/layer/Layer.java b/deep/src/main/java/smile/deep/layer/Layer.java
index 59ce3f3b0..ec817f540 100644
--- a/deep/src/main/java/smile/deep/layer/Layer.java
+++ b/deep/src/main/java/smile/deep/layer/Layer.java
@@ -41,7 +41,10 @@ public interface Layer {
      */
     Tensor forward(Tensor input);
 
-    /** Returns the PyTorch Module object. */
+    /**
+     * Returns the PyTorch Module object.
+     * @return the PyTorch Module object.
+     */
     Module asTorch();
 
     /**
@@ -225,7 +228,7 @@ static FullyConnectedLayer hardShrink(int in, int out) {
      * @param size the window size.
      * @return a convolutional layer.
      */
-    static Conv2dLayer conv2d(int in, int out, int size, int pool) {
+    static Conv2dLayer conv2d(int in, int out, int size) {
         return new Conv2dLayer(in, out, size, 1, 1, 1, true);
     }
 
diff --git a/deep/src/main/java/smile/deep/metric/Metric.java b/deep/src/main/java/smile/deep/metric/Metric.java
index 428a904a7..2d0485b6f 100644
--- a/deep/src/main/java/smile/deep/metric/Metric.java
+++ b/deep/src/main/java/smile/deep/metric/Metric.java
@@ -26,7 +26,10 @@
  * @author Haifeng Li
  */
 public interface Metric {
-    /** Returns the name of metric. */
+    /**
+     * Returns the name of metric.
+     * @return the name of metric.
+     */
     String name();
 
     /**
@@ -39,9 +42,10 @@ public interface Metric {
     void update(Tensor output, Tensor target);
 
     /**
-     * Computes the metric values from the metric state, which are updated by
+     * Computes the metric value from the metric state, which are updated by
      * previous update() calls. The compute frequency can be less than the
      * update frequency.
+     * @return the metric value.
      */
     double compute();
 
diff --git a/deep/src/main/java/smile/deep/tensor/Device.java b/deep/src/main/java/smile/deep/tensor/Device.java
index 874b38f99..82d17109b 100644
--- a/deep/src/main/java/smile/deep/tensor/Device.java
+++ b/deep/src/main/java/smile/deep/tensor/Device.java
@@ -53,17 +53,26 @@ public String toString() {
         return value.str().getString();
     }
 
-    /** Returns true if the device is CUDA. */
+    /**
+     * Returns true if the device is CUDA.
+     * @return true if the device is CUDA.
+     */
     public boolean isCUDA() {
         return value.is_cuda();
     }
 
-    /** Returns true if the device is CPU. */
+    /**
+     * Returns true if the device is CPU.
+     * @return true if the device is CPU.
+     */
     public boolean isCPU() {
         return value.is_cpu();
     }
 
-    /** Returns true if the device is MPS. */
+    /**
+     * Returns true if the device is MPS.
+     * @return true if the device is MPS.
+     */
     public boolean isMPS() {
         return value.is_mps();
     }
@@ -77,12 +86,18 @@ public void emptyCache() {
         }
     }
 
-    /** Returns the PyTorch device object. */
+    /**
+     * Returns the PyTorch device object.
+     * @return the PyTorch device object.
+     */
     public org.bytedeco.pytorch.Device asTorch() {
         return this.value;
     }
 
-    /** Returns the preferred (most powerful) device. */
+    /**
+     * Returns the preferred (most powerful) device.
+     * @return the preferred (most powerful) device.
+     */
     public static Device preferredDevice() {
         if (torch.cuda_is_available()) {
             return Device.CUDA();
@@ -145,6 +160,10 @@ public void setDefaultDevice() {
         torch.device(value);
     }
 
+    /**
+     * Returns the device type.
+     * @return the device type.
+     */
     public DeviceType type() {
         return type;
     }
diff --git a/deep/src/main/java/smile/deep/tensor/DeviceType.java b/deep/src/main/java/smile/deep/tensor/DeviceType.java
index e68f0eade..a5c393e0a 100644
--- a/deep/src/main/java/smile/deep/tensor/DeviceType.java
+++ b/deep/src/main/java/smile/deep/tensor/DeviceType.java
@@ -24,6 +24,7 @@
  * @author Haifeng Li
  */
 public enum DeviceType {
+    /** CPU */
     CPU(torch.DeviceType.CPU),
     /** NVIDIA GPU */
     CUDA(torch.DeviceType.CUDA),
@@ -41,6 +42,7 @@ public enum DeviceType {
     /**
      * Returns the byte value of device type,
      * which is compatible with PyTorch.
+     * @return the byte value of device type.
      */
     public byte value() {
         return value.value;
diff --git a/deep/src/main/java/smile/deep/tensor/Tensor.java b/deep/src/main/java/smile/deep/tensor/Tensor.java
index 83aeccc5a..52cb70233 100644
--- a/deep/src/main/java/smile/deep/tensor/Tensor.java
+++ b/deep/src/main/java/smile/deep/tensor/Tensor.java
@@ -51,7 +51,10 @@ public Tensor clone() {
         return Tensor.of(value.to());
     }
 
-    /** Returns the PyTorch tensor object. */
+    /**
+     * Returns the PyTorch tensor object.
+     * @return the PyTorch tensor object.
+     */
     public org.bytedeco.pytorch.Tensor asTorch() {
         return this.value;
     }
@@ -148,6 +151,7 @@ public long size(long dim) {
      * but with the specified shape. This method returns a view
      * if shape is compatible with the current shape.
      *
+     * @param shape the new shape of tensor.
      * @return the tensor with the specified shape.
      */
     public Tensor reshape(long... shape) {
@@ -322,22 +326,34 @@ public double getDouble(long... index) {
         return x.item_double();
     }
 
-    /** Returns the int value when the tensor holds a single value. */
+    /**
+     * Returns the int value when the tensor holds a single value.
+     * @return the int value when the tensor holds a single value.
+     */
     public int toInt() {
         return value.item_int();
     }
 
-    /** Returns the long value when the tensor holds a single value. */
+    /**
+     * Returns the long value when the tensor holds a single value.
+     * @return the long value when the tensor holds a single value.
+     */
     public long toLong() {
         return value.item_long();
     }
 
-    /** Returns the float value when the tensor holds a single value. */
+    /**
+     * Returns the float value when the tensor holds a single value.
+     * @return the float value when the tensor holds a single value.
+     */
     public float toFloat() {
         return value.item_float();
     }
 
-    /** Returns the double value when the tensor holds a single value. */
+    /**
+     * Returns the double value when the tensor holds a single value.
+     * @return the double value when the tensor holds a single value.
+     */
     public double toDouble() {
         return value.item_double();
     }
@@ -935,6 +951,7 @@ public Tensor asin_() {
 
     /**
      * Returns logical AND of two boolean tensors.
+     * @param other another tensor.
      * @return a new tensor of logical and results.
      */
     public Tensor and(Tensor other) {
@@ -943,6 +960,7 @@ public Tensor and(Tensor other) {
 
     /**
      * Returns logical AND of two boolean tensors.
+     * @param other another tensor.
      * @return this tensor.
      */
     public Tensor and_(Tensor other) {
@@ -952,6 +970,7 @@ public Tensor and_(Tensor other) {
 
     /**
      * Returns logical OR of two boolean tensors.
+     * @param other another tensor.
      * @return a new tensor of logical and results.
      */
     public Tensor or(Tensor other) {
@@ -960,6 +979,7 @@ public Tensor or(Tensor other) {
 
     /**
      * Returns logical OR of two boolean tensors.
+     * @param other another tensor.
      * @return this tensor.
      */
     public Tensor or_(Tensor other) {
@@ -968,7 +988,7 @@ public Tensor or_(Tensor other) {
     }
 
     /**
-     * Randomly zeroes some of the elements of the input tensor
+     * Randomly zeroes some elements of the input tensor
      * with probability p.
      *
      * @param p the probability of an element to be zeroed.
@@ -979,7 +999,7 @@ public Tensor dropout(double p) {
     }
 
     /**
-     * Randomly zeroes some of the elements in place
+     * Randomly zeroes some elements in place
      * with probability p.
      *
      * @param p the probability of an element to be zeroed.
@@ -1261,7 +1281,11 @@ public Options() {
             this.value = new TensorOptions();
         }
 
-        /** Sets the data type of the elements stored in the tensor. */
+        /**
+         * Sets the data type of the elements stored in the tensor.
+         * @param type the data type.
+         * @return this options object.
+         */
         public Options dtype(ScalarType type) {
             value = value.dtype(new ScalarTypeOptional(type.value));
             return this;
@@ -1277,13 +1301,22 @@ public Options device(Device device) {
             return this;
         }
 
-        /** Sets strided (dense) or sparse tensor. */
+        /**
+         * Sets strided (dense) or sparse tensor.
+         * @param layout the tensor layout.
+         * @return this options object.
+         */
         public Options layout(Layout layout) {
             value = value.layout(new LayoutOptional(layout.value));
             return this;
         }
 
-        /** Set true if gradients need to be computed for this Tensor. */
+        /**
+         * Set true if gradients need to be computed for this tensor.
+         * @param required the flag indicating if gradients need to be
+         *                computed for this tensor.
+         * @return this options object.
+         */
         public Options requireGradients(boolean required) {
             value = value.requires_grad(new BoolOptional(required));
             return this;