diff --git a/examples/dynamics_analysis/highdim_RNN_Analysis.py b/examples/dynamics_analysis/highdim_RNN_Analysis.py
index 87cb41ae3..a96a1c1d9 100644
--- a/examples/dynamics_analysis/highdim_RNN_Analysis.py
+++ b/examples/dynamics_analysis/highdim_RNN_Analysis.py
@@ -36,7 +36,7 @@ def __init__(
             self.alpha = 1
         else:
             self.alpha = dt / self.tau
-        self.rng = bm.random.RandomState(seed=seed)
+        self.rng = bm.random.RandomState(seed)
 
         # input weight
         self.w_ir = bm.TrainVar(bp.init.parameter(w_ir, (num_input, num_hidden)))
diff --git a/examples/dynamics_simulation/whole_brain_simulation_with_fhn.py b/examples/dynamics_simulation/whole_brain_simulation_with_fhn.py
index bd2ff7c62..b6aeab940 100644
--- a/examples/dynamics_simulation/whole_brain_simulation_with_fhn.py
+++ b/examples/dynamics_simulation/whole_brain_simulation_with_fhn.py
@@ -1,6 +1,8 @@
 # -*- coding: utf-8 -*-
 
 
+import os
+
 import matplotlib.pyplot as plt
 import numpy as np
 
@@ -25,11 +27,11 @@ class Network(bp.DynSysGroup):
     def __init__(self, signal_speed=20.):
         super(Network, self).__init__()
 
-        # Please download the processed data "hcp.npz" of the
-        # ConnectomeDB of the Human Connectome Project (HCP)
-        # from the following link:
-        # - https://share.weiyun.com/wkPpARKy
-        hcp = np.load('../../tests/simulation/data/hcp.npz')
+        # HCP connectome data ("hcp.npz", processed from the ConnectomeDB of
+        # the Human Connectome Project) is bundled with BrainPy at
+        # ``brainpy/dyn/rates/data/hcp.npz``.
+        hcp_path = os.path.join(os.path.dirname(bp.__file__), 'dyn', 'rates', 'data', 'hcp.npz')
+        hcp = np.load(hcp_path)
         conn_mat = bm.asarray(hcp['Cmat'])
         bm.fill_diagonal(conn_mat, 0)
         delay_mat = bm.round(hcp['Dmat'] / signal_speed / bm.dt).astype(bm.int_)
diff --git a/examples/dynamics_simulation/whole_brain_simulation_with_sl_oscillator.py b/examples/dynamics_simulation/whole_brain_simulation_with_sl_oscillator.py
index eece6cb8b..509ec5e79 100644
--- a/examples/dynamics_simulation/whole_brain_simulation_with_sl_oscillator.py
+++ b/examples/dynamics_simulation/whole_brain_simulation_with_sl_oscillator.py
@@ -1,5 +1,7 @@
 # -*- coding: utf-8 -*-
 
+import os
+
 import matplotlib.pyplot as plt
 import numpy as np
 
@@ -26,11 +28,11 @@ class Network(bp.DynSysGroup):
     def __init__(self, noise=0.14):
         super(Network, self).__init__()
 
-        # Please download the processed data "hcp.npz" of the
-        # ConnectomeDB of the Human Connectome Project (HCP)
-        # from the following link:
-        # - https://share.weiyun.com/wkPpARKy
-        hcp = np.load('../../tests/simulation/data/hcp.npz')
+        # HCP connectome data ("hcp.npz", processed from the ConnectomeDB of
+        # the Human Connectome Project) is bundled with BrainPy at
+        # ``brainpy/dyn/rates/data/hcp.npz``.
+        hcp_path = os.path.join(os.path.dirname(bp.__file__), 'dyn', 'rates', 'data', 'hcp.npz')
+        hcp = np.load(hcp_path)
         conn_mat = bm.asarray(hcp['Cmat'])
         bm.fill_diagonal(conn_mat, 0)
         gc = 0.6  # global coupling strength
diff --git a/examples/dynamics_training/Song_2016_EI_RNN.py b/examples/dynamics_training/Song_2016_EI_RNN.py
index c7209823c..6feb3b22e 100644
--- a/examples/dynamics_training/Song_2016_EI_RNN.py
+++ b/examples/dynamics_training/Song_2016_EI_RNN.py
@@ -38,7 +38,7 @@ def __init__(
         self.i_size = num_hidden - self.e_size
         self.alpha = dt / self.tau
         self.sigma_rec = (2 * self.alpha) ** 0.5 * sigma_rec  # Recurrent noise
-        self.rng = bm.random.RandomState(seed=seed)
+        self.rng = bm.random.RandomState(seed)
 
         # hidden mask
         mask = np.tile([1] * self.e_size + [-1] * self.i_size, (num_hidden, 1))
diff --git a/examples/dynamics_training/integrate_flax_into_brainpy.py b/examples/dynamics_training/integrate_flax_into_brainpy.py
index acb0d1e0a..6dccd09f3 100644
--- a/examples/dynamics_training/integrate_flax_into_brainpy.py
+++ b/examples/dynamics_training/integrate_flax_into_brainpy.py
@@ -30,7 +30,7 @@ class Network(bp.DynamicalSystem):
     def __init__(self):
         super(Network, self).__init__()
         self.cnn = bp.layers.FromFlax(CNN(), bm.ones([1, 4, 28, 1]))
-        self.rnn = bp.layers.GRUCell(256, 100)
+        self.rnn = bp.dyn.GRUCell(256, 100)
         self.linear = bp.layers.Dense(100, 10)
 
     def update(self, x):
diff --git a/examples/training_ann_models/mnist_ResNet.py b/examples/training_ann_models/mnist_ResNet.py
index 9355bce72..19d9860cc 100644
--- a/examples/training_ann_models/mnist_ResNet.py
+++ b/examples/training_ann_models/mnist_ResNet.py
@@ -89,11 +89,11 @@ def update(self, s, x):
 
 
 class ResNet(bp.DynamicalSystem):
-    def __init__(self, block, num_blocks, num_classes=10, zero_init_residual=False):
+    def __init__(self, block, num_blocks, num_classes=10, in_channels=3, zero_init_residual=False):
         super(ResNet, self).__init__()
         self.in_planes = 64
 
-        self.conv1 = bp.layers.Conv2D(3, 64, kernel_size=(3, 3), strides=(1, 1), padding=(1, 1),
+        self.conv1 = bp.layers.Conv2D(in_channels, 64, kernel_size=(3, 3), strides=(1, 1), padding=(1, 1),
                                       w_initializer=weight_init)
         self.bn1 = bp.layers.BatchNorm2D(64)
         self.layer1 = self._make_layer(block, 64, num_blocks[0], stride=1)
@@ -208,7 +208,7 @@ def main():
     y_test = bm.asarray(test_set.targets, dtype=bm.int_)
 
     with bm.training_environment():
-        net = ResNet18(num_classes=10)
+        net = ResNet18(num_classes=10, in_channels=1)
 
     # loss function
     def loss_fun(X, Y, fit=True):