diff --git a/data/model/dcrnn_config.yaml b/data/model/dcrnn_config.yaml
index 60b1d5a..27b2bf0 100644
--- a/data/model/dcrnn_config.yaml
+++ b/data/model/dcrnn_config.yaml
@@ -14,7 +14,6 @@ model:
   input_dim: 2
   l1_decay: 0
   max_diffusion_step: 2
-  max_grad_norm: 5
   num_nodes: 207
   num_rnn_layers: 2
   output_dim: 1
@@ -27,10 +26,13 @@ train:
   dropout: 0
   epoch: 0
   epochs: 100
+  epsilon: 1.0e-3
   global_step: 0
   lr_decay_ratio: 0.1
-  steps: [20, 30, 40, 50]
+  max_grad_norm: 5
   max_to_keep: 100
   min_learning_rate: 2.0e-06
+  optimizer: adam
   patience: 50
+  steps: [20, 30, 40, 50]
   test_every_n_epochs: 10
\ No newline at end of file
diff --git a/data/model/pretrained/config.yaml b/data/model/pretrained/config.yaml
new file mode 100644
index 0000000..621ba8d
--- /dev/null
+++ b/data/model/pretrained/config.yaml
@@ -0,0 +1,40 @@
+base_dir: data/model
+data:
+  batch_size: 64
+  dataset_dir: data/METR-LA
+  graph_pkl_filename: data/sensor_graph/adj_mx.pkl
+  test_batch_size: 64
+model:
+  cl_decay_steps: 2000
+  filter_type: dual_random_walk
+  horizon: 12
+  input_dim: 2
+  l1_decay: 0
+  max_diffusion_step: 2
+  num_nodes: 207
+  num_rnn_layers: 2
+  output_dim: 1
+  rnn_units: 64
+  seq_len: 12
+  use_curriculum_learning: true
+train:
+  base_lr: 0.01
+  dropout: 0
+  epoch: 64
+  epochs: 100
+  epsilon: 0.001
+  global_step: 24375
+  log_dir: data/model/pretrained/
+  lr_decay_ratio: 0.1
+  max_grad_norm: 5
+  max_to_keep: 100
+  min_learning_rate: 2.0e-06
+  model_filename: data/model/pretrained/models-2.7422-24375
+  optimizer: adam
+  patience: 50
+  steps:
+  - 20
+  - 30
+  - 40
+  - 50
+  test_every_n_epochs: 10
diff --git a/data/model/pretrained/models-2.7422-24375.data-00000-of-00001 b/data/model/pretrained/models-2.7422-24375.data-00000-of-00001
new file mode 100644
index 0000000..c5862c0
Binary files /dev/null and b/data/model/pretrained/models-2.7422-24375.data-00000-of-00001 differ
diff --git a/data/model/pretrained/models-2.7422-24375.index b/data/model/pretrained/models-2.7422-24375.index
new file mode 100644
index 0000000..feee3e1
Binary files /dev/null and b/data/model/pretrained/models-2.7422-24375.index differ
diff --git a/run_demo.py b/run_demo.py
index 5b8fedd..c24a5b0 100644
--- a/run_demo.py
+++ b/run_demo.py
@@ -1,6 +1,6 @@
 import argparse
+import numpy as np
 import os
-import pandas as pd
 import sys
 import tensorflow as tf
 import yaml
@@ -21,21 +21,17 @@ def run_dcrnn(args):
     with tf.Session(config=tf_config) as sess:
         supervisor = DCRNNSupervisor(adj_mx=adj_mx, **config)
         supervisor.restore(sess, config=config)
-        df_preds = supervisor.test_and_write_result(sess, config['global_step'])
-        # TODO (yaguang): save this file to the npz file.
-        for horizon_i in df_preds:
-            df_pred = df_preds[horizon_i]
-            filename = os.path.join('data/results/', 'dcrnn_prediction_%d.h5' % (horizon_i + 1))
-            df_pred.to_hdf(filename, 'results')
-        print('Predictions saved as data/results/dcrnn_seq2seq_prediction_[1-12].h5...')
+        outputs = supervisor.test_and_write_result(sess, config['train']['global_step'])
+        np.savez_compressed(args.output_filename, **outputs)
+        print('Predictions saved as {}.'.format(args.output_filename))
 
 
 if __name__ == '__main__':
     sys.path.append(os.getcwd())
     parser = argparse.ArgumentParser()
-    parser.add_argument('--traffic_df_filename', default='data/df_highway_2012_4mon_sample.h5',
-                        type=str, help='Traffic data file.')
     parser.add_argument('--use_cpu_only', default=False, type=str, help='Whether to run tensorflow on cpu.')
-    parser.add_argument('--config_filename', default=None, type=str, help='Config file for pretrained model.')
+    parser.add_argument('--config_filename', default='data/model/pretrained/config.yaml', type=str,
+                        help='Config file for pretrained model.')
+    parser.add_argument('--output_filename', default='data/dcrnn_predictions.npz')
     args = parser.parse_args()
     run_dcrnn(args)