Adds support for python 3.5 and python 3.6.

README.md

@@ -14,7 +14,7 @@ Yaguang Li, Rose Yu, Cyrus Shahabi, Yan Liu, [Diffusion Convolutional Recurrent
 - numpy>=1.12.1
 - pandas>=0.19.2
 - tensorflow>=1.3.0
-- python 2.7
+
 
 Dependency can be installed using the following command:
 ```bash
@@ -38,7 +38,7 @@ python gen_adj_mx.py  --sensor_ids_filename=data/sensor_graph/graph_sensor_ids.t
 
 ## Train the Model
 ```bash
-python dcrnn_train.py --config_filename=data/model/dcrnn_config.json
+python dcrnn_train.py --config_filename=data/model/dcrnn_config.yaml
 ```
 
 
@@ -59,7 +59,7 @@ If you find this repository useful in your research, please cite the following p
 @inproceedings{li2018dcrnn_traffic,
   title={Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting},
   author={Li, Yaguang and Yu, Rose and Shahabi, Cyrus and Liu, Yan},
-  booktitle={International Conference on Learning Representations},
+  booktitle={International Conference on Learning Representations (ICLR '18)},
   year={2018}
 }
 ```

gen_adj_mx.py

@@ -59,5 +59,5 @@ if __name__ == '__main__':
     distance_df = pd.read_csv(FLAGS.distances_filename, dtype={'from': 'str', 'to': 'str'})
     _, sensor_id_to_ind, adj_mx = get_adjacency_matrix(distance_df, sensor_ids)
     # Save to pickle file.
-    with open(FLAGS.output_pkl_filename, 'w') as f:
-        pickle.dump([sensor_ids, sensor_id_to_ind, adj_mx], f)
+    with open(FLAGS.output_pkl_filename, 'wb') as f:
+        pickle.dump([sensor_ids, sensor_id_to_ind, adj_mx], f, protocol=2)

lib/dcrnn_utils.py

@@ -4,11 +4,11 @@ import scipy.sparse as sp
 
 from scipy.sparse import linalg
 from lib.tf_utils import sparse_matrix_to_tf_sparse_tensor
+from lib.utils import load_pickle
 
 
 def load_graph_data(pkl_filename):
-    with open(pkl_filename) as f:
-        sensor_ids, sensor_id_to_ind, adj_mx = pickle.load(f)
+    sensor_ids, sensor_id_to_ind, adj_mx = load_pickle(pkl_filename)
     return sensor_ids, sensor_id_to_ind, adj_mx
 
 

lib/utils.py

@@ -1,6 +1,7 @@
 import datetime
 import numpy as np
 import pandas as pd
+import pickle
 
 
 class StandardScaler:
@@ -155,6 +156,18 @@ def generate_graph_seq2seq_io_data_with_time(df, batch_size, seq_len, horizon, n
     return x, y
 
 
+def load_pickle(pickle_file):
+    try:
+        with open(pickle_file, 'rb') as f:
+            pickle_data = pickle.load(f)
+    except UnicodeDecodeError as e:
+        with open(pickle_file, 'rb') as f:
+            pickle_data = pickle.load(f, encoding='latin1')
+    except Exception as e:
+        print('Unable to load data ', pickle_file, ':', e)
+        raise
+    return pickle_data
+
 def round_down(num, divisor):
     return num - (num % divisor)