diff --git a/.idea/workspace.xml b/.idea/workspace.xml
index 60984d7..af11f3c 100644
--- a/.idea/workspace.xml
+++ b/.idea/workspace.xml
@@ -3,6 +3,7 @@
+
@@ -47,7 +48,7 @@
-
+
@@ -56,7 +57,7 @@
-
+
@@ -91,8 +92,8 @@
-
-
+
+
@@ -164,6 +165,7 @@
train_num
label_in_feature_columns
epoch
+ timeseries_to_supervised
@@ -224,7 +226,7 @@
-
+
@@ -265,12 +267,12 @@
1588152877746
-
+
-
+
@@ -395,8 +397,8 @@
-
-
+
+
diff --git a/train.py b/train.py
index f9f6745..71bd77a 100644
--- a/train.py
+++ b/train.py
@@ -86,7 +86,8 @@ def experiment(repeats, series, seed):
raw_values = series.values
diff_values = difference(raw_values, 1)
# transform data to be supervised learning
- supervised = timeseries_to_supervised(diff_values, 4)
+ lag2 = 4
+ supervised = timeseries_to_supervised(diff_values, lag2)
supervised_values = supervised.values
# split data into train and test-sets
train, test = supervised_values[0:-12], supervised_values[-12:]
@@ -104,7 +105,7 @@ def experiment(repeats, series, seed):
lstm_model = fit_lstm(train_trimmed, batch_size, 30, 4)
# forecast the entire training dataset to build up state for forecasting
if seed:
- train_reshaped = train_trimmed[:, 0].reshape(len(train_trimmed), 1, 1)
+ train_reshaped = train_trimmed[:, 0].reshape(len(train_trimmed), 1, lag2)
lstm_model.predict(train_reshaped, batch_size=batch_size)
# forecast test dataset
test_reshaped = test_scaled[:, 0:-1]