timer

src/main.py

@@ -39,33 +39,19 @@ def main():
     model.train(input_fn=lambda: esmm_input_fn(train_df, shuffle=True), steps=5000)
     metrics = model.evaluate(input_fn=lambda: esmm_input_fn(val_df, False), steps=5000)
     print("metrics: " + str(metrics))
+
     model_export_path = str(Path("~/data/models/").expanduser())
     save_path = model_export(model, all_features, model_export_path)
     print("save to: " + save_path)
 
-    # predictions = model.predict(input_fn=lambda: esmm_input_fn(test_df, False))
-    # print(next(iter(predictions)))
-
-    test_300 = test_df.sample(300)
-    test_300_2 = test_df.sample(300)
-    test_300_3 = test_df.sample(300)
-
     predict_fn = tf.contrib.predictor.from_saved_model(save_path)
 
-    time_1 = timeit.default_timer()
-    model_predict(test_300, predict_fn)
-    total_1 = (timeit.default_timer() - time_1)
-    print("prediction cost {:.5f} s".format(total_1))
-
-    time_2 = timeit.default_timer()
-    model_predict(test_300_2, predict_fn)
-    total_2 = (timeit.default_timer() - time_2)
-    print("prediction cost {:.5f} s".format(total_2))
-
-    time_3 = timeit.default_timer()
-    model_predict(test_300_3, predict_fn)
-    total_3 = (timeit.default_timer() - time_3)
-    print("prediction cost {:.5f} s".format(total_3))
+    for i in range(10):
+        test_300 = test_df.sample(300)
+        time_1 = timeit.default_timer()
+        model_predict(test_300, predict_fn)
+        total_1 = (timeit.default_timer() - time_1)
+        print("prediction cost {:.5f} s".format(total_1))
 
     total_time = (time.time() - time_begin) / 60
     print("cost {:.2f} mins at {}".format(total_time, datetime.now()))