different frequencies prediction corrected

deployment/nbeats/main.py

@@ -97,11 +97,11 @@ def main():
     )
 
     # msg1 = Msg()
-    # msg1.body = '[{"metric": "cpu_usage", "level": 3, "publish_rate": 60000}]'
+    # msg1.body = '[{"metric": "cpu_usage", "level": 3, "publish_rate": 30000}, {"metric": "response_time", "level": 3, "publish_rate": 60000}]'
     # msg2 = Msg()
     # msg2.body = (
     #     "{"
-    #     + f'"metrics": ["cpu_usage"],"timestamp": {int(time.time())}, "epoch_start": {int(time.time()) + 30}, "number_of_forward_predictions": 8,"prediction_horizon": 60'
+    #     + f'"metrics": ["cpu_usage", "response_time"],"timestamp": {int(time.time())}, "epoch_start": {int(time.time()) + 30}, "number_of_forward_predictions": 8,"prediction_horizon": 60'
     #     + "}"
     # )
 

deployment/nbeats/src/model_predict.py

@@ -79,13 +79,6 @@ def predict(
 
     prediction = model.predict(prediction_input, mode="raw")["prediction"]
 
-    it = iter(prediction_input)
-    first = next(it)
-    model.train()  # turning on dropout (in order to obtaining different prediction and calculate confidence interval)
-    predictions_with_dropout = [
-        model.forward(first[0])["prediction"][-1].item() for _ in range(20)
-    ]
-
     predictions_with_dropout = []
     model.train()
     model.loss = RMSE()
@@ -94,7 +87,8 @@ def predict(
             for x, _ in prediction_input:
                 # make prediction
                 out = model(x)  # raw output is dictionary
-                out = model.transform_output(out).item()
+                out = torch.flatten(model.transform_output(out))[-1]
+                out = out.item()
                 predictions_with_dropout.append(out)
 
     # print(model.to_prediction(model.forward(first[0])), "TRANSFORMED")
@@ -105,9 +99,8 @@ def predict(
         scale=st.sem(predictions_with_dropout),
     )
 
-    predicted_values = list(conf_intervals) + [prediction[-1].item()]
+    predicted_values = list(conf_intervals) + [torch.flatten(prediction)[-1].item()]
     predicted_values.sort()  # ensure that predictions and confidence intervals are in correct order
-    print(predicted_values)
 
     msg = {
         target_column: {
@@ -128,5 +121,5 @@ def predict(
     logging.debug(f"prediction msg: {msg}")
 
     future_df["split"] = "val"
-    future_df[target_column] = prediction[-1].item()
+    future_df[target_column] = torch.flatten(prediction).numpy()
     return (msg, future_df)

deployment/tft/main.py

@@ -96,7 +96,7 @@ def main():
     )
 
     # msg1 = Msg()
-    # msg1.body = '[{"metric": "cpu_usage", "level": 3, "publish_rate": 60000}]'
+    # msg1.body = '[{"metric": "cpu_usage", "level": 3, "publish_rate": 45000}]'
     # msg2 = Msg()
     # msg2.body = (
     #     "{"