Added extra branch to main autoencoder for rule_based prediction

deepof/data.py

@@ -331,7 +331,7 @@ class project:
 
             for k, value in tab_dict.items():
                 imputed = IterativeImputer(
-                    max_iter=1000, skip_complete=True
+                    max_iter=1, skip_complete=True ## CHANGE. TESTING ONLY
                 ).fit_transform(value)
                 tab_dict[k] = pd.DataFrame(
                     imputed, index=value.index, columns=value.columns

deepof/train_utils.py

@@ -100,7 +100,7 @@ def get_callbacks(
     elif reg_cat_clusters and reg_cluster_variance:
         latreg = "categorical+variance"
 
-    run_ID = "{}{}{}{}{}{}{}_{}".format(
+    run_ID = "{}{}{}{}{}{}{}{}{}{}{}{}".format(
         ("GMVAE" if variational else "AE"),
         ("_input_type={}".format(input_type) if input_type else "coords"),
         ("_NextSeqPred={}".format(next_sequence_prediction) if variational else ""),
@@ -115,6 +115,8 @@ def get_callbacks(
         ("_{}".format(datetime.now().strftime("%Y%m%d-%H%M%S")) if not run else ""),
     )
 
+    print(run_ID)
+
     log_dir = os.path.abspath(os.path.join(outpath, "fit", run_ID))
     tensorboard_callback = tf.keras.callbacks.TensorBoard(
         log_dir=log_dir,
@@ -599,7 +601,7 @@ def tune_search(
             directory=os.path.join(
                 outpath, "HyperBandx_{}_{}".format(loss, str(date.today()))
             ),
-            max_epochs=35,
+            max_epochs=50,
             hyperband_iterations=hypertun_trials,
             factor=3,
             **hpt_params