new version with minor bug fixes

Author: AutoViML

deep_autoviml/__init__.py

@@ -45,7 +45,7 @@
 		model_options=model_options, save_model_flag=True, use_my_model='',
 		model_use_case='', verbose=0)
 
-predictions = deepauto.predict(model, project_name, test_dataset=test,
+predictions = deepauto.predict(model, project_name="deep_autoviml", test_dataset=test,
                                  keras_model_type=keras_model_type, 
                                  cat_vocab_dict=cat_vocab_dict)
                                 """ %(module_type, version_number))

deep_autoviml/__version__.py

@@ -20,6 +20,6 @@
 __author__ = "Ram Seshadri"
 __description__ = "deep_autoviml - build and test multiple Tensorflow 2.0 models and pipelines"
 __url__ = "https://github.com/Auto_ViML/deep_autoviml.git"
-__version__ = "0.0.82"
+__version__ = "0.0.84"
 __license__ = "Apache License 2.0"
 __copyright__ = "2020-21 Google"

deep_autoviml/deep_autoviml.py

@@ -212,7 +212,7 @@ def fit(train_data_or_file, target, keras_model_type="basic", project_name="deep
             "compression": None => you can set it to zip or other file compression formats if your data is compressed
             "csv_encoding": default 'utf-8'. But you can set it to any other csv encoding format your data is in
             "label_encode_flag": False. But you can set it to True if you want it encoded.
-            "max_trials": default = 30 ## number of Storm Tuner trials ### Lower this for faster processing.
+            "max_trials": default = 5 ## number of Storm Tuner trials ### Lower this for faster processing.
             "tuner": default = 'storm'  ## Storm Tuner is the default tuner. Optuna is the other option.
             "embedding_size": default = 50 ## this is the NLP embedding size minimum
             "tf_hub_model": default "" (empty string). If you want to supply TF hub model, provide URL here.
@@ -361,7 +361,7 @@ def fit(train_data_or_file, target, keras_model_type="basic", project_name="deep
                             "patience", "epochs", "steps_per_epoch", "optimizer",
                             "kernel_initializer", "num_layers", "class_weight",
                             "loss", "metrics", "monitor","mode", "lr_scheduler","early_stopping",
-                            "class_weight"]
+                            ]
 
     keras_options = copy.deepcopy(keras_options_defaults)
     if len(keras_options_copy) > 0:
@@ -389,7 +389,7 @@ def fit(train_data_or_file, target, keras_model_type="basic", project_name="deep
     model_options_defaults['compression'] = None ## is is needed in case to read Zip files
     model_options_defaults["label_encode_flag"] = '' ## User can set it to True or False depending on their need.
     model_options_defaults["header"] = 0 ### this is the header row for pandas to read
-    model_options_defaults["max_trials"] = 30 ## number of Storm Tuner trials ###
+    model_options_defaults["max_trials"] = 5 ## The number of Storm Tuner trials - make it small ###
     model_options_defaults['tuner'] = 'storm'  ## Storm Tuner is the default tuner. Optuna is the other option.
     model_options_defaults["embedding_size"] = "" ## this is the NLP embedding size minimum
     model_options_defaults["tf_hub_model"] = "" ## If you want to use a pretrained Hub model, provide URL here.

deep_autoviml/modeling/train_model.py

@@ -157,6 +157,7 @@ def train_model(deep_model, full_ds, target, keras_model_type, keras_options,
     callbacks_dict, tb_logpath = get_callbacks(val_mode, val_monitor, patience, learning_rate, 
                             save_weights_only, onecycle_steps, save_model_path)
 
+    early_stopping = check_keras_options(keras_options, "early_stopping", False)
     if keras_options['lr_scheduler'] in ['expo', 'ExponentialDecay', 'exponentialdecay']:
         if early_stopping:
             callbacks_list = [callbacks_dict['early_stop'], callbacks_dict['print']]

deep_autoviml/utilities/utilities.py

@@ -820,7 +820,11 @@ def add_outputs_to_model_body(model_body, meta_outputs):
     ##### This is the simplest way to convert a sequential model to functional!
     for num, each_layer in enumerate(model_body.layers):
         if num == 0:
-            final_outputs = each_layer(meta_outputs)
+            if isinstance(meta_outputs,list):
+                combined_input = layers.concatenate(meta_outputs, name='auto_combined_layer')
+                final_outputs = each_layer(combined_input)
+            else:
+                final_outputs = each_layer(meta_outputs)
         else:
             final_outputs = each_layer(final_outputs)
     return final_outputs

setup.py

@@ -20,7 +20,7 @@
 
 setuptools.setup(
     name="deep_autoviml",
-    version="0.0.82",
+    version="0.0.84",
     author="Ram Seshadri",
     # author_email="[email protected]",
     description="Automatically Build Deep Learning Models and Pipelines fast!",