Merge pull request #146 from Thilakraj1998/main

Minor fix
blobcity · Oct 18, 2021 · c594f25 · c594f25
2 parents 343be68 + 6e792eb
commit c594f25
Show file tree

Hide file tree

Showing 3 changed files with 11 additions and 8 deletions.
diff --git a/blobcity/config/classifier_config.py b/blobcity/config/classifier_config.py
@@ -130,7 +130,8 @@ class classifier_config:
                 'learning_rate': {'float':[1e-3,0.1]},
                 'reg_alpha': {'int':[1, 1.5]},
                 'reg_lambda': {'int':[1, 1.5]},
-                'booster':{'str':['gbtree', 'gblinear','dart']}
+                'booster':{'str':['gbtree', 'gblinear','dart']},
+                'verbosity':{'str':[0]}
             }
         ],
         "RadiusNeighborsClassifier":[

diff --git a/blobcity/main/modelSelection.py b/blobcity/main/modelSelection.py
@@ -16,6 +16,7 @@
 import os
 import warnings
 import itertools
+from tqdm import tqdm,tqdm_notebook
 from blobcity.store import Model
 from blobcity.config import tuner as Tuner
 from blobcity.config import classifier_config,regressor_config
@@ -83,7 +84,7 @@ def trainOnSample(dataframe,target,models,DictClass):
     df=dataframe.sample(frac=0.1,random_state=123)
     X,Y=df.drop(target,axis=1),df[target]
     k=getKFold(X)
-    modelScore={m:cvScore(models[m][0](),X,Y,k) for m in models }
+    modelScore={m:cvScore(models[m][0](),X,Y,k) for m in tqdm_notebook(models) }
     return dict(itertools.islice(sortScore(modelScore).items(), 5))
 
 def trainOnFull(dataframe,target,models,best,DictClass):
@@ -99,7 +100,7 @@ def trainOnFull(dataframe,target,models,best,DictClass):
     """
     X,Y=dataframe.drop(target,axis=1),dataframe[target]
     k=getKFold(X)
-    modelScore={m:cvScore(models[m][0](),X,Y,k) for m in best }
+    modelScore={m:cvScore(models[m][0](),X,Y,k) for m in tqdm_notebook(best) }
     return dict(itertools.islice(sortScore(modelScore).items(), 1))
 
 def modelSearch(dataframe,target,DictClass):

diff --git a/blobcity/utils/AutoFeatureSelection.py b/blobcity/utils/AutoFeatureSelection.py
@@ -37,8 +37,10 @@ def dropHighCorrelationFeatures(X):
         cor_matrix = X.corr()
         upper_tri = cor_matrix.where(np.triu(np.ones(cor_matrix.shape),k=1).astype(np.bool))
         to_drop = [column for column in upper_tri.columns if any(upper_tri[column] > 0.95)]
-        X = X.drop(to_drop, axis=1)
-        return X
+        if to_drop!=[]:
+            return X.drop(to_drop, axis=1)
+        else:
+            return X
 
     def dropConstantFeatures(X):
         """
@@ -52,8 +54,7 @@ def dropConstantFeatures(X):
         cols=X.columns
         constant_filter = VarianceThreshold(threshold=0).fit(X)
         constcols=[col for col in cols if col not in cols[constant_filter.get_support()]]
-        if(constcols!=[]):
-            X.drop(constcols,axis=1,inplace=True)
+        if(constcols!=[]): X.drop(constcols,axis=1,inplace=True)
         return X
 
     def MainScore(resultscore,dc):
@@ -99,7 +100,7 @@ def get_feature_importance(X,Y,score_func,dc):
         if the dataframe has less then equal to 2 features return orignal dataframe. else return a short listed dataframe on the basis of 
         categorical features.
         """
-        if(X.shape[1]<=3):
+        if(X.shape[1]<3):
             return X
         else:
             fit = SelectKBest(score_func=score_func, k=X.shape[1]).fit(X,Y)