diff --git a/constants_crunchy.py b/constants_crunchy.py
index add67d3dee88047b5750a3dcdce9a276443c27ca..5776a30ee57fa1f2fc8833befaf0a2140ed95760 100644
--- a/constants_crunchy.py
+++ b/constants_crunchy.py
@@ -5,14 +5,14 @@ def mkdir_if_not_exists(dir_name):
     if not os.path.exists(dir_name):
         os.mkdir(dir_name)
         
-experiment_name = 'trial_new_rel_feat'
+experiment_name = 'ans_through_obj_atr_pretrained_no_rel_bin_feats'
 
 ##########################################################################
 #                    Machine Specific Paths                              #
 ##########################################################################
 
 # Global output directory (all subexperiments will be saved here)
-global_output_dir = '/home/tanmay/Code/GenVQA/Exp_Results/VQA'
+global_output_dir = '/home/tanmay/Code/GenVQA/Exp_Results/models_cvpr/'
 
 global_experiment_dir = os.path.join(
     global_output_dir,
@@ -42,9 +42,9 @@ word2vec_binary = '/home/tanmay/Code/word2vec/word2vec-api-master/' + \
 vqa_basedir = '/home/ssd/VQA/'
 
 # Pretrained obj atr model to be restored
-pretrained_model = '/home/tanmay/Code/GenVQA/Exp_Results/VQA/' + \
-    'object_attribute_classifier_wordvec_xform/' + \
-    'object_attribute_classifiers/model-102000'
+pretrained_model = '/home/tanmay/Code/GenVQA/Exp_Results/models_cvpr/' + \
+                   'obj_atr_through_none_single_feat/answer_classifiers/' + \
+                   'model-72000'
 
 ##########################################################################
 #                         Model Parameters                               #
@@ -93,6 +93,10 @@ regions_json = os.path.join(
     data_absolute_path,
     'restructured/region_with_hypernym_labels.json')
 
+eval_regions_json = os.path.join(
+    data_absolute_path,
+    'restructured/region_with_labels.json')
+
 mean_image_filename = os.path.join(
     data_absolute_path,
     'restructured/mean_image.jpg')
@@ -132,7 +136,7 @@ region_model = os.path.join(
 # Object Attribute Classifier Training Params
 region_batch_size = 200
 region_num_epochs = 20
-region_queue_size = 400
+region_queue_size = 300
 region_regularization_coeff = 1e-5
 region_lr = 1e-3
 
@@ -169,7 +173,12 @@ region_attribute_scores_dirname = os.path.join(
     region_output_dir,
     'attribute_scores')
 
+region_object_scores_dirname = os.path.join(
+    region_output_dir,
+    'object_scores')
+
 mkdir_if_not_exists(region_attribute_scores_dirname)
+mkdir_if_not_exists(region_object_scores_dirname)
 
 ##########################################################################
 #                         VQA Parameters                                 #
@@ -253,11 +262,11 @@ answer_model = os.path.join(
     'model')
 
 # Answer classifier training params
-answer_train_from_scratch = True
+answer_train_from_scratch = False
 
 answer_batch_size = 50
 answer_num_epochs = 20
-answer_queue_size = 500
+answer_queue_size = 200
 answer_regularization_coeff = 1e-5
 answer_lr = 1e-3
 
@@ -285,8 +294,8 @@ model_accuracies_txt = os.path.join(
     'model_accuracies.txt')
 
 # Answer eval params
-answer_eval_on = 'testdev'
-answer_model_to_eval = answer_model + '-43000'
+answer_eval_on = 'val'
+answer_model_to_eval = answer_model + '-68000'
 
 vqa_results_dir = os.path.join(
     answer_output_dir,
@@ -315,6 +324,6 @@ raw_vqa_test_ques_json = os.path.join(
     vqa_basedir,
     'MultipleChoice_mscoco_test2015_questions.json')
 
-raw_vqa_val_anno_json = os.path.join(
+raw_vqa_test_anno_json = os.path.join(
     vqa_basedir,
     'mscoco_test2015_annotations.json')