analyticsindiamagazine

train_InputExamples = train.apply(lambda x: bert.run_classifier.InputExample(guid=None,
                                                                   text_a = x[DATA_COLUMN], 
                                                                   text_b = None, 
                                                                   label = x[LABEL_COLUMN]), axis = 1)
val_InputExamples = val.apply(lambda x: bert.run_classifier.InputExample(guid=None, 
                                                                   text_a = x[DATA_COLUMN], 
                                                                   text_b = None, 
                                                                   label = x[LABEL_COLUMN]), axis = 1)

analyticsindiamagazine

# This is a path to an uncased (all lowercase) version of BERT
BERT_MODEL_HUB = "https://tfhub.dev/google/bert_uncased_L-12_H-768_A-12/1"

def create_tokenizer_from_hub_module():
  """Get the vocab file and casing info from the Hub module."""
  with tf.Graph().as_default():
    bert_module = hub.Module(BERT_MODEL_HUB)
    tokenization_info = bert_module(signature="tokenization_info", as_dict=True)
    with tf.Session() as sess:
      vocab_file, do_lower_case = sess.run([tokenization_info["vocab_file"],

analyticsindiamagazine

def create_model(is_predicting, input_ids, input_mask, segment_ids, labels,
                 num_labels):
  
  bert_module = hub.Module(
      BERT_MODEL_HUB,
      trainable=True)
  bert_inputs = dict(
      input_ids=input_ids,
      input_mask=input_mask,
      segment_ids=segment_ids)

analyticsindiamagazine

FROM tensorflow/tensorflow:1.12.0-py3
ENV LANG=C.UTF-8
RUN mkdir /gpt-2
WORKDIR /gpt-2
ADD . /gpt-2
RUN pip3 install -r requirements.txt

analyticsindiamagazine

analyticsindiamagazine

#Loading Data
train = pd.read_excel("/GD/.../DataSets/Data_Train.xlsx")

#display the dataframe head(10) returns the first 10 rows
train.head(10)

#Shape of a table (number or rows, number of columns)
train.shape

#Display the names of all columns in a dataframe

analyticsindiamagazine

#A function to find the maximun number of features in a single cell
def max_features_in_single_row(train, test, delimiter):
  max_info = 0 
  item_lis = list(train.append(test))
  for i in item_lis:
    if len(i.split("{}".format(delimiter))) > max_info:
      max_info = len(i.split("{}".format(delimiter)))
  print("\n","-"*35)    
  print("Max_Features in One Observation = ", max_info)
  return max_info

analyticsindiamagazine

#A function to find all the non numeric values
def non_numerals(series):
  non_numerals = []
  for i in series.unique():
    try :
      i = float(i)
    except:
      non_numerals.append(i)
  return non_numerals

analyticsindiamagazine

#Filling Nulls/Nans
train.fillna('NAN', inplace = True)

#Label Encoding
from sklearn.preprocessing import LabelEncoder
le_c = LabelEncoder().fit(unique_cuisines)
train['Cuisines'] = le_l.transform(train['Cuisines'])

#Scaling
cols = [list of columns to apply standard scaler]

analyticsindiamagazine

#Splitting training set into training and validation sets
from sklearn.model_selection import train_test_split
train, val = train_test_split(train_sample, test_size = 0.1, random_state = 123)

#Seperating the independent and dependent variables
cols = [list of column names in order] #last column corresponds to y or target variable
#Training set
X_train = train[cols[:-1]]
Y_train = train[cols[-1]]
#Validation set