print("Train wordbag regressor")
wb_regressor = wordbag_regressor.WordbagRegressor(
"../models/wordbag_model.pkl.gz", tripadvisor_dir)
#wb_regressor= wordbag_regressor.WordbagRegressor("../models/wordbag_model.pkl.gz")
df['wordbag_score'] = wb_regressor.predict(df['text'].values)
import wordhash_regressor
print("Train wordhash regressor")
wh_regressor = wordhash_regressor.WordhashRegressor(
"../models/wordhash_model.pkl.gz", tripadvisor_dir)
#wh_regressor= wordhash_regressor.WordhashRegressor("../models/wordhash_model.pkl.gz")
df['wordhash_score'] = wh_regressor.predict(df['text'].values)
import wordseq_regressor
print("Train wordseq regressor")
ws_regressor = wordseq_regressor.WordseqRegressor(
"../models/wordseq_model.pkl.gz", tripadvisor_dir)
#ws_regressor = wordseq_regressor.WordseqRegressor("../models/wordseq_model.pkl.gz")
df['wordseq_score'] = ws_regressor.predict_batch(df['text'].values)
import wordvec_regressor
print("Train wordvec regressor")
wv_regressor = wordvec_regressor.WordvecRegressor(
"../models/wordvec_model.pkl.gz", tripadvisor_dir)
#wv_regressor= wordvec_regressor.WordvecRegressor("../models/wordvec_model.pkl.gz")
df['wordvec_score'] = wv_regressor.predict(df['text'].values)
df['tweet_len'] = df['text'].map(lambda x: log(1 + len(x)))
df['tweet_wordcount'] = df['text'].map(lambda x: log(1 + len(x.split())))
print(df)
full_preds = np.zeros(df.shape[0])
import wordbag_regressor
print "Train wordbag regressor"
wordbag_regressor= wordbag_regressor.WordbagRegressor("../models/wordbag_model.pkl.gz", tripadvisor_dir)
#wordbag_regressor= wordbag_regressor.WordbagRegressor("../models/wordbag_model.pkl.gz")
df['wordbag_score']= wordbag_regressor.predict(df['text'].values)
import wordhash_regressor
print "Train wordhash regressor"
wordhash_regressor= wordhash_regressor.WordhashRegressor("../models/wordhash_model.pkl.gz", tripadvisor_dir)
#wordhash_regressor= wordhash_regressor.WordhashRegressor("../models/wordhash_model.pkl.gz")
df['wordhash_score']= wordhash_regressor.predict(df['text'].values)
import wordseq_regressor
print "Train wordseq regressor"
wordseq_regressor= wordseq_regressor.WordseqRegressor("../models/wordseq_model.neo", tripadvisor_dir)
#wordseq_regressor= wordseq_regressor.WordseqRegressor("../models/wordseq_model.neo")
df['wordseq_score']= wordseq_regressor.predict_batch(df['text'].values)
import wordvec_regressor
print "Train wordvec regressor"
wordvec_regressor= wordvec_regressor.WordvecRegressor("../models/wordseq_model.pkl.gz", tripadvisor_dir)
#wordvec_regressor= wordvec_regressor.WordvecRegressor("../models/wordseq_model.pkl.gz")
df['wordvec_score'] = wordvec_regressor.predict(df['text'].values)
df['tweet_len']= df['text'].map(lambda x: log(1+len(x)))
df['tweet_wordcount']= df['text'].map(lambda x: log(1+len(x.split())))
full_preds= np.zeros(df.shape[0])
columns_pick= ['tweet_len', 'tweet_wordcount', 'wordbag_score', 'wordhash_score', 'wordseq_score', 'wordvec_score', 'textblob_score'] #Mean Squared Error: 0.297226914949
#columns_pick= ['tweet_len', 'tweet_wordcount', 'wordhash_score', 'wordseq_score', 'wordvec_score', 'textblob_score'] #Mean Squared Error: 0.306232998673