def predict_with_fixed_value(forward,comment,like,submission=True):
# type check
if isinstance(forward,int) and isinstance(forward,int) and isinstance(forward,int):
pass
else:
raise TypeError("forward,comment,like should be type 'int' ")
traindata,testdata = loadData()
# score on the training set
train_real_pred = traindata[['forward','comment','like']]
train_real_pred['fp'],train_real_pred['cp'],train_real_pred['lp'] = forward,comment,like
print ("Score on the training set:{0:.2f}%".format(precision(train_real_pred.values)*100))
# predict on the test data with fixed value, generate submission file
if submission:
test_pred = testdata[['uid','mid']]
test_pred['fp'],test_pred['cp'],test_pred['lp'] = forward,comment,like
result = []
filename = "weibo_predict_{}_{}_{}.txt".format(forward,comment,like)
for _,row in test_pred.iterrows():
result.append("{0}\t{1}\t{2},{3},{4}\n".format(row[0],row[1],row[2],row[3],row[4]))
f = open(filename,'w')
f.writelines(result)
f.close()
print ('generate submission file "{}"'.format(filename))
def predict_with_fixed_value(forward,comment,like,submission=True):
# type check
if isinstance(forward,int) and isinstance(forward,int) and isinstance(forward,int):
pass
else:
raise TypeError("forward,comment,like should be type 'int' ")
traindata,testdata = loadData()
#score on the training set
train_real_pred = traindata[['forward','comment','like']]
train_real_pred['fp'],train_real_pred['cp'],train_real_pred['lp'] = forward,comment,like
print "Score on the training set:{0:.2f}%".format(precision(train_real_pred.values)*100)
#predict on the test data with fixed value, generate submission file
if submission:
test_pred = testdata[['uid','mid']]
test_pred['fp'],test_pred['cp'],test_pred['lp'] = forward,comment,like
result = []
filename = "weibo_predict_{}_{}_{}.txt".format(forward,comment,like)
for _,row in test_pred.iterrows():
result.append("{0}\t{1}\t{2},{3},{4}\n".format(row[0],row[1],row[2],row[3],row[4]))
f = open(filename,'w')
f.writelines(result)
f.close()
print 'generate submission file "{}"'.format(filename)
def predict_with_stat(stat="median",submission=True):
"""
stat:
string
min,max,mean,median
"""
stat_dic = genUidStat()
# 載入資料 并設定欄位
traindata,testdata = loadData()
# get stat for each uid
forward,comment,like = [],[],[]
for uid in traindata['uid']:
if uid in stat_dic:
forward.append(int(stat_dic[uid]["forward_"+stat]))
comment.append(int(stat_dic[uid]["comment_"+stat]))
like.append(int(stat_dic[uid]["like_"+stat]))
else:
forward.append(0)
comment.append(0)
like.append(0)
# score on the training set
train_real_pred = traindata[['forward','comment','like']]
train_real_pred['fp'],train_real_pred['cp'],train_real_pred['lp'] = forward,comment,like
print ("Score on the training set:{0:.2f}%".format(precision(train_real_pred.values)*100))
#predict on the test data with fixed value, generate submission file
if submission:
test_pred = testdata[['uid','mid']]
forward,comment,like = [],[],[]
for uid in testdata['uid']:
if uid in stat_dic:
forward.append(int(stat_dic[uid]["forward_"+stat]))
comment.append(int(stat_dic[uid]["comment_"+stat]))
like.append(int(stat_dic[uid]["like_"+stat]))
else:
forward.append(0)
comment.append(0)
like.append(0)
test_pred['fp'],test_pred['cp'],test_pred['lp'] = forward,comment,like
result = []
filename = "weibo_predict_{}.txt".format(stat)
for _,row in test_pred.iterrows():
result.append("{0}\t{1}\t{2},{3},{4}\n".format(row[0],row[1],row[2],row[3],row[4]))
f = open(filename,'w')
f.writelines(result)
f.close()
print ('generate submission file "{}"'.format(filename))
def predict_with_stat(stat="median",submission=True):
"""
stat:
string
min,max,mean,median
"""
stat_dic = genUidStat()
traindata,testdata = loadData()
#get stat for each uid
forward,comment,like = [],[],[]
for uid in traindata['uid']:
if stat_dic.has_key(uid):
forward.append(int(stat_dic[uid]["forward_"+stat]))
comment.append(int(stat_dic[uid]["comment_"+stat]))
like.append(int(stat_dic[uid]["like_"+stat]))
else:
forward.append(0)
comment.append(0)
like.append(0)
#score on the training set
train_real_pred = traindata[['forward','comment','like']]
train_real_pred['fp'],train_real_pred['cp'],train_real_pred['lp'] = forward,comment,like
print "Score on the training set:{0:.2f}%".format(precision(train_real_pred.values)*100)
#predict on the test data with fixed value, generate submission file
if submission:
test_pred = testdata[['uid','mid']]
forward,comment,like = [],[],[]
for uid in testdata['uid']:
if stat_dic.has_key(uid):
forward.append(int(stat_dic[uid]["forward_"+stat]))
comment.append(int(stat_dic[uid]["comment_"+stat]))
like.append(int(stat_dic[uid]["like_"+stat]))
else:
forward.append(0)
comment.append(0)
like.append(0)
test_pred['fp'],test_pred['cp'],test_pred['lp'] = forward,comment,like
result = []
filename = "weibo_predict_{}.txt".format(stat)
for _,row in test_pred.iterrows():
result.append("{0}\t{1}\t{2},{3},{4}\n".format(row[0],row[1],row[2],row[3],row[4]))
f = open(filename,'w')
f.writelines(result)
f.close()
print 'generate submission file "{}"'.format(filename)
def search_all_uid():
"""
traindata,testdata = loadData()
stat_dic = genUidStat()
#for each uid,search its best fp,cp,lp
uid_best_pred = {}
for uid in stat_dic:
print "search uid: {}".format(uid)
uid_data = traindata[traindata.uid == uid]
args = stat_dic[uid][['forward_min','forward_median','forward_max','comment_min',\
'comment_median','comment_max','like_min','like_median','like_max']]
args = tuple([int(i) for i in args])
fp = search(uid_data,'forward',args)
cp = search(uid_data,'comment',args)
lp = search(uid_data,'like',args)
uid_best_pred[uid] = [fp,cp,lp]
"""
#multiprocessing version for geting uid_best_pred
traindata,testdata = loadData()
stat_dic = genUidStat()
uid_best_pred = {}
pool = Pool()
uids,f,c,l = [],[],[],[]
for uid in stat_dic:
print "search uid:{}".format(uid)
uid_data = traindata[traindata.uid == uid]
arguments = stat_dic[uid][['forward_min','forward_median','forward_max','comment_min',\
'comment_median','comment_max','like_min','like_median','like_max']]
arguments = tuple([int(i) for i in arguments])
f.append(pool.apply_async(search,args=(uid_data,'forward',arguments)))
c.append(pool.apply_async(search,args=(uid_data,'comment',arguments)))
l.append(pool.apply_async(search,args=(uid_data,'like',arguments)))
uids.append(uid)
pool.close()
pool.join()
f = [i.get() for i in f]
c = [i.get() for i in c]
l = [i.get() for i in l]
for i in range(len(uids)):
uid_best_pred[uids[i]] = [f[i],c[i],l[i]]
try:
cPickle.dump(uid_best_pred,open('uid_best_pred.pkl','w'))
except Exception:
pass
return uid_best_pred
def predict_by_search(submission=True):
traindata,testdata = loadData()
uid_best_pred = search_all_uid()
print "search done,now predict on traindata and testdata..."
#predict traindata with uid's best fp,cp,lp
forward,comment,like = [],[],[]
for uid in traindata['uid']:
if uid_best_pred.has_key(uid):
forward.append(int(uid_best_pred[uid][0]))
comment.append(int(uid_best_pred[uid][1]))
like.append(int(uid_best_pred[uid][2]))
else:
forward.append(0)
comment.append(0)
like.append(0)
#score on the traindata
train_real_pred = traindata[['forward','comment','like']]
train_real_pred['fp'],train_real_pred['cp'],train_real_pred['lp'] = forward,comment,like
print "Score on the training set:{0:.2f}%".format(precision(train_real_pred.values)*100)
if submission:
test_pred = testdata[['uid','mid']]
forward,comment,like = [],[],[]
for uid in testdata['uid']:
if uid_best_pred.has_key(uid):
forward.append(int(uid_best_pred[uid][0]))
comment.append(int(uid_best_pred[uid][1]))
like.append(int(uid_best_pred[uid][2]))
else:
forward.append(0)
comment.append(0)
like.append(0)
test_pred['fp'],test_pred['cp'],test_pred['lp'] = forward,comment,like
#generate submission file
result = []
filename = "weibo_predict_search.txt"
for _,row in test_pred.iterrows():
result.append("{0}\t{1}\t{2},{3},{4}\n".format(row[0],row[1],row[2],row[3],row[4]))
f = open(filename,'w')
f.writelines(result)
f.close()
print 'generate submission file "{}"'.format(filename)
