def train_model(use_char = False ):
print('load data')
import data_utils, training_utils
conf = data_utils.TrainConfigure()
data_dict = data_utils.pickle_load(conf.char_file)
print('loading embed ...')
if use_char:
vocab_dict = data_utils.pickle_load(conf.char_dict)
embed_matrix = data_utils.load_embedding(vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/char_embed.pkl')
MAX_LEN = 600
name = 'deepcnn_model.h5'
x = data_dict['x']
else:
x = data_utils.pickle_load(conf.term_file)
vocab_dict = data_utils.pickle_load(conf.term_dict)
embed_matrix = data_utils.load_embedding(vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/term_embed.pkl')
MAX_LEN = 300
name = 'deepcnn_model_term.h5'
print('load embed done.')
y = to_categorical(data_dict['y'])
x_tn, y_tn, x_ts, y_ts = training_utils.split(x, y, shuffle=False)
x_tn, y_tn, x_val, y_val = training_utils.split(x_tn, y_tn, shuffle=False)
print('train')
print('define model')
model = DeepCNNModel(embed_matrix=embed_matrix, MAX_LEN=MAX_LEN, name=name)
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
def load_condition_stacking_main():
tn_conf = TrainConfigure()
term_vocab_dict = data_utils.pickle_load(tn_conf.term_dict)
term_embed_matrix = data_utils.load_embedding(term_vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/term_embed.pkl')
# term_embed_matrix = data_utils.load_embedding(term_vocab_dict,
# 'data/sgns.target.word-word.dynwin5.thr10.neg5.dim300.iter5',
# dump_path='data/term_embed_ww.pkl')
char_vocab_dict = data_utils.pickle_load(tn_conf.char_dict)
char_embed_matrix = data_utils.load_embedding(char_vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/char_embed.pkl')
name = 'model/stack_condition_model.pkl'
model_dir = 'model/stack/'
n_fold = 3
name = 'model/stack_condition_model5.pkl'
model_dir = 'model/stack5/'
n_fold = 5
conf = conditionmodelbase.ModelConfigure()
stk_model = stacking(n_fold, name=name, is_condition=True)
stk_model.add_model(ConditionConvModel, {"conf":conf,"char_embed_matrix": char_embed_matrix,
"term_embed_matrix": term_embed_matrix,
"name": model_dir + 'conditionconvmodel_PE.h5'})
stk_model.add_model(ConditionDPCNNModel, {"conf":conf,"char_embed_matrix": char_embed_matrix,
"term_embed_matrix": term_embed_matrix,
"name": model_dir + 'conditiondpcnnmodel_PE.h5'})
stk_model.add_model(ConditionGatedConvModel, {"conf":conf,"char_embed_matrix": char_embed_matrix,
"term_embed_matrix": term_embed_matrix,
"name": model_dir + 'conditiongatedconvmodel_PE.h5'})
stk_model.add_model(ConditionGatedDeepCNNModel, {"conf":conf,"char_embed_matrix": char_embed_matrix,
"term_embed_matrix": term_embed_matrix,
"name": model_dir + 'conditiongateddeepcnnmodel_PE.h5'})
stk_model.load( )
return stk_model
def train_model_pe():
print('load data')
import data_utils, training_utils
conf = data_utils.TrainConfigure()
data_dict = data_utils.pickle_load(conf.char_file)
print('loading embed ...')
vocab_dict = data_utils.pickle_load(conf.char_dict)
embed_matrix = data_utils.load_embedding(
vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/char_embed.pkl')
print('load embed done.')
y = to_categorical(data_dict['y'])
xe = [[i for i in range(600)] for _ in range(y.shape[0])]
xe = np.array(xe)
x_tn, y_tn, x_ts, y_ts = training_utils.split([data_dict['x'], xe],
y,
shuffle=False)
x_tn, y_tn, x_val, y_val = training_utils.split(x_tn, y_tn, shuffle=False)
print('train')
print('define model')
model = SSCharModel(embed_matrix=embed_matrix,
name='sscharmodel_PE.h5',
PE=True)
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
del model
model = SSCharModel(embed_matrix=embed_matrix,
name='sscharmodel_PE.h5',
PE=True,
train_embed=True)
model.load_weights()
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
def train_model_pe( use_char = True):
print('define model')
print('load data')
import data_utils, training_utils
tn_conf = data_utils.TrainConfigure()
data_dict = data_utils.pickle_load('data.dict')
y = to_categorical(data_dict['y'])
if use_char:
vocab_dict = data_utils.pickle_load(tn_conf.char_dict)
embed_matrix = data_utils.load_embedding(vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/char_embed.pkl')
x = data_dict['x']
MAX_LEN = 600
name = 'conditionmodel_PE.h5'
else:
x = data_utils.pickle_load(tn_conf.term_file)
vocab_dict = data_utils.pickle_load(tn_conf.term_dict)
embed_matrix = data_utils.load_embedding(vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/term_embed.pkl')
MAX_LEN = 300
name = 'conditionmodel_term_PE.h5'
xe = [[i for i in range(MAX_LEN)] for _ in range(y.shape[0])]
xe = np.array(xe)
x_tn, y_tn, x_ts, y_ts = training_utils.split([x, xe], y, shuffle=False)
x_tn, y_tn, x_val, y_val = training_utils.split(x_tn, y_tn, shuffle=False)
print('train')
model = ConditionModel(embed_matrix=embed_matrix, MAX_LEN=MAX_LEN, name=name, PE=True)
x_tn, y_tn = model.gen_train(x_tn, y_tn)
x_val, y_val = model.gen_train(x_val, y_val)
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
def train_main_pe():
print('load data')
import data_utils, training_utils
tn_conf = TrainConfigure()
data_dict = data_utils.pickle_load(tn_conf.char_file)
y = to_categorical(data_dict['y'])
x = data_dict['x']
xterm = data_utils.pickle_load(tn_conf.term_file)
xfeat = data_utils.pickle_load(tn_conf.feat_file)
# normalization
from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
scaler.fit(xfeat)
data_utils.pickle_dump(scaler, tn_conf.feat_norm)
xfeat = scaler.transform(xfeat)
print('loading embed ...')
term_vocab_dict = data_utils.pickle_load(tn_conf.term_dict)
term_embed_matrix = data_utils.load_embedding(
term_vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/term_embed.pkl')
# term_embed_matrix = data_utils.load_embedding(term_vocab_dict,
# 'data/sgns.target.word-word.dynwin5.thr10.neg5.dim300.iter5',
# dump_path='data/term_embed_ww.pkl')
char_vocab_dict = data_utils.pickle_load(tn_conf.char_dict)
char_embed_matrix = data_utils.load_embedding(
char_vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/char_embed.pkl')
print('load embed done.')
xe_char = [[i for i in range(600)] for _ in range(y.shape[0])]
xe_char = np.array(xe_char)
xe_term = [[i for i in range(300)] for _ in range(y.shape[0])]
xe_term = np.array(xe_term)
x_tn, y_tn, x_ts, y_ts = training_utils.split(
[x, xterm, xfeat, xe_char, xe_term], y, shuffle=False)
x_tn, y_tn, x_val, y_val = training_utils.split(x_tn, y_tn, shuffle=False)
print('define model')
model = HybridModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
name='hybridmodel_PE.h5',
PE=True) # +37
print('feat shape', xfeat.shape)
import sys
print('train')
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
def load_data_cache(name, sep_sent = True , sample = False):
if sep_sent:
filename = './data/'+name+'_sent.pkl'
else:
filename = './data/'+name+'_word.pkl'
r = data_utils.pickle_load( filename )
x_tn = r['x_tn']
y_tn = r['y_tn']
x_ts = r['x_ts']
y_ts = r['y_ts']
embedding_matrix = r['embed_mat']
y_tn = to_categorical(y_tn)
y_ts = to_categorical(y_ts)
indices = np.arange(y_tn.shape[0])
np.random.shuffle(indices)
x_tn = x_tn[indices]
y_tn = y_tn[indices]
if sample:
sample_num = min( 500000, y_tn.shape[0] )
# indices = np.arange( y_train.shape[0] )
# np.random.shuffle( indices )
# indices = indices[0:sample_num]
x_tn = x_tn[0:sample_num]
y_tn = y_tn[0:sample_num]
sample_num = min(100000, y_ts.shape[0] )
indices = np.arange( y_ts.shape[0] )
np.random.shuffle( indices )
indices = indices[0:sample_num ]
y_ts = y_ts[indices]
x_ts = x_ts[indices]
return x_tn, y_tn, x_ts, y_ts, embedding_matrix
def train_model_pe(use_char=False):
print('load data')
import data_utils, training_utils
conf = data_utils.TrainConfigure()
data_dict = data_utils.pickle_load(conf.char_file)
print('loading embed ...')
if use_char:
vocab_dict = data_utils.pickle_load(conf.char_dict)
embed_matrix = data_utils.load_embedding(
vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/char_embed.pkl')
MAX_LEN = 600
name = 'gateddeepcnnmodel_PE.h5'
x = data_dict['x']
else:
x = data_utils.pickle_load(conf.term_file)
vocab_dict = data_utils.pickle_load(conf.term_dict)
embed_matrix = data_utils.load_embedding(
vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/term_embed.pkl')
MAX_LEN = 300
name = 'gateddeepcnnmodel_term_PE.h5'
print('load embed done.')
y = to_categorical(data_dict['y'])
xe = [[i for i in range(MAX_LEN)] for _ in range(y.shape[0])]
xe = np.array(xe)
x_tn, y_tn, x_ts, y_ts = training_utils.split([x, xe], y, shuffle=False)
x_tn, y_tn, x_val, y_val = training_utils.split(x_tn, y_tn, shuffle=False)
print('train')
print('define model')
model = GatedDeepCNNModel(embed_matrix=embed_matrix,
MAX_LEN=MAX_LEN,
name=name,
PE=True)
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
del model
model = GatedDeepCNNModel(embed_matrix=embed_matrix,
MAX_LEN=MAX_LEN,
name=name,
PE=True,
train_embed=True)
model.load_weights()
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
def prepare_tn_data(filename='data/topic_train.txt'):
tn_conf = TrainConfigure()
data_dict = data_utils.pickle_load(tn_conf.char_file)
xterm = data_utils.pickle_load(tn_conf.term_file)
term_vocab_dict = data_utils.pickle_load(tn_conf.term_dict)
reverse_dict = dict()
for k, v in term_vocab_dict.items():
reverse_dict[v] = k
N = xterm.shape[0]
with open(filename, 'w') as fout:
for i in range(N):
xi = xterm[i]
term_list = []
for idx in xi:
if idx != 0 and idx != 1 and idx in reverse_dict:
term_list.append(reverse_dict[idx])
fout.write(' '.join(term_list) + '\n')
print('prepare data done.')
def train_model():
print('load data')
import data_utils, training_utils
conf = data_utils.TrainConfigure()
data_dict = data_utils.pickle_load(conf.char_file)
y = to_categorical(data_dict['y'])
x = data_dict['x']
xterm = data_utils.pickle_load(conf.term_file)
print('loading embed ...')
vocab_dict = data_utils.pickle_load(conf.term_dict)
embed_matrix = data_utils.load_embedding(
vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/term_embed.pkl')
print('load embed done.')
x_tn, y_tn, x_ts, y_ts = training_utils.split(xterm, y, shuffle=True)
x_tn, y_tn, x_val, y_val = training_utils.split(x_tn, y_tn, shuffle=True)
print('train')
print('define model')
model = TermModel(embed_matrix=embed_matrix)
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
def train_model_n200(model_conf, name, ModelClass):
print(name)
import data_utils200 as data_utils
conf = data_utils.TrainConfigure()
data_dict = data_utils.pickle_load(conf.char_file)
print('loading embed ...')
vocab_dict = data_utils.pickle_load(conf.char_dict)
char_embed_matrix = data_utils.load_embedding(
vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/char_embed.pkl')
MAX_LEN = conf.MAX_LEN
x = data_dict['x']
xterm = data_utils.pickle_load(conf.term_file)
term_vocab_dict = data_utils.pickle_load(conf.term_dict)
term_embed_matrix = data_utils.load_embedding(
term_vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/term_embed.pkl')
MAX_LEN_TERM = conf.MAX_LEN
print('load embed done.')
y = to_categorical(data_dict['y'])
xt = data_utils.pickle_load('data/lda_vec.pkl')
xfeat = data_utils.pickle_load(conf.feat_file)
# normalization
from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
scaler.fit(xfeat)
data_utils.pickle_dump(scaler, conf.feat_norm)
xfeat = scaler.transform(xfeat)
xe = [[i for i in range(MAX_LEN)] for _ in range(y.shape[0])]
xe = np.array(xe)
xe_term = [[i for i in range(MAX_LEN_TERM)] for _ in range(y.shape[0])]
xe_term = np.array(xe_term)
x_tn, y_tn, x_ts, y_ts = training_utils.split(
[x, xe, xterm, xe_term, xfeat, xt], y, shuffle=False)
x_tn, y_tn, x_val, y_val = training_utils.split(x_tn, y_tn, shuffle=False)
print('train')
print('define model')
model = ModelClass(model_conf,
char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
name=name)
x_tn, y_tn = model.gen_train(x_tn, y_tn)
x_val, y_val = model.gen_train(x_val, y_val)
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
del model
model = ModelClass(model_conf,
char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
name=name,
train_embed=True,
train_top=False)
model.load_weights()
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
def train_model(use_char=True):
print('train condition conv model.\nload data')
import data_utils, training_utils
conf = data_utils.TrainConfigure()
data_dict = data_utils.pickle_load(conf.char_file)
print('loading embed ...')
vocab_dict = data_utils.pickle_load(conf.char_dict)
char_embed_matrix = data_utils.load_embedding(
vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/char_embed.pkl')
MAX_LEN = 600
x = data_dict['x']
xterm = data_utils.pickle_load(conf.term_file)
term_vocab_dict = data_utils.pickle_load(conf.term_dict)
term_embed_matrix = data_utils.load_embedding(
term_vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/term_embed.pkl')
MAX_LEN_TERM = 300
name = 'conditionconvmodel.h5'
print('load embed done.')
y = to_categorical(data_dict['y'])
xt = data_utils.pickle_load('data/lda_vec.pkl')
xfeat = data_utils.pickle_load(conf.feat_file)
# normalization
from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
scaler.fit(xfeat)
data_utils.pickle_dump(scaler, conf.feat_norm)
xfeat = scaler.transform(xfeat)
x_tn, y_tn, x_ts, y_ts = training_utils.split([x, xterm, xfeat, xt],
y,
shuffle=False)
x_tn, y_tn, x_val, y_val = training_utils.split(x_tn, y_tn, shuffle=False)
print('train')
print('define model')
model = ConditionConvModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
MAX_LEN=MAX_LEN,
NUM_FEAT=8,
name=name)
x_tn, y_tn = model.gen_train(x_tn, y_tn)
x_val, y_val = model.gen_train(x_val, y_val)
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
del model
model = ConditionConvModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
MAX_LEN=MAX_LEN,
NUM_FEAT=8,
name=name,
train_embed=True,
train_top=False,
lr=0.001)
model.load_weights()
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
def get_vec(out_file, mode="tn"):
index = 0
vec_len = 20
dictionary = corpora.Dictionary.load('./data/lda.dict')
lda = gensim.models.LdaModel.load('data/LDA20.model')
if mode == "train":
print('train')
tn_conf = TrainConfigure()
elif mode == "val":
print('val')
tn_conf = data_utils.ValidConfigure()
else:
print("test")
tn_conf = data_utils.TestConfigure()
# data_dict = data_utils.pickle_load(tn_conf.char_file)
xterm = data_utils.pickle_load(tn_conf.term_file)
term_vocab_dict = data_utils.pickle_load(tn_conf.term_dict)
reverse_dict = dict()
for k, v in term_vocab_dict.items():
reverse_dict[v] = k
all_lda = []
for xi in xterm:
doc = []
for idx in xi:
if idx != 0 and idx != 1 and idx in reverse_dict:
doc.append(reverse_dict[idx])
doc_bow = dictionary.doc2bow(doc)
lda_vec_tmp = lda[doc_bow]
lda_vec = np.zeros(vec_len)
for (index, p) in lda_vec_tmp:
lda_vec[index] = p
index += 1
all_lda.append(lda_vec)
data_utils.pickle_dump(np.array(all_lda), out_file)
print('done.')
def train_model_cv( cv_index, cv_num ):
print('load data')
import data_utils, training_utils
conf = data_utils.TrainConfigure()
data_dict = data_utils.pickle_load(conf.char_file)
print('loading embed ...')
vocab_dict = data_utils.pickle_load(conf.char_dict)
char_embed_matrix = data_utils.load_embedding(vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/char_embed.pkl')
MAX_LEN = 600
x = data_dict['x']
xterm = data_utils.pickle_load(conf.term_file)
term_vocab_dict = data_utils.pickle_load(conf.term_dict)
term_embed_matrix = data_utils.load_embedding(term_vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/term_embed.pkl')
MAX_LEN_TERM = 300
name = 'hybriddensemodel_cv{}.h5'.format(cv_index)
print('load embed done.')
y = to_categorical(data_dict['y'])
xt = data_utils.pickle_load('data/lda_vec.pkl')
xfeat = data_utils.pickle_load(conf.feat_file)
# normalization
from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
scaler.fit(xfeat)
data_utils.pickle_dump(scaler, conf.feat_norm)
xfeat = scaler.transform(xfeat)
xe = [[i for i in range(MAX_LEN)] for _ in range(y.shape[0])]
xe = np.array(xe)
xe_term = [[i for i in range(MAX_LEN_TERM)] for _ in range(y.shape[0])]
xe_term = np.array(xe_term)
x_tn, y_tn, x_ts, y_ts = training_utils.split_cv([x, xe, xterm, xe_term, xfeat, xt], y, cv_index=cv_index,cv_num=cv_num)
x_tn, y_tn, x_val, y_val = training_utils.split(x_tn, y_tn, shuffle=False)
print('train')
print('define model')
model = HybridDenseModel(char_embed_matrix=char_embed_matrix, term_embed_matrix=term_embed_matrix,
MAX_LEN=MAX_LEN, NUM_FEAT=8, PE=True, name=name)
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
del model
model = HybridDenseModel(char_embed_matrix=char_embed_matrix, term_embed_matrix=term_embed_matrix,
MAX_LEN=MAX_LEN, NUM_FEAT=8, PE=True,
name=name, train_embed=True, train_top=False, lr=0.001)
model.load_weights()
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
del model
self.model.fit(x,
y,
validation_data=[x_val, y_val],
batch_size=512,
epochs=20,
callbacks=[early_stop, save_best])
metric = self.model.evaluate(x_ts, y_ts)
print(metric)
self.model.load_weights(save_path)
metric = self.model.evaluate(x_ts, y_ts, batch_size=512)
print(metric)
y_pred = self.model.predict(x_ts, batch_size=512)
cnf_matrix = confusion_matrix(convert_y(y_ts), convert_y(y_pred))
print(cnf_matrix)
if __name__ == '__main__':
import sys
print('define char model')
model = HanModel(name='hanmodel.h5')
print('load data')
import data_utils, training_utils
data_dict = data_utils.pickle_load('data_sent.dict')
y = to_categorical(data_dict['y'])
x_tn, y_tn, x_ts, y_ts = training_utils.split(data_dict['x'],
y,
shuffle=False)
x_tn, y_tn, x_val, y_val = training_utils.split(x_tn, y_tn, shuffle=False)
print('train')
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
def predict():
"""
根据概率集成
:return:
"""
print('load data')
tn_conf = TrainConfigure()
data_dict = data_utils.pickle_load(tn_conf.char_file)
y = to_categorical(data_dict['y'])
x = data_dict['x']
xterm = data_utils.pickle_load(tn_conf.term_file)
xfeat = data_utils.pickle_load(tn_conf.feat_file)
# normalization
from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
scaler.fit(xfeat)
data_utils.pickle_dump(scaler, tn_conf.feat_norm)
xfeat = scaler.transform(xfeat)
print('loading embed ...')
term_vocab_dict = data_utils.pickle_load(tn_conf.term_dict)
term_embed_matrix = data_utils.load_embedding(
term_vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/term_embed.pkl')
# term_embed_matrix = data_utils.load_embedding(term_vocab_dict,
# 'data/sgns.target.word-word.dynwin5.thr10.neg5.dim300.iter5',
# dump_path='data/term_embed_ww.pkl')
char_vocab_dict = data_utils.pickle_load(tn_conf.char_dict)
char_embed_matrix = data_utils.load_embedding(
char_vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/char_embed.pkl')
print('load embed done.')
val_conf = ValidConfigure()
data_dict = data_utils.pickle_load(val_conf.char_file)
y = to_categorical(data_dict['y'])
x = data_dict['x']
ids = data_dict['id']
xterm = data_utils.pickle_load(val_conf.term_file)
xfeat = data_utils.pickle_load(val_conf.feat_file)
xfeat = scaler.transform(xfeat)
print('feat shape', xfeat.shape)
xtopic = data_utils.pickle_load('data/lda_vec_val.pkl')
xe = [[i for i in range(600)] for _ in range(y.shape[0])]
xe = np.array(xe)
xe_term = [[i for i in range(300)] for _ in range(y.shape[0])]
xe_term = np.array(xe_term)
import data_utils100
val_conf100 = data_utils100.ValidConfigure()
data_dict100 = data_utils.pickle_load(val_conf100.char_file)
x100 = data_dict100['x']
xterm100 = data_utils.pickle_load(val_conf100.term_file)
xe100 = [[i for i in range(100)] for _ in range(y.shape[0])]
xe100 = np.array(xe100)
xe_term100 = [[i for i in range(100)] for _ in range(y.shape[0])]
xe_term100 = np.array(xe_term100)
import data_utils200
val_conf200 = data_utils200.ValidConfigure()
data_dict200 = data_utils.pickle_load(val_conf200.char_file)
x200 = data_dict200['x']
xterm200 = data_utils.pickle_load(val_conf200.term_file)
xe200 = [[i for i in range(200)] for _ in range(y.shape[0])]
xe200 = np.array(xe200)
xe_term200 = [[i for i in range(200)] for _ in range(y.shape[0])]
xe_term200 = np.array(xe_term200)
ys = []
print('define model')
model = HybridDenseModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='hybriddensemodel_PE.h5')
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
model = HybridDenseMAModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='hybriddensemodelma_PE.h5')
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('dense model done.')
model = HybridSEModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='hybridsemodel_PE.h5')
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('se model done.')
# print('start len 100 model')
# model = HybridConvModel(char_embed_matrix=char_embed_matrix,
# term_embed_matrix=term_embed_matrix, MAX_LEN=100, MAX_LEN_TERM=100,NUM_FEAT=8,
# PE=True, name='hybridconvmodel_n100.h5')
# model.load_weights()
# y = model.predict([x100, xe100, xterm100, xe_term100, xfeat, xtopic])
# ys.append(y)
# del model
# print('hybrid conv model done.')
#
# model = HybridGatedDeepCNNModel(char_embed_matrix=char_embed_matrix,
# term_embed_matrix=term_embed_matrix, MAX_LEN=100, MAX_LEN_TERM=100,NUM_FEAT=8,
# PE=True, name='hybridgateddeepcnnmodel_n100.h5')
# model.load_weights()
# y = model.predict([x100, xe100, xterm100, xe_term100, xfeat, xtopic])
# ys.append(y)
# del model
# print('hybrid gated deep cnn model done.')
#
# model = HybridRCNNModel(char_embed_matrix=char_embed_matrix,
# term_embed_matrix=term_embed_matrix, MAX_LEN=100, MAX_LEN_TERM=100,NUM_FEAT=8,
# PE=True, name='hybridrcnnmodel_n100.h5')
# model.load_weights()
# y = model.predict([x100, xe100, xterm100, xe_term100, xfeat, xtopic])
# ys.append(y)
# del model
# print('hybrid RCNN model done.')
print('start len 200 model')
model = HybridConvModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
MAX_LEN=200,
MAX_LEN_TERM=200,
NUM_FEAT=8,
PE=True,
name='hybridconvmodel_n200.h5')
model.load_weights()
y = model.predict([x200, xe200, xterm200, xe_term200, xfeat, xtopic])
ys.append(y)
del model
print('hybrid conv model done.')
model = HybridDPCNNModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
MAX_LEN=200,
MAX_LEN_TERM=200,
NUM_FEAT=8,
PE=True,
name='hybriddpcnnmodel_n200.h5')
model.load_weights()
y = model.predict([x200, xe200, xterm200, xe_term200, xfeat, xtopic])
ys.append(y)
del model
print('hybrid dpcnn model done.')
model = HybridGatedConvTopicModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
MAX_LEN=200,
MAX_LEN_TERM=200,
NUM_FEAT=8,
PE=True,
name='hybridgatedconvtopicmodel_n200.h5')
model.load_weights()
y = model.predict([x200, xe200, xterm200, xe_term200, xfeat, xtopic])
ys.append(y)
del model
print('hybrid dpcnn model done.')
model = HybridGatedDeepCNNModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
MAX_LEN=200,
MAX_LEN_TERM=200,
NUM_FEAT=8,
PE=True,
name='hybridgateddeepcnnmodel_n200.h5')
model.load_weights()
y = model.predict([x200, xe200, xterm200, xe_term200, xfeat, xtopic])
ys.append(y)
del model
print('hybrid gated deep cnn model done.')
model = HybridRCNNModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
MAX_LEN=200,
MAX_LEN_TERM=200,
NUM_FEAT=8,
PE=True,
name='hybridrcnnmodel_n200.h5')
model.load_weights()
y = model.predict([x200, xe200, xterm200, xe_term200, xfeat, xtopic])
ys.append(y)
del model
#这个模型太慢
model = ConditionAttModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='conditionattmodel_PE.h5',
lr=0.001)
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
print('condition att model done.')
model = ConditionConvModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='conditionconvmodel_PE.h5',
lr=0.001)
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('condition conv model done.')
model = ConditionDPCNNModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='conditiondpcnnmodel_PE.h5',
lr=0.001)
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('condition dpcnn model done.')
model = ConditionGatedConvModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='conditiongatedconvmodel_PE.h5',
lr=0.001)
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('condition gated conv model done.')
model = ConditionGatedDeepCNNModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='conditiongateddeepcnnmodel_PE.h5',
lr=0.001)
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('condition gated deepcnn model done.')
model = HybridAttModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='hybridattmodel_PE.h5')
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
print('hybrid att model done.')
model = HybridConvModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='hybridconvmodel_PE.h5')
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('hybrid conv model done.')
model = HybridDPCNNModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='hybriddpcnnmodel_PE.h5')
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('hybrid dpcnn model done.')
model = HybridGatedDeepCNNModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='hybridgateddeepcnnmodel_PE.h5')
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('hybrid gated deep cnn model done.')
model = HybridRCNNModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='hybridrcnnmodel_PE.h5')
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('hybrid rcnn model done.')
model = HybridGatedConvTopicModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='hybridgatedconvtopicmodel_PE.h5')
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
print('hybrid gated conv topic done.')
y = fasttextmodel.predict_char()
ys.append(y)
y = fasttextmodel.predict_term()
ys.append(y)
print(y.shape)
print('fast text done.')
#hybrid model
# model = HybridModel(char_embed_matrix=char_embed_matrix, term_embed_matrix=term_embed_matrix, NUM_FEAT=8)# + 37
# model.load_weights()
# y = model.predict([x, xterm, xfeat])
# ys.append( y )
# print(y.shape)
# print('hybrid model done.')
labels = ['人类作者', '自动摘要', '机器作者', '机器翻译']
y_pred = np.mean(ys, axis=0)
y_pred = convert_y(y_pred)
out_file = 'result.csv'
with open(out_file, 'w', encoding='utf-8') as fout:
for id, yi in zip(ids, y_pred):
label = labels[yi]
fout.write('{},{}\n'.format(id, label))
print('done.')
def predict2():
"""
根据分类结果集成
:return:
"""
print('load data')
tn_conf = TrainConfigure()
data_dict = data_utils.pickle_load(tn_conf.char_file)
y = to_categorical(data_dict['y'])
x = data_dict['x']
xterm = data_utils.pickle_load(tn_conf.term_file)
xfeat = data_utils.pickle_load(tn_conf.feat_file)
# normalization
from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
scaler.fit(xfeat)
data_utils.pickle_dump(scaler, tn_conf.feat_norm)
xfeat = scaler.transform(xfeat)
print('loading embed ...')
term_vocab_dict = data_utils.pickle_load(tn_conf.term_dict)
term_embed_matrix = data_utils.load_embedding(
term_vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/term_embed.pkl')
# term_embed_matrix = data_utils.load_embedding(term_vocab_dict,
# 'data/sgns.target.word-word.dynwin5.thr10.neg5.dim300.iter5',
# dump_path='data/term_embed_ww.pkl')
char_vocab_dict = data_utils.pickle_load(tn_conf.char_dict)
char_embed_matrix = data_utils.load_embedding(
char_vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/char_embed.pkl')
print('load embed done.')
val_conf = ValidConfigure()
data_dict = data_utils.pickle_load(val_conf.char_file)
y = to_categorical(data_dict['y'])
x = data_dict['x']
ids = data_dict['id']
xterm = data_utils.pickle_load(val_conf.term_file)
xfeat = data_utils.pickle_load(val_conf.feat_file)
xfeat = scaler.transform(xfeat)
xe = [[i for i in range(600)] for _ in range(y.shape[0])]
xe = np.array(xe)
ys = []
print('define model')
#hybrid model
model = HybridModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8) # + 37
print('feat shape', xfeat.shape)
model.load_weights()
y = model.predict([x, xterm, xfeat])
ys.append(convert_onehot(y))
print('hybrid model done.')
#CNN model (char)
model = CharModel(embed_matrix=char_embed_matrix)
model.load_weights()
y = model.predict(x)
ys.append(convert_onehot(y))
print('char model done.')
model = CharModel(embed_matrix=char_embed_matrix,
name='charmodel_PE.h5',
PE=True)
model.load_weights()
y = model.predict([x, xe])
ys.append(convert_onehot(y))
print('char model done.')
model = CharModel(embed_matrix=char_embed_matrix,
name='charmodel_PE_OE.h5',
PE=True)
model.load_weights()
y = model.predict([x, xe])
ys.append(convert_onehot(y))
print('char model done.')
#CNN (term)
model = TermModel(embed_matrix=term_embed_matrix)
model.load_weights()
y = model.predict(xterm)
ys.append(convert_onehot(y))
print('term model done.')
model = DeepCNNModel(embed_matrix=char_embed_matrix)
model.load_weights()
y = model.predict(x)
ys.append(convert_onehot(y))
print('deep cnn done.')
# # attention model (char)
# model = AttModel(MAX_LEN=600, name='charattmodel.h5', embed_matrix=char_embed_matrix)
# model.load_weights()
# y = model.predict(x)
# ys.append(convert_onehot(y))
# # attention model (term)
# model = AttModel(MAX_LEN=300, embed_matrix=term_embed_matrix)
# model.load_weights()
# y = model.predict(xterm)
# ys.append(convert_onehot(y))
#
# model = ConditionModel(embed_matrix=char_embed_matrix)
# model.load_weights()
# y = model.predict(x)
# ys.append(convert_onehot(y))
model = SSCharModel(embed_matrix=char_embed_matrix,
name='sscharmodel_PE.h5',
PE=True,
train_embed=True)
model.load_weights()
y = model.predict([x, xe])
ys.append(convert_onehot(y))
model = SSCharModel(embed_matrix=char_embed_matrix, train_embed=True)
model.load_weights()
y = model.predict(x)
ys.append(convert_onehot(y))
model = GatedConvModel(embed_matrix=char_embed_matrix,
name='gatedconvmodel_PE.h5',
PE=True)
model.load_weights()
y = model.predict([x, xe])
ys.append(convert_onehot(y))
model = GatedConvModel(embed_matrix=char_embed_matrix, train_embed=True)
model.load_weights()
y = model.predict(x)
ys.append(convert_onehot(y))
model = GatedDeepCNNModel(embed_matrix=char_embed_matrix,
name='gateddeepcnnmodel_PE.h5',
PE=True,
train_embed=True)
model.load_weights()
y = model.predict([x, xe])
ys.append(convert_onehot(y))
model = GatedDeepCNNModel(embed_matrix=char_embed_matrix, train_embed=True)
model.load_weights()
y = model.predict(x)
ys.append(convert_onehot(y))
labels = ['人类作者', '自动摘要', '机器作者', '机器翻译']
y_pred = np.mean(ys, axis=0)
y_pred = convert_y(y_pred)
out_file = 'result.csv'
with open(out_file, 'w', encoding='utf-8') as fout:
for id, yi in zip(ids, y_pred):
label = labels[yi]
fout.write('{},{}\n'.format(id, label))
print('done.')
def predict(self, x):
y_pred = self.model.predict(x, batch_size=512)
return y_pred
if __name__ == '__main__':
import sys
tn_conf = TrainConfigure()
if len(sys.argv) > 1 and sys.argv[1] == 'char':
if len(sys.argv) > 2 and sys.argv[2] == 'pe':
print('define char model with position embedding')
print('load data')
import data_utils, training_utils
data_dict = data_utils.pickle_load(tn_conf.char_file)
y = to_categorical(data_dict['y'])
char_vocab_dict = data_utils.pickle_load(tn_conf.char_dict)
char_embed_matrix = data_utils.load_embedding(
char_vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/char_embed.pkl')
xe = [[i for i in range(600)] for _ in range(y.shape[0])]
xe = np.array(xe)
x_tn, y_tn, x_ts, y_ts = training_utils.split([data_dict['x'], xe],
y,
shuffle=False)
x_tn, y_tn, x_val, y_val = training_utils.split(x_tn,
y_tn,
shuffle=False)
print('train')
def train_model_peoe( ):
print('load data')
import data_utils, training_utils
conf = data_utils.TrainConfigure()
data_dict = data_utils.pickle_load(conf.char_file)
print('loading embed ...')
vocab_dict = data_utils.pickle_load(conf.char_dict)
char_embed_matrix = data_utils.load_embedding(vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/char_embed.pkl')
MAX_LEN = 600
x = data_dict['x']
xterm = data_utils.pickle_load(conf.term_file)
term_vocab_dict = data_utils.pickle_load(conf.term_dict)
term_embed_matrix = data_utils.load_embedding(term_vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/term_embed.pkl')
char_embeds = []
char_embed_matrix_oe = data_utils.load_our_embedding(vocab_dict)
char_embeds.append( char_embed_matrix_oe )
for windows in [3, 5, 8]:
sg = 1
# for sg in [0,1]:
embed_file = 'data/char_embed_{}_{}.model'.format(windows, sg)
char_embed_tmp = data_utils.load_our_embedding(vocab_dict, model_file = embed_file,
dump_path = 'data/our_char_embed_{}_{}.pkl'.format(windows, sg))
char_embeds.append(char_embed_tmp)
term_embeds = []
term_embed_matrix_oe = data_utils.load_our_embedding(term_vocab_dict, model_file='data/term_embed.model',
dump_path='data/our_term_embed.pkl')
term_embeds.append( term_embed_matrix_oe )
for windows in [3, 5, 8]:
sg = 1
# for sg in [0,1]:
embed_file = 'data/term_embed_{}_{}.model'.format(windows, sg)
term_embed_tmp = data_utils.load_our_embedding(term_vocab_dict, model_file=embed_file,
dump_path='data/our_term_embed_{}_{}.pkl'.format(windows,
sg))
term_embeds.append(term_embed_tmp)
MAX_LEN_TERM = 300
name = 'hybriddpcnnmodel_PEOE.h5'
print('load embed done.')
y = to_categorical(data_dict['y'])
xt = data_utils.pickle_load('data/lda_vec.pkl')
xfeat = data_utils.pickle_load(conf.feat_file)
# normalization
from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
scaler.fit(xfeat)
data_utils.pickle_dump(scaler, conf.feat_norm)
xfeat = scaler.transform(xfeat)
xe = [[i for i in range(MAX_LEN)] for _ in range(y.shape[0])]
xe = np.array(xe)
xe_term = [[i for i in range(MAX_LEN_TERM)] for _ in range(y.shape[0])]
xe_term = np.array(xe_term)
x_tn, y_tn, x_ts, y_ts = training_utils.split([x, xe, xterm, xe_term, xfeat, xt], y, shuffle=False)
x_tn, y_tn, x_val, y_val = training_utils.split(x_tn, y_tn, shuffle=False)
print('train')
print('define model')
#加入更多embedding模型以后，学习率要降低才能正常学习
model = HybridDPCNNModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix, MAX_LEN=MAX_LEN, NUM_FEAT=8, PE=True,
name=name, char_embeds=char_embeds, term_embeds = term_embeds, lr=0.0004)
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
del model
model = HybridDPCNNModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix, MAX_LEN=MAX_LEN, NUM_FEAT=8, PE=True,
name=name, train_embed=True, train_top=False, lr=0.001, char_embeds=char_embeds, term_embeds = term_embeds)
model.load_weights()
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
def predict(tn_conf, lda_file, val_conf, val_conf100, val_conf200):
"""
根据概率集成
:return:
"""
print('load data')
data_dict = data_utils.pickle_load(tn_conf.char_file)
y = to_categorical(data_dict['y'])
x = data_dict['x']
xterm = data_utils.pickle_load(tn_conf.term_file)
xfeat = data_utils.pickle_load(tn_conf.feat_file)
# normalization
from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
scaler.fit(xfeat)
data_utils.pickle_dump(scaler, tn_conf.feat_norm)
xfeat = scaler.transform(xfeat)
xe = [[i for i in range(600)] for _ in range(y.shape[0])]
xe = np.array(xe)
xe_term = [[i for i in range(300)] for _ in range(y.shape[0])]
xe_term = np.array(xe_term)
xtopic = data_utils.pickle_load(lda_file)
print('loading embed ...')
term_vocab_dict = data_utils.pickle_load(tn_conf.term_dict)
term_embed_matrix = data_utils.load_embedding(
term_vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/term_embed.pkl')
# term_embed_matrix = data_utils.load_embedding(term_vocab_dict,
# 'data/sgns.target.word-word.dynwin5.thr10.neg5.dim300.iter5',
# dump_path='data/term_embed_ww.pkl')
char_vocab_dict = data_utils.pickle_load(tn_conf.char_dict)
char_embed_matrix = data_utils.load_embedding(
char_vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/char_embed.pkl')
print('load embed done.')
data_dict = data_utils.pickle_load(val_conf.char_file)
y = to_categorical(data_dict['y'])
x = data_dict['x']
ids = data_dict['id']
xterm = data_utils.pickle_load(val_conf.term_file)
xfeat = data_utils.pickle_load(val_conf.feat_file)
xfeat = scaler.transform(xfeat)
print('feat shape', xfeat.shape)
data_dict100 = data_utils.pickle_load(val_conf100.char_file)
x100 = data_dict100['x']
xterm100 = data_utils.pickle_load(val_conf100.term_file)
xe100 = [[i for i in range(100)] for _ in range(y.shape[0])]
xe100 = np.array(xe100)
xe_term100 = [[i for i in range(100)] for _ in range(y.shape[0])]
xe_term100 = np.array(xe_term100)
data_dict200 = data_utils.pickle_load(val_conf200.char_file)
x200 = data_dict200['x']
xterm200 = data_utils.pickle_load(val_conf200.term_file)
xe200 = [[i for i in range(200)] for _ in range(y.shape[0])]
xe200 = np.array(xe200)
xe_term200 = [[i for i in range(200)] for _ in range(y.shape[0])]
xe_term200 = np.array(xe_term200)
ys = []
print('define model')
model = HybridDenseModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='hybriddensemodel_PE.h5')
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
model = HybridDenseMAModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='hybriddensemodelma_PE.h5')
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('dense model done.')
model = HybridSEModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='hybridsemodel_PE.h5')
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('se model done.')
print('start len 100 model')
model = HybridConvModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
MAX_LEN=100,
MAX_LEN_TERM=100,
NUM_FEAT=8,
PE=True,
name='hybridconvmodel_n100.h5')
model.load_weights()
y = model.predict([x100, xe100, xterm100, xe_term100, xfeat, xtopic])
ys.append(y)
del model
print('hybrid conv model done.')
model = HybridGatedDeepCNNModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
MAX_LEN=100,
MAX_LEN_TERM=100,
NUM_FEAT=8,
PE=True,
name='hybridgateddeepcnnmodel_n100.h5')
model.load_weights()
y = model.predict([x100, xe100, xterm100, xe_term100, xfeat, xtopic])
ys.append(y)
del model
print('hybrid gated deep cnn model done.')
model = HybridRCNNModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
MAX_LEN=100,
MAX_LEN_TERM=100,
NUM_FEAT=8,
PE=True,
name='hybridrcnnmodel_n100.h5')
model.load_weights()
y = model.predict([x100, xe100, xterm100, xe_term100, xfeat, xtopic])
ys.append(y)
del model
print('hybrid RCNN model done.')
# print('start len 200 model')
# model = HybridConvModel(char_embed_matrix=char_embed_matrix,
# term_embed_matrix=term_embed_matrix, MAX_LEN=200, MAX_LEN_TERM=200,NUM_FEAT=8,
# PE=True, name='hybridconvmodel_n200.h5')
# model.load_weights()
# y = model.predict([x200, xe200, xterm200, xe_term200, xfeat, xtopic])
# ys.append(y)
# del model
# print('hybrid conv model done.')
#
# model = HybridDPCNNModel(char_embed_matrix=char_embed_matrix,
# term_embed_matrix=term_embed_matrix, MAX_LEN=200, MAX_LEN_TERM=200,NUM_FEAT=8,
# PE=True, name='hybriddpcnnmodel_n200.h5')
# model.load_weights()
# y = model.predict([x200, xe200, xterm200, xe_term200, xfeat, xtopic])
# ys.append(y)
# del model
# print('hybrid dpcnn model done.')
#
# model = HybridGatedDeepCNNModel(char_embed_matrix=char_embed_matrix,
# term_embed_matrix=term_embed_matrix, MAX_LEN=200, MAX_LEN_TERM=200,NUM_FEAT=8,
# PE=True, name='hybridgateddeepcnnmodel_n200.h5')
# model.load_weights()
# y = model.predict([x200, xe200, xterm200, xe_term200, xfeat, xtopic])
# ys.append(y)
# del model
# print('hybrid gated deep cnn model done.')
#
# model = HybridRCNNModel(char_embed_matrix=char_embed_matrix,
# term_embed_matrix=term_embed_matrix, MAX_LEN=200, MAX_LEN_TERM=200,NUM_FEAT=8,
# PE=True, name='hybridrcnnmodel_n200.h5')
# model.load_weights()
# y = model.predict([x200, xe200, xterm200, xe_term200, xfeat, xtopic])
# ys.append(y)
# del model
#这个模型太慢
# model = ConditionAttModel(char_embed_matrix=char_embed_matrix,
# term_embed_matrix=term_embed_matrix, NUM_FEAT=8, PE=True,
# name='conditionattmodel_PE.h5', lr=0.001)
# model.load_weights()
# y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
# ys.append(y)
# print('condition att model done.')
model = ConditionConvModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='conditionconvmodel_PE.h5',
lr=0.001)
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('condition conv model done.')
model = ConditionDPCNNModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='conditiondpcnnmodel_PE.h5',
lr=0.001)
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('condition dpcnn model done.')
model = ConditionGatedConvModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='conditiongatedconvmodel_PE.h5',
lr=0.001)
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('condition gated conv model done.')
model = ConditionGatedDeepCNNModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='conditiongateddeepcnnmodel_PE.h5',
lr=0.001)
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('condition gated deepcnn model done.')
model = HybridAttModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='hybridattmodel_PE.h5')
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
print('hybrid att model done.')
model = HybridConvModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='hybridconvmodel_PE.h5')
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('hybrid conv model done.')
model = HybridDPCNNModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='hybriddpcnnmodel_PE.h5')
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('hybrid dpcnn model done.')
model = HybridGatedDeepCNNModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='hybridgateddeepcnnmodel_PE.h5')
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('hybrid gated deep cnn model done.')
model = HybridRCNNModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='hybridrcnnmodel_PE.h5')
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
del model
print('hybrid rcnn model done.')
model = HybridGatedConvTopicModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8,
PE=True,
name='hybridgatedconvtopicmodel_PE.h5')
model.load_weights()
y = model.predict([x, xe, xterm, xe_term, xfeat, xtopic])
ys.append(y)
print('hybrid gated conv topic done.')
y = fasttextmodel.predict_char()
ys.append(y)
y = fasttextmodel.predict_term()
ys.append(y)
print(y.shape)
print('fast text done.')
#hybrid model
model = HybridModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8) # + 37
model.load_weights()
y = model.predict([x, xterm, xfeat])
ys.append(y)
print(y.shape)
print('hybrid model done.')
ys = np.array(ys)
print(ys.shape)
return ys
def load_grid(num_psf):
"""
Load grid
"""
log.info("Load Grid data")
return pickle_load('grid_{}.pickle.gz'.format(num_psf), compressed=True)
def stacking_main_condition():
print('load data')
tn_conf = TrainConfigure()
data_dict = data_utils.pickle_load(tn_conf.char_file)
y = to_categorical(data_dict['y'])
x = data_dict['x']
xterm = data_utils.pickle_load(tn_conf.term_file)
xfeat = data_utils.pickle_load(tn_conf.feat_file)
# normalization
from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
scaler.fit(xfeat)
data_utils.pickle_dump(scaler, tn_conf.feat_norm)
xfeat = scaler.transform(xfeat)
xe = [[i for i in range(600)] for _ in range(y.shape[0])]
xe = np.array(xe)
xe_term = [[i for i in range(300)] for _ in range(y.shape[0])]
xe_term = np.array(xe_term)
xtopic = data_utils.pickle_load('data/lda_vec.pkl')
print('loading embed ...')
term_vocab_dict = data_utils.pickle_load(tn_conf.term_dict)
term_embed_matrix = data_utils.load_embedding(term_vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/term_embed.pkl')
# term_embed_matrix = data_utils.load_embedding(term_vocab_dict,
# 'data/sgns.target.word-word.dynwin5.thr10.neg5.dim300.iter5',
# dump_path='data/term_embed_ww.pkl')
char_vocab_dict = data_utils.pickle_load(tn_conf.char_dict)
char_embed_matrix = data_utils.load_embedding(char_vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/char_embed.pkl')
print('load embed done.')
name = 'model/stack_condition_model.pkl'
model_dir = 'model/stack/'
n_fold = 3
name = 'model/stack_condition_model5.pkl'
model_dir = 'model/stack5/'
n_fold = 5
stk_model = stacking(n_fold, name=name, is_condition=True)
conf = conditionmodelbase.ModelConfigure()
conf.PE = True
stk_model.add_model(ConditionConvModel, {"conf":conf,"char_embed_matrix":char_embed_matrix,
"term_embed_matrix":term_embed_matrix,
"name":model_dir+'conditionconvmodel_PE.h5'})
stk_model.add_model(ConditionGatedConvModel, {"conf":conf,"char_embed_matrix": char_embed_matrix,
"term_embed_matrix": term_embed_matrix,
"name": model_dir+'conditiongatedconvmodel_PE.h5'})
stk_model.add_model(ConditionGatedDeepCNNModel, {"conf":conf,"char_embed_matrix": char_embed_matrix,
"term_embed_matrix": term_embed_matrix,
"name": model_dir+'conditiongateddeepcnnmodel_PE.h5'})
conf.lr = 0.0005
stk_model.add_model(ConditionDPCNNModel, {"conf": conf, "char_embed_matrix": char_embed_matrix,
"term_embed_matrix": term_embed_matrix,
"name": model_dir + 'conditiondpcnnmodel_PE.h5'})
#采样0.1用于测试
# x_tn, y_tn, x_ts, y_ts = training_utils.split([x, xe, xterm, xe_term, xfeat, xtopic], y, split_ratio=0.005, shuffle=False)
# x_tn, y_tn, x_ts, y_ts = training_utils.split(x_tn, y_tn, shuffle=False)
x_tn, y_tn, x_ts, y_ts = training_utils.split([x, xe, xterm, xe_term, xfeat, xtopic], y, split_ratio=0.95)
stk_model.fit(x_tn, y_tn)
# joblib.dump(stk_model, 'model/stack_model_3.pkl')
y_pred = stk_model.predict(x_ts)
acc = accuracy_score(training_utils.convert_y(y_pred), training_utils.convert_y(y_ts) )
print(acc)
cnf_matrix = confusion_matrix(training_utils.convert_y(y_pred), training_utils.convert_y(y_ts) )
print(cnf_matrix)
stk_model.save( )
def train_model_ftoe(model_conf,
model_name='hybridconvmodel_FTOE.h5',
ModelClass=HybridModelBase,
char_embed_file=None,
term_embed_file=None):
print(model_name)
print('load data')
import data_utils, training_utils
conf = data_utils.TrainConfigure()
data_dict = data_utils.pickle_load(conf.char_file)
print('loading embed ...')
vocab_dict = data_utils.pickle_load(conf.char_dict)
# char_embed_matrix = data_utils.load_embedding(vocab_dict,
# 'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
# dump_path='data/char_embed.pkl')
MAX_LEN = 600
x = data_dict['x']
xterm = data_utils.pickle_load(conf.term_file)
term_vocab_dict = data_utils.pickle_load(conf.term_dict)
# term_embed_matrix = data_utils.load_embedding(term_vocab_dict,
# 'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
# dump_path='data/term_embed.pkl')
char_embed_matrix = data_utils.load_embedding(
vocab_dict,
char_embed_file,
dump_path='data/{}.pkl'.format(char_embed_file[5:]))
term_embed_matrix = data_utils.load_embedding(
term_vocab_dict,
term_embed_file,
dump_path='data/{}.pkl'.format(term_embed_file[5:]))
MAX_LEN_TERM = 300
print('load embed done.')
y = to_categorical(data_dict['y'])
xt = data_utils.pickle_load('data/lda_vec.pkl')
xfeat = data_utils.pickle_load(conf.feat_file)
# normalization
from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
scaler.fit(xfeat)
data_utils.pickle_dump(scaler, conf.feat_norm)
xfeat = scaler.transform(xfeat)
xe = [[i for i in range(MAX_LEN)] for _ in range(y.shape[0])]
xe = np.array(xe)
xe_term = [[i for i in range(MAX_LEN_TERM)] for _ in range(y.shape[0])]
xe_term = np.array(xe_term)
x_tn, y_tn, x_ts, y_ts = training_utils.split(
[x, xe, xterm, xe_term, xfeat, xt], y, split_ratio=0.95, shuffle=False)
x_tn, y_tn, x_val, y_val = training_utils.split(x_tn,
y_tn,
split_ratio=0.95,
shuffle=False)
print('train')
print('define model')
model = ModelClass(model_conf,
char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
name=model_name,
train_embed=False,
train_top=True)
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
del model
model_conf.lr *= 0.5
model = ModelClass(model_conf,
char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
name=model_name,
train_embed=True,
train_top=False)
model.load_weights()
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
del model
def train_main():
print('load data')
import data_utils, training_utils
tn_conf = TrainConfigure()
data_dict = data_utils.pickle_load(tn_conf.char_file)
y = to_categorical(data_dict['y'])
x = data_dict['x']
xterm = data_utils.pickle_load(tn_conf.term_file)
xfeat = data_utils.pickle_load(tn_conf.feat_file)
# normalization
from sklearn.preprocessing import MinMaxScaler
scaler = MinMaxScaler()
scaler.fit(xfeat)
data_utils.pickle_dump(scaler, tn_conf.feat_norm)
xfeat = scaler.transform(xfeat)
print('loading embed ...')
term_vocab_dict = data_utils.pickle_load(tn_conf.term_dict)
term_embed_matrix = data_utils.load_embedding(
term_vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/term_embed.pkl')
# term_embed_matrix = data_utils.load_embedding(term_vocab_dict,
# 'data/sgns.target.word-word.dynwin5.thr10.neg5.dim300.iter5',
# dump_path='data/term_embed_ww.pkl')
char_vocab_dict = data_utils.pickle_load(tn_conf.char_dict)
char_embed_matrix = data_utils.load_embedding(
char_vocab_dict,
'data/sgns.target.word-character.char1-2.dynwin5.thr10.neg5.dim300.iter5',
dump_path='data/char_embed.pkl')
print('load embed done.')
x_tn, y_tn, x_ts, y_ts = training_utils.split([x, xterm, xfeat],
y,
shuffle=False)
x_tn, y_tn, x_val, y_val = training_utils.split(x_tn, y_tn, shuffle=False)
print('define model')
model = HybridModel(char_embed_matrix=char_embed_matrix,
term_embed_matrix=term_embed_matrix,
NUM_FEAT=8) # +37
print('feat shape', xfeat.shape)
import sys
if len(sys.argv) <= 1 or sys.argv[1] == 'train':
print('train')
model.train(x_tn, y_tn, x_val, y_val, x_ts, y_ts)
if len(sys.argv) > 1 and sys.argv[1] == 'val':
val_conf = ValidConfigure()
data_dict = data_utils.pickle_load(val_conf.char_file)
y = to_categorical(data_dict['y'])
x = data_dict['x']
ids = data_dict['id']
xterm = data_utils.pickle_load(val_conf.term_file)
xfeat = data_utils.pickle_load(val_conf.feat_file)
xfeat = scaler.transform(xfeat)
model.load_weights()
model.test([x, xterm, xfeat], ids, val_conf.out_file)
if len(sys.argv) > 1 and sys.argv[1] == 'error':
start_index = y_tn.shape[0] + y_val.shape[0]
texts = data_utils.load_all_text(tn_conf)
model.load_weights()
model.error_analysis(x_ts, y_ts, texts, start_index)
