kernel_id int64 24.2k 23.3M | prompt stringlengths 8 1.85M | completetion stringlengths 1 182k | comp_name stringlengths 5 57 |
|---|---|---|---|
943,976 | seed_everything()
glove_embeddings = load_glove(word_index)
paragram_embeddings = load_para(word_index)
fasttext_embeddings = load_fasttext(word_index)
embedding_matrix = np.mean([glove_embeddings, paragram_embeddings, fasttext_embeddings], axis=0)
del glove_embeddings, paragram_embeddings, fasttext_embeddings
gc.collect()
np.shape(embedding_matrix )<split> | X_train = df_train[CATEGORY_COLUMNS].fillna(-1000)
y_train = df_train["Survived"]
X_test = df_test[CATEGORY_COLUMNS].fillna(-1000)
randomf=RandomForestClassifier(criterion='gini', n_estimators=700, min_samples_split=10,min_samples_leaf=1,max_features='auto',oob_score=True,random_state=1,n_jobs=-1)
randomf.fit(X_train, y_train)
Submission['Survived']=randomf.predict(X_test)
acc_gd_cv = round(accuracy_score(y_pred, y_val)* 100, 2)
print('Accuracy')
print(acc_gd_cv)
print(Submission.head(10))
Submission.to_csv('finalrandomforest01.csv',sep=',')
print('Random Forest prediction created' ) | Titanic - Machine Learning from Disaster |
943,976 | splits = list(StratifiedKFold(n_splits=n_splits, shuffle=True, random_state=SEED ).split(x_train, y_train))
splits[:3]<choose_model_class> | MLA = [
ensemble.ExtraTreesClassifier() ,
ensemble.GradientBoostingClassifier() ,
ensemble.RandomForestClassifier() ,
linear_model.LogisticRegressionCV() ,
neighbors.KNeighborsClassifier() ,
svm.SVC(probability=True),
]
index = 1
for alg in MLA:
predicted = alg.fit(X_train, y_train ).predict(X_test)
fp, tp, th = roc_curve(y_test, predicted)
roc_auc_mla = auc(fp, tp)
MLA_name = alg.__class__.__name__
plt.plot(fp, tp, lw=2, alpha=0.3, label='ROC %s(AUC = %0.2f)' %(MLA_name, roc_auc_mla))
index+=1
plt.title('ROC Curve comparison')
plt.legend(bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0.)
plt.plot([0,1],[0,1],'r--')
plt.xlim([0,1])
plt.ylim([0,1])
plt.ylabel('True Positive Rate')
plt.xlabel('False Positive Rate')
plt.show() | Titanic - Machine Learning from Disaster |
943,976 | class CyclicLR(object):
def __init__(self, optimizer, base_lr=1e-3, max_lr=6e-3,
step_size=2000, mode='triangular', gamma=1.,
scale_fn=None, scale_mode='cycle', last_batch_iteration=-1):
if not isinstance(optimizer, Optimizer):
raise TypeError('{} is not an Optimizer'.format(
type(optimizer ).__name__))
self.optimizer = optimizer
if isinstance(base_lr, list)or isinstance(base_lr, tuple):
if len(base_lr)!= len(optimizer.param_groups):
raise ValueError("expected {} base_lr, got {}".format(
len(optimizer.param_groups), len(base_lr)))
self.base_lrs = list(base_lr)
else:
self.base_lrs = [base_lr] * len(optimizer.param_groups)
if isinstance(max_lr, list)or isinstance(max_lr, tuple):
if len(max_lr)!= len(optimizer.param_groups):
raise ValueError("expected {} max_lr, got {}".format(
len(optimizer.param_groups), len(max_lr)))
self.max_lrs = list(max_lr)
else:
self.max_lrs = [max_lr] * len(optimizer.param_groups)
self.step_size = step_size
if mode not in ['triangular', 'triangular2', 'exp_range'] \
and scale_fn is None:
raise ValueError('mode is invalid and scale_fn is None')
self.mode = mode
self.gamma = gamma
if scale_fn is None:
if self.mode == 'triangular':
self.scale_fn = self._triangular_scale_fn
self.scale_mode = 'cycle'
elif self.mode == 'triangular2':
self.scale_fn = self._triangular2_scale_fn
self.scale_mode = 'cycle'
elif self.mode == 'exp_range':
self.scale_fn = self._exp_range_scale_fn
self.scale_mode = 'iterations'
else:
self.scale_fn = scale_fn
self.scale_mode = scale_mode
self.batch_step(last_batch_iteration + 1)
self.last_batch_iteration = last_batch_iteration
def batch_step(self, batch_iteration=None):
if batch_iteration is None:
batch_iteration = self.last_batch_iteration + 1
self.last_batch_iteration = batch_iteration
for param_group, lr in zip(self.optimizer.param_groups, self.get_lr()):
param_group['lr'] = lr
def _triangular_scale_fn(self, x):
return 1.
def _triangular2_scale_fn(self, x):
return 1 /(2.**(x - 1))
def _exp_range_scale_fn(self, x):
return self.gamma**(x)
def get_lr(self):
step_size = float(self.step_size)
cycle = np.floor(1 + self.last_batch_iteration /(2 * step_size))
x = np.abs(self.last_batch_iteration / step_size - 2 * cycle + 1)
lrs = []
param_lrs = zip(self.optimizer.param_groups, self.base_lrs, self.max_lrs)
for param_group, base_lr, max_lr in param_lrs:
base_height =(max_lr - base_lr)* np.maximum(0,(1 - x))
if self.scale_mode == 'cycle':
lr = base_lr + base_height * self.scale_fn(cycle)
else:
lr = base_lr + base_height * self.scale_fn(self.last_batch_iteration)
lrs.append(lr)
return lrs
<choose_model_class> | REVISED_NUMERIC_COLUMNS=['Pclass','Age','SibSp','Parch','Family_Survival','Alone','Title_Master', 'Title_Miss','Title_Mr', 'Title_Mrs', 'Title_Millitary','Embarked']
SIMPLE_COLUMNS=['Pclass','Age','SibSp','Parch','Family_Survival','Alone','Sex_female','Sex_male','Title_Master', 'Title_Miss','Title_Mr', 'Title_Mrs', 'Title_Millitary','Embarked']
INTERESTING_COLUMNS=['Survived','Pclass','Age','SibSp','Parch','Title','Alone','Mother','Family Size','Family_Survival','Embarked','FareBand','TicketRef']
CATEGORY_COLUMNS=['Pclass','SibSp','Parch','Family Size','Family_Survival','Alone','Mother','Sex_female','Sex_male','AgeBand_Child',
'AgeBand_Young Adult', 'AgeBand_Adult', 'AgeBand_Older Adult',
'AgeBand_Senior','Title_Master', 'Title_Miss','Title_Mr', 'Title_Mrs', 'Title_Millitary','NameBand_1',
'NameBand_2', 'NameBand_3', 'NameBand_4', 'NameBand_5','Embarked','TicketRef_A', 'TicketRef_C', 'TicketRef_F', 'TicketRef_L',
'TicketRef_P', 'TicketRef_S', 'TicketRef_W', 'TicketRef_X','HadCabin','Free']
data_to_train = df_train[REVISED_NUMERIC_COLUMNS].fillna(-1000)
data_to_test = df_test[REVISED_NUMERIC_COLUMNS].fillna(-1000)
prediction = df_train["Survived"]
X_train, X_val, y_train, y_val = train_test_split(data_to_train, prediction, test_size = 0.3,random_state=21, stratify=prediction)
print('Data Split' ) | Titanic - Machine Learning from Disaster |
943,976 | embedding_dim = 300
embedding_path = '.. /save/embedding_matrix.npy'
use_pretrained_embedding = True
hidden_size = 60
gru_len = hidden_size
Routings = 4
Num_capsule = 5
Dim_capsule = 5
dropout_p = 0.25
rate_drop_dense = 0.28
LR = 0.001
T_epsilon = 1e-7
num_classes = 30
class Embed_Layer(nn.Module):
def __init__(self, embedding_matrix=None, vocab_size=None, embedding_dim=300):
super(Embed_Layer, self ).__init__()
self.encoder = nn.Embedding(vocab_size + 1, embedding_dim)
if use_pretrained_embedding:
self.encoder.weight.data.copy_(t.from_numpy(embedding_matrix))
def forward(self, x, dropout_p=0.25):
return nn.Dropout(p=dropout_p )(self.encoder(x))
class GRU_Layer(nn.Module):
def __init__(self):
super(GRU_Layer, self ).__init__()
self.gru = nn.GRU(input_size=300,
hidden_size=gru_len,
bidirectional=True)
def init_weights(self):
ih =(param.data for name, param in self.named_parameters() if 'weight_ih' in name)
hh =(param.data for name, param in self.named_parameters() if 'weight_hh' in name)
b =(param.data for name, param in self.named_parameters() if 'bias' in name)
for k in ih:
nn.init.xavier_uniform_(k)
for k in hh:
nn.init.orthogonal_(k)
for k in b:
nn.init.constant_(k, 0)
def forward(self, x):
return self.gru(x)
class Caps_Layer(nn.Module):
def __init__(self, input_dim_capsule=gru_len * 2, num_capsule=Num_capsule, dim_capsule=Dim_capsule, \
routings=Routings, kernel_size=(9, 1), share_weights=True,
activation='default', **kwargs):
super(Caps_Layer, self ).__init__(**kwargs)
self.num_capsule = num_capsule
self.dim_capsule = dim_capsule
self.routings = routings
self.kernel_size = kernel_size
self.share_weights = share_weights
if activation == 'default':
self.activation = self.squash
else:
self.activation = nn.ReLU(inplace=True)
if self.share_weights:
self.W = nn.Parameter(
nn.init.xavier_normal_(t.empty(1, input_dim_capsule, self.num_capsule * self.dim_capsule)))
else:
self.W = nn.Parameter(
t.randn(BATCH_SIZE, input_dim_capsule, self.num_capsule * self.dim_capsule))
def forward(self, x):
if self.share_weights:
u_hat_vecs = t.matmul(x, self.W)
else:
print('add later')
batch_size = x.size(0)
input_num_capsule = x.size(1)
u_hat_vecs = u_hat_vecs.view(( batch_size, input_num_capsule,
self.num_capsule, self.dim_capsule))
u_hat_vecs = u_hat_vecs.permute(0, 2, 1, 3)
b = t.zeros_like(u_hat_vecs[:, :, :, 0])
for i in range(self.routings):
b = b.permute(0, 2, 1)
c = F.softmax(b, dim=2)
c = c.permute(0, 2, 1)
b = b.permute(0, 2, 1)
outputs = self.activation(t.einsum('bij,bijk->bik',(c, u_hat_vecs)))
if i < self.routings - 1:
b = t.einsum('bik,bijk->bij',(outputs, u_hat_vecs))
return outputs
def squash(self, x, axis=-1):
s_squared_norm =(x ** 2 ).sum(axis, keepdim=True)
scale = t.sqrt(s_squared_norm + T_epsilon)
return x / scale
class Capsule_Main(nn.Module):
def __init__(self, embedding_matrix=None, vocab_size=None):
super(Capsule_Main, self ).__init__()
self.embed_layer = Embed_Layer(embedding_matrix, vocab_size)
self.gru_layer = GRU_Layer()
self.gru_layer.init_weights()
self.caps_layer = Caps_Layer()
self.dense_layer = Dense_Layer()
def forward(self, content):
content1 = self.embed_layer(content)
content2, _ = self.gru_layer(
content1)
content3 = self.caps_layer(content2)
output = self.dense_layer(content3)
return output
<normalization> | logreg = LogisticRegression(C=10, solver='newton-cg')
logreg.fit(X_train, y_train)
y_pred_train_logreg = cross_val_predict(logreg,X_val, y_val)
y_pred_test_logreg = logreg.predict(X_test)
print('logreg first layer predicted')
tree = DecisionTreeClassifier(random_state=8,min_samples_leaf=6, max_features= 7, max_depth= 4, criterion='gini', splitter='best')
tree.fit(X_train, y_train)
y_pred_train_tree = cross_val_predict(tree,X_val,y_val)
y_pred_test_tree = tree.predict(X_test)
print('decision tree first layer predicted')
randomforest = RandomForestClassifier(random_state=8, n_estimators=15, min_samples_leaf= 4, max_features= 6, max_depth=4,criterion='gini')
randomforest.fit(X_train, y_train)
y_pred_train_randomforest = cross_val_predict(randomforest, X_val, y_val)
y_pred_test_randomforest = randomforest.predict(X_test)
print('random forest first layer predicted')
gbk = GradientBoostingClassifier(min_samples_leaf=3, max_features= 3, max_depth= 3)
gbk.fit(X_train, y_train)
y_pred_train_gbk = cross_val_predict(gbk, X_val, y_val)
y_pred_test_gbk = gbk.predict(X_test)
print('gbk first layer predicted')
knn = KNeighborsClassifier(algorithm='auto', leaf_size=36, metric='minkowski',metric_params=None, n_jobs=1, n_neighbors=12, p=2,weights='uniform')
knn.fit(X_train, y_train)
y_pred_train_knn = cross_val_predict(knn, X_val, y_val)
y_pred_test_knn = gbk.predict(X_test)
print('knn first layer predicted')
clf = SVC(C=3, degree=1, kernel='linear', max_iter=1, shrinking=0)
clf.fit(X_train, y_train)
y_pred_train_clf = cross_val_predict(clf, X_val, y_val)
y_pred_test_clf = clf.predict(X_test)
print('clf first layer predicted' ) | Titanic - Machine Learning from Disaster |
943,976 | class Attention(nn.Module):
def __init__(self, feature_dim, step_dim, bias=True, **kwargs):
super(Attention, self ).__init__(**kwargs)
self.supports_masking = True
self.bias = bias
self.feature_dim = feature_dim
self.step_dim = step_dim
self.features_dim = 0
weight = torch.zeros(feature_dim, 1)
nn.init.xavier_uniform_(weight)
self.weight = nn.Parameter(weight)
if bias:
self.b = nn.Parameter(torch.zeros(step_dim))
def forward(self, x, mask=None):
feature_dim = self.feature_dim
step_dim = self.step_dim
eij = torch.mm(
x.contiguous().view(-1, feature_dim),
self.weight
).view(-1, step_dim)
if self.bias:
eij = eij + self.b
eij = torch.tanh(eij)
a = torch.exp(eij)
if mask is not None:
a = a * mask
a = a / torch.sum(a, 1, keepdim=True)+ 1e-10
weighted_input = x * torch.unsqueeze(a, -1)
return torch.sum(weighted_input, 1)
class NeuralNet(nn.Module):
def __init__(self):
super(NeuralNet, self ).__init__()
fc_layer = 16
fc_layer1 = 16
self.embedding = nn.Embedding(max_features, embed_size)
self.embedding.weight = nn.Parameter(torch.tensor(embedding_matrix, dtype=torch.float32))
self.embedding.weight.requires_grad = False
self.embedding_dropout = nn.Dropout2d(0.1)
self.lstm = nn.LSTM(embed_size, hidden_size, bidirectional=True, batch_first=True)
self.gru = nn.GRU(hidden_size * 2, hidden_size, bidirectional=True, batch_first=True)
self.lstm2 = nn.LSTM(hidden_size * 2, hidden_size, bidirectional=True, batch_first=True)
self.lstm_attention = Attention(hidden_size * 2, maxlen)
self.gru_attention = Attention(hidden_size * 2, maxlen)
self.bn = nn.BatchNorm1d(16, momentum=0.5)
self.linear = nn.Linear(hidden_size*8+3, fc_layer1)
self.relu = nn.ReLU()
self.dropout = nn.Dropout(0.1)
self.fc = nn.Linear(fc_layer**2,fc_layer)
self.out = nn.Linear(fc_layer, 1)
self.lincaps = nn.Linear(Num_capsule * Dim_capsule, 1)
self.caps_layer = Caps_Layer()
def forward(self, x):
h_embedding = self.embedding(x[0])
h_embedding = torch.squeeze(
self.embedding_dropout(torch.unsqueeze(h_embedding, 0)))
h_lstm, _ = self.lstm(h_embedding)
h_gru, _ = self.gru(h_lstm)
content3 = self.caps_layer(h_gru)
content3 = self.dropout(content3)
batch_size = content3.size(0)
content3 = content3.view(batch_size, -1)
content3 = self.relu(self.lincaps(content3))
h_lstm_atten = self.lstm_attention(h_lstm)
h_gru_atten = self.gru_attention(h_gru)
avg_pool = torch.mean(h_gru, 1)
max_pool, _ = torch.max(h_gru, 1)
f = torch.tensor(x[1], dtype=torch.float ).cuda()
conc = torch.cat(( h_lstm_atten, h_gru_atten,content3, avg_pool, max_pool,f), 1)
conc = self.relu(self.linear(conc))
conc = self.bn(conc)
conc = self.dropout(conc)
out = self.out(conc)
return out<categorify> | votingC = VotingClassifier(estimators=[('logreg', logreg_cv.best_estimator_),('gbk', gbk_cv.best_estimator_),
('tree', tree_cv.best_estimator_),('randomforest',randomforest_cv.best_estimator_),('knn',knn_cv.best_estimator_)], voting='soft', n_jobs=4)
votingC = votingC.fit(X_train, y_train)
Submission['Survived'] = votingC.predict(X_test)
Submission.to_csv('Votingclassifier02.csv',sep=',')
print('Voting Classifier Ensemble File created')
print(Submission.head() ) | Titanic - Machine Learning from Disaster |
943,976 | class MyDataset(Dataset):
def __init__(self,dataset):
self.dataset = dataset
def __getitem__(self, index):
data, target = self.dataset[index]
return data, target, index
def __len__(self):
return len(self.dataset )<define_variables> | second_layer_train = pd.DataFrame({'Logistic Regression': y_pred_train_logreg.ravel() ,
'Gradient Boosting': y_pred_train_gbk.ravel() ,
'Decision Tree': y_pred_train_tree.ravel() ,
'Random Forest': y_pred_train_randomforest.ravel()
})
X_train_second = np.concatenate(( y_pred_train_logreg.reshape(-1, 1), y_pred_train_gbk.reshape(-1, 1),
y_pred_train_tree.reshape(-1, 1), y_pred_train_randomforest.reshape(-1, 1)) ,
axis=1)
X_test_second = np.concatenate(( y_pred_test_logreg.reshape(-1, 1), y_pred_test_gbk.reshape(-1, 1),
y_pred_test_tree.reshape(-1, 1), y_pred_test_randomforest.reshape(-1, 1)) ,
axis=1)
tree = DecisionTreeClassifier(random_state=8,min_samples_leaf=6, max_depth= 4, criterion='gini' ).fit(X_train_second,y_val)
Submission['Survived'] = tree.predict(X_test_second)
print(Submission.head())
print('Tuned Ensemble model prediction complete' ) | Titanic - Machine Learning from Disaster |
943,976 | <data_type_conversions><EOS> | Submission.to_csv('tunedensemblesubmission04.csv',sep=',')
print('tuned Ensemble File created' ) | Titanic - Machine Learning from Disaster |
5,214,044 | <SOS> metric: categorizationaccuracy Kaggle data source: titanic-machine-learning-from-disaster<compute_test_metric> | train_data = pd.read_csv('/kaggle/input/train.csv')
test_data = pd.read_csv('/kaggle/input/test.csv' ) | Titanic - Machine Learning from Disaster |
5,214,044 | def bestThresshold(y_train,train_preds):
tmp = [0,0,0]
delta = 0
for tmp[0] in tqdm(np.arange(0.1, 0.501, 0.01)) :
tmp[1] = f1_score(y_train, np.array(train_preds)>tmp[0])
if tmp[1] > tmp[2]:
delta = tmp[0]
tmp[2] = tmp[1]
print('best threshold is {:.4f} with F1 score: {:.4f}'.format(delta, tmp[2]))
return delta
delta = bestThresshold(y_train,train_preds )<save_to_csv> | full_data = [train_data, test_data]
for idx, dataset in enumerate(full_data):
dataset = pd.concat([pd.get_dummies(dataset['Pclass'], prefix='Pclass'), dataset], axis=1)
dataset = dataset.drop(['Pclass', 'Pclass_3'], axis=1)
dataset['Title'] = dataset['Name'].apply(get_title)
dataset['Title'] = dataset['Title'].replace(['Lady', 'Countess','Capt', 'Col','Don', 'Dr', 'Major', 'Rev', 'Sir', 'Jonkheer', 'Dona'], 'Rare')
dataset['Title'] = dataset['Title'].replace('Mlle', 'Miss')
dataset['Title'] = dataset['Title'].replace('Ms', 'Miss')
dataset['Title'] = dataset['Title'].replace('Mme', 'Mrs')
title_mapping = {"Mr": 1, "Miss": 2, "Mrs": 3, "Master": 4, "Rare": 5}
dataset['Title'] = dataset['Title'].map(title_mapping)
dataset['Title'] = dataset['Title'].fillna(0)
dataset = dataset.drop(['Name'], axis=1)
dataset = pd.concat([pd.get_dummies(dataset['Sex'], prefix='Sex'), dataset], axis=1)
dataset = dataset.drop(['Sex', 'Sex_male'], axis=1)
dataset['Age'].fillna(dataset['Age'].median() , inplace = True)
dataset['AgeBin'] = pd.cut(dataset['Age'].astype(int), 5, labels=[1, 2, 3, 4, 5])
dataset = dataset.drop(['Age'], axis=1)
dataset = dataset.drop(['Embarked'], axis=1)
dataset['FamilySize'] = dataset['SibSp'] + dataset['Parch'] + 1
dataset['IsAlone'] = 1
dataset['IsAlone'].loc[dataset['FamilySize'] > 1] = 0
dataset = dataset.drop(['SibSp', 'Parch'], axis=1)
dataset = dataset.drop(['Fare'], axis=1)
dataset = dataset.drop(['Cabin'], axis=1)
dataset = dataset.drop(['Ticket'], axis=1)
if idx == 0:
X_train, y_train = dataset.loc[:, dataset.columns != 'Survived'], dataset.loc[:, dataset.columns == 'Survived']
else:
submission_id = dataset['PassengerId']
X_test = dataset
| Titanic - Machine Learning from Disaster |
5,214,044 | submission = df_test[['qid']].copy()
submission['prediction'] =(test_preds > delta ).astype(int)
submission.to_csv('submission.csv', index=False )<load_from_csv> | kf = KFold(n_splits=5, random_state = 0)
clf = clf = RandomForestClassifier(n_estimators=400, max_depth=4, min_samples_split=4, random_state=0)
for train_index, test_index in kf.split(X_train):
__kf_X_train, __kf_X_test = X_train.values[train_index], X_train.values[test_index]
__kf_y_train, __kf_y_test = y_train.values[train_index].ravel() , y_train.values[test_index].ravel()
clf.fit(__kf_X_train, __kf_y_train)
print(accuracy_score(__kf_y_test, clf.predict(__kf_X_test)) ) | Titanic - Machine Learning from Disaster |
5,214,044 | <feature_engineering><EOS> | y_pred = pd.Series(clf.predict(X_test))
submission = pd.concat([submission_id, y_pred], axis=1)
submission = submission.rename(columns={0:'Survived'})
submission.to_csv('submisson.csv', index=False ) | Titanic - Machine Learning from Disaster |
6,678,936 | <SOS> metric: categorizationaccuracy Kaggle data source: titanic-machine-learning-from-disaster<train_model> | %matplotlib inline
warnings.filterwarnings("ignore" ) | Titanic - Machine Learning from Disaster |
6,678,936 | class EarlyStopping:
def __init__(self, patience=7, verbose=False):
self.patience = patience
self.verbose = verbose
self.counter = 0
self.best_score = None
self.early_stop = False
self.val_loss_min = np.Inf
def __call__(self, val_loss, model):
score = -val_loss
if self.best_score is None:
self.best_score = score
if self.verbose:
print(f'Validation loss decreased({self.val_loss_min:.6f} --> {val_loss:.6f} ).Saving model...')
self.save_checkpoint(val_loss, model)
elif score < self.best_score:
self.counter += 1
if self.verbose:
print(f'EarlyStopping counter: {self.counter} out of {self.patience}')
if self.counter >= self.patience:
self.early_stop = True
else:
self.best_score = score
self.save_checkpoint(val_loss, model)
self.counter = 0
def save_checkpoint(self, val_loss, model):
if self.verbose:
print(f'Validation loss decreased({self.val_loss_min:.6f} --> {val_loss:.6f} ).Saving model...')
torch.save(model.state_dict() , 'checkpoint.pt')
self.val_loss_min = val_loss<set_options> | path = ".. /input/titanic/" | Titanic - Machine Learning from Disaster |
6,678,936 | torch.cuda.init()
torch.cuda.empty_cache()
print('CUDA MEM:',torch.cuda.memory_allocated())
print('cuda:', torch.cuda.is_available())
print('cude index:',torch.cuda.current_device())
batch_size = 512
print('batch_size:',batch_size)
print('---')
train_loader = torchtext.data.BucketIterator(dataset=train,
batch_size=batch_size,
shuffle=True,
sort=False)
val_loader = torchtext.data.BucketIterator(dataset=val,
batch_size=batch_size,
shuffle=False,
sort=False)
test_loader = torchtext.data.BucketIterator(dataset=test,
batch_size=batch_size,
shuffle=False,
sort=False)
class Sentiment(nn.Module):
def __init__(self,vocab_vectors,padding_idx,batch_size):
super(Sentiment,self ).__init__()
print('Vocab vectors size:',vocab_vectors.shape)
self.batch_size = batch_size
self.hidden_dim = 128
self.embedding = nn.Embedding.from_pretrained(vocab_vectors)
self.embedding.weight.requires_grad = False
self.embedding.padding_idx = padding_idx
self.cnns = nn.ModuleList([nn.Conv1d(in_channels=vocab_vectors.shape[1], out_channels=32, kernel_size=k)for k in [3,4,5]])
self.lstm = nn.LSTM(input_size=32,
hidden_size=self.hidden_dim,
bidirectional=True,
batch_first=True)
self.linear1 = nn.Linear(2*self.hidden_dim,self.hidden_dim)
self.linear2 = nn.Linear(self.hidden_dim,1)
self.dropout = nn.Dropout(0.2)
@staticmethod
def conv_and_max_pool(x, conv):
return F.max_pool1d(F.elu(conv(x)) ,1 ).permute(0, 2, 1)
def forward(self,x):
hidden =(torch.zeros(2, x.shape[0], self.hidden_dim ).cuda() ,
torch.zeros(2, x.shape[0], self.hidden_dim ).cuda())
e = self.embedding(x)
e = e.permute(0,2,1)
cnn_outs = []
for conv in self.cnns:
f =self.conv_and_max_pool(e,conv)
cnn_outs.append(f)
out = torch.cat(cnn_outs, dim=1 ).cuda()
_, hidden = self.lstm(out,hidden)
out = torch.cat(( hidden[0][-2,:,:], hidden[0][-1,:,:]), dim=1 ).cuda()
return self.linear2(self.dropout(F.relu(self.linear1(out))))
model = Sentiment(text.vocab.vectors, padding_idx=text.vocab.stoi[text.pad_token], batch_size=batch_size ).cuda()
print(model)
print('-'*80)
early_stopping = EarlyStopping(patience=2,verbose=True)
loss_function = nn.BCEWithLogitsLoss().cuda()
optimizer = optim.Adam(model.parameters() ,lr=1e-3)
def get_lr(optimizer):
for param_group in optimizer.param_groups:
return param_group['lr']
losses = []
val_losses=[]
epoch_acc=[]
epoch_val_acc=[]
lrs = []
for epoch in range(100):
epoch_losses=[]
epoch_val_losses = []
preds = []
val_preds=[]
targets = []
acc = []
model.train()
for batch,train_batch in enumerate(list(iter(train_loader)) ,1):
optimizer.zero_grad()
y_pred = model(train_batch.text.cuda() ).squeeze(1)
y_numpy_pred =torch.sigmoid(y_pred ).cpu().detach().numpy()
preds += y_numpy_pred.tolist()
y_true = train_batch.target.float().cuda()
y_numpy_true = train_batch.target.cpu().detach().numpy()
targets += y_numpy_true.tolist()
loss = loss_function(y_pred,y_true)
epoch_losses.append(loss.item())
loss.backward()
optimizer.step()
lrs.append(get_lr(optimizer))
acc.append(accuracy_score(y_numpy_true,np.round(y_numpy_pred)))
if batch % 100 == 0:
print('\rtraining(batch,loss,acc)| ',batch,' ===>',loss.item() ,' acc ',np.mean(acc),end='')
losses.append(np.mean(epoch_losses))
targets = np.array(targets)
preds = np.array(preds)
search_result = threshold_search(targets, preds)
train_f1 = search_result['f1']
epoch_acc.append(np.mean(acc))
targets = []
val_acc=[]
model.eval()
with torch.no_grad() :
for batch,val_batch in enumerate(list(val_loader),1):
y_pred = model(val_batch.text.cuda() ).squeeze(1)
y_numpy_pred = torch.sigmoid(y_pred ).cpu().detach().numpy()
val_preds += y_numpy_pred.tolist()
y_true = val_batch.target.float().cuda()
y_numpy_true = val_batch.target.cpu().detach().numpy()
targets += y_numpy_true.tolist()
val_loss = loss_function(y_pred,y_true)
epoch_val_losses.append(val_loss.item())
val_acc.append(accuracy_score(y_numpy_true,np.round(y_numpy_pred)))
if batch % 100 == 0:
print('\rvalidation(batch,acc)| ',batch,' ===>', np.mean(val_acc),end='')
val_losses.append(np.mean(epoch_val_losses))
epoch_val_acc.append(np.mean(val_acc))
targets = np.array(targets)
val_preds = np.array(val_preds)
search_result = threshold_search(targets, val_preds)
val_f1 = search_result['f1']
print('
EPOCH: ',epoch,'
has acc of ',epoch_acc[-1],' ,has loss of ',losses[-1], ' ,f1 of ',train_f1,'
val acc of ',epoch_val_acc[-1],' ,val loss of ',val_losses[-1],' ,val f1 of ',val_f1)
print('-'*80)
if early_stopping.early_stop:
print("Early stopping at ",epoch," epoch")
break
else:
early_stopping(1.-val_f1, model)
print('Training finished.... ' )<load_pretrained> | cv_n_split = 2
random_state = 0
test_train_split_part = 0.15 | Titanic - Machine Learning from Disaster |
6,678,936 | print(os.listdir())
model = Sentiment(text.vocab.vectors, padding_idx=text.vocab.stoi[text.pad_token], batch_size=batch_size ).cuda()
model.load_state_dict(torch.load('checkpoint.pt'))<init_hyperparams> | metrics_all = {1 : 'r2_score', 2: 'acc', 3 : 'rmse', 4 : 're'}
metrics_now = [1, 2, 3, 4] | Titanic - Machine Learning from Disaster |
6,678,936 | print('Threshold:',search_result['threshold'])
submission_list = list(torchtext.data.BucketIterator(dataset=submission_x,
batch_size=batch_size,
sort=False,
train=False))
pred = []
with torch.no_grad() :
for submission_batch in submission_list:
model.eval()
x = submission_batch.text.cuda()
pred += torch.sigmoid(model(x ).squeeze(1)).cpu().data.numpy().tolist()
pred = np.array(pred)
df_subm = pd.DataFrame()
df_subm['qid'] = [qid.vocab.itos[j] for i in submission_list for j in i.qid.view(-1 ).numpy() ]
df_subm['prediction'] =(pred > search_result['threshold'] ).astype(int)
print(df_subm.head())
df_subm.to_csv('submission.csv', index=False )<set_options> | traindf = pd.read_csv(path + 'train.csv' ).set_index('PassengerId')
testdf = pd.read_csv(path + 'test.csv' ).set_index('PassengerId')
submission = pd.read_csv(path + 'gender_submission.csv' ) | Titanic - Machine Learning from Disaster |
6,678,936 | def seed_everything(seed=1234):
random.seed(seed)
os.environ['PYTHONHASHSEED'] = str(seed)
np.random.seed(seed)
torch.manual_seed(seed)
torch.cuda.manual_seed(seed)
torch.backends.cudnn.deterministic = True
seed_everything(6017)
print('Seeding done...' )<load_from_csv> | target_name = 'Survived' | Titanic - Machine Learning from Disaster |
6,678,936 | text = torchtext.data.Field(lower=True, batch_first=True, tokenize=word_tokenize, fix_length=100)
qid = torchtext.data.Field()
target = torchtext.data.Field(sequential=False, use_vocab=False, is_target=True)
train_dataset = torchtext.data.TabularDataset(path='.. /input/train.csv', format='csv',
fields={'question_text':('text',text),
'target':('target',target)})
train, test = train_dataset.split(split_ratio=[0.8,0.2],stratified=True,strata_field='target',random_state=random.getstate())
submission_x = torchtext.data.TabularDataset(path='.. /input/test.csv', format='csv',
fields={'qid':('qid', qid),
'question_text':('text', text)})
text.build_vocab(train_dataset, submission_x, min_freq=3)
qid.build_vocab(submission_x)
print('train dataset len:',len(train_dataset))
print('train len:',len(train))
print('test len:',len(test))<feature_engineering> | df = pd.concat([traindf, testdf], axis=0, sort=False)
df['Title'] = df.Name.str.split(',' ).str[1].str.split('.' ).str[0].str.strip()
df['Title'] = df.Name.str.split(',' ).str[1].str.split('.' ).str[0].str.strip()
df['IsWomanOrBoy'] =(( df.Title == 'Master')|(df.Sex == 'female'))
df['LastName'] = df.Name.str.split(',' ).str[0]
family = df.groupby(df.LastName ).Survived
df['WomanOrBoyCount'] = family.transform(lambda s: s[df.IsWomanOrBoy].fillna(0 ).count())
df['WomanOrBoyCount'] = df.mask(df.IsWomanOrBoy, df.WomanOrBoyCount - 1, axis=0)
df['FamilySurvivedCount'] = family.transform(lambda s: s[df.IsWomanOrBoy].fillna(0 ).sum())
df['FamilySurvivedCount'] = df.mask(df.IsWomanOrBoy, df.FamilySurvivedCount - \
df.Survived.fillna(0), axis=0)
df['WomanOrBoySurvived'] = df.FamilySurvivedCount / df.WomanOrBoyCount.replace(0, np.nan)
df.WomanOrBoyCount = df.WomanOrBoyCount.replace(np.nan, 0)
df['Alone'] =(df.WomanOrBoyCount == 0 ) | Titanic - Machine Learning from Disaster |
6,678,936 | glove = torchtext.vocab.Vectors('.. /input/embeddings/glove.840B.300d/glove.840B.300d.txt')
text.vocab.set_vectors(glove.stoi, glove.vectors, dim=300 )<define_variables> | df['Title'] = df['Title'].replace('Ms','Miss')
df['Title'] = df['Title'].replace('Mlle','Miss')
df['Title'] = df['Title'].replace('Mme','Mrs')
df['Embarked'] = df['Embarked'].fillna('S')
med_fare = df.groupby(['Pclass', 'Parch', 'SibSp'] ).Fare.median() [3][0][0]
df['Fare'] = df['Fare'].fillna(med_fare)
df['famous_cabin'] = df["Cabin"].apply(lambda x: 0 if type(x)== float else 1)
df['Deck'] = df['Cabin'].apply(lambda s: s[0] if pd.notnull(s)else 'M')
df.loc[(df['Deck'] == 'T'), 'Deck'] = 'A'
df['Family_Size'] = df['SibSp'] + df['Parch'] + 1
df['Name_length'] = df['Name'].apply(len)
df.WomanOrBoySurvived = df.WomanOrBoySurvived.fillna(0)
df.WomanOrBoyCount = df.WomanOrBoyCount.fillna(0)
df.FamilySurvivedCount = df.FamilySurvivedCount.fillna(0)
df.Alone = df.Alone.fillna(0 ) | Titanic - Machine Learning from Disaster |
6,678,936 | batch_size = 512
print('batch_size:',batch_size)
print('---')
train_loader = torchtext.data.BucketIterator(dataset=train,
batch_size=batch_size,
shuffle=True,
sort=False)
test_loader = torchtext.data.BucketIterator(dataset=test,
batch_size=batch_size,
shuffle=False,
sort=False)
<set_options> | cols_to_drop = ['Name','Ticket','Cabin']
df = df.drop(cols_to_drop, axis=1 ) | Titanic - Machine Learning from Disaster |
6,678,936 | torch.cuda.init()
torch.cuda.empty_cache()
print('CUDA MEM:',torch.cuda.memory_allocated())
print('cuda:', torch.cuda.is_available())
print('cude index:',torch.cuda.current_device())
class SentimentLSTM(nn.Module):
def __init__(self,vocab_vectors,padding_idx,batch_size):
super(SentimentLSTM,self ).__init__()
print('Vocab vectors size:',vocab_vectors.shape)
self.batch_size = batch_size
self.hidden_dim = 128
self.n_layers = 2
self.embedding = nn.Embedding.from_pretrained(vocab_vectors)
self.embedding.weight.requires_grad = False
self.embedding.padding_idx = padding_idx
self.lstm = nn.LSTM(input_size=vocab_vectors.shape[1], hidden_size=self.hidden_dim, bidirectional=True,batch_first=True)
self.linear1 = nn.Linear(self.n_layers*self.hidden_dim,self.hidden_dim)
self.linear2 = nn.Linear(self.hidden_dim,1)
self.dropout = nn.Dropout(0.2)
def forward(self,x):
hidden =(torch.zeros(self.n_layers, x.shape[0], self.hidden_dim ).cuda() ,
torch.zeros(self.n_layers, x.shape[0], self.hidden_dim ).cuda())
e = self.embedding(x)
_, hidden = self.lstm(e, hidden)
out = torch.cat(( hidden[0][-2,:,:], hidden[0][-1,:,:]), dim=1 ).cuda()
out = self.linear1(out)
return self.linear2(self.dropout(F.relu(out)))
class SentimentBase(nn.Module):
def __init__(self):
super(SentimentBase,self ).__init__()
self.embedding = nn.Embedding(75966,300)
self.linear1 = nn.Linear(300*100,128)
self.linear2 = nn.Linear(128,1)
def forward(self,x):
emb = self.embedding(x)
pooled = emb.reshape(( emb.shape[0],emb.shape[1]*emb.shape[2]))
out = self.linear1(pooled)
out = self.linear2(F.relu(out))
return out
class SentimentCNN(nn.Module):
def __init__(self,vocab_vectors,padding_idx,batch_size):
super(SentimentCNN,self ).__init__()
print('Vocab vectors size:',vocab_vectors.shape)
self.batch_size = batch_size
self.hidden_dim = 128
self.embedding = nn.Embedding.from_pretrained(vocab_vectors)
self.embedding.weight.requires_grad = False
self.embedding.padding_idx = padding_idx
self.cnns = nn.ModuleList([nn.Conv1d(in_channels=vocab_vectors.shape[1], out_channels=self.hidden_dim, kernel_size=k)for k in [3,4,5]])
self.linear1 = nn.Linear(3*self.hidden_dim,self.hidden_dim)
self.linear2 = nn.Linear(self.hidden_dim,1)
self.dropout = nn.Dropout(0.2)
@staticmethod
def conv_and_max_pool(x, conv):
return F.relu(conv(x ).permute(0, 2, 1 ).max(1)[0])
def forward(self,x):
e = self.embedding(x)
e = e.permute(0,2,1)
cnn_outs = []
for conv in self.cnns:
f =self.conv_and_max_pool(e,conv)
cnn_outs.append(f)
out = torch.cat(cnn_outs, dim=1 ).cuda()
out = self.linear1(out)
return self.linear2(self.dropout(F.relu(out)))
class SentimentGRU(nn.Module):
def __init__(self,vocab_vectors,padding_idx,batch_size):
super(SentimentGRU,self ).__init__()
print('Vocab vectors size:',vocab_vectors.shape)
self.batch_size = batch_size
self.hidden_dim = 128
self.n_layers = 2
self.embedding = nn.Embedding.from_pretrained(vocab_vectors)
self.embedding.weight.requires_grad = False
self.embedding.padding_idx = padding_idx
self.gru = nn.GRU(input_size=vocab_vectors.shape[1], hidden_size=self.hidden_dim, bidirectional=True,batch_first=True)
self.linear1 = nn.Linear(self.n_layers*self.hidden_dim,self.hidden_dim)
self.linear2 = nn.Linear(self.hidden_dim,1)
self.dropout = nn.Dropout(0.2)
def forward(self,x):
hidden = torch.zeros(self.n_layers, x.shape[0], self.hidden_dim ).cuda()
e = self.embedding(x)
_, hidden = self.gru(e, hidden)
out = torch.cat(( hidden[-2,:,:], hidden[-1,:,:]), dim=1 ).cuda()
out = self.linear1(out)
return self.linear2(self.dropout(F.relu(out)))
def train(model,filename,epochs=3):
loss_function = nn.BCEWithLogitsLoss().cuda()
optimizer = optim.Adam(model.parameters())
for epoch in range(epochs):
model.train()
avg_loss = 0
for batch,train_batch in enumerate(list(iter(train_loader)) ,1):
optimizer.zero_grad()
y_pred = model(train_batch.text.cuda() ).squeeze(1)
y_true = train_batch.target.float().cuda()
loss = loss_function(y_pred,y_true)
avg_loss += loss.item()
loss.backward()
optimizer.step()
print('EPOCH: ',epoch,': ',avg_loss/batch)
print('-'*80)
torch.save(model.state_dict() , filename)
print('Training finished.... ')
<choose_model_class> | Y = df.Survived.loc[traindf.index].astype(int)
X_train, X_test = df.loc[traindf.index], df.loc[testdf.index]
X_test = X_test.drop(['Survived'], axis = 1 ) | Titanic - Machine Learning from Disaster |
6,678,936 | model = SentimentLSTM(text.vocab.vectors, padding_idx=text.vocab.stoi[text.pad_token], batch_size=batch_size ).cuda()
print(model)
print('-'*80)
train(model,'lstm.pt',3)
print('-'*80 )<train_model> | print(X_train.isnull().sum())
| Titanic - Machine Learning from Disaster |
6,678,936 | model = SentimentBase().cuda()
print(model)
print('-'*80)
train(model,'base.pt',5)
print('-'*80 )<train_on_grid> | numerics = ['int8', 'int16', 'int32', 'int64', 'float16', 'float32', 'float64']
categorical_columns = []
features = X_train.columns.values.tolist()
for col in features:
if X_train[col].dtype in numerics: continue
categorical_columns.append(col)
categorical_columns | Titanic - Machine Learning from Disaster |
6,678,936 | model = SentimentCNN(text.vocab.vectors, padding_idx=text.vocab.stoi[text.pad_token], batch_size=batch_size ).cuda()
print(model)
print('-'*80)
train(model,'cnn.pt',3)
print('-'*80 )<train_on_grid> | for col in categorical_columns:
if col in X_train.columns:
le = LabelEncoder()
le.fit(list(X_train[col].astype(str ).values)+ list(X_test[col].astype(str ).values))
X_train[col] = le.transform(list(X_train[col].astype(str ).values))
X_test[col] = le.transform(list(X_test[col].astype(str ).values)) | Titanic - Machine Learning from Disaster |
6,678,936 | model = SentimentGRU(text.vocab.vectors, padding_idx=text.vocab.stoi[text.pad_token], batch_size=batch_size ).cuda()
print(model)
print('-'*80)
train(model,'gru.pt',3)
print('-'*80 )<choose_model_class> | X_train = X_train.reset_index()
X_test = X_test.reset_index()
X_dropna_categor = X_train.dropna().astype(int)
Xtest_dropna_categor = X_test.dropna().astype(int ) | Titanic - Machine Learning from Disaster |
6,678,936 | def disable_grad(layer):
for p in layer.parameters() :
p.requires_grad=False
class Ensemble(nn.Module):
def __init__(self,vocab_vectors,padding_idx,batch_size):
super(Ensemble,self ).__init__()
self.lstm = SentimentLSTM(text.vocab.vectors, padding_idx=text.vocab.stoi[text.pad_token], batch_size=batch_size ).cuda()
self.lstm.load_state_dict(torch.load('lstm.pt'))
disable_grad(self.lstm)
self.gru = SentimentGRU(text.vocab.vectors, padding_idx=text.vocab.stoi[text.pad_token], batch_size=batch_size ).cuda()
self.gru.load_state_dict(torch.load('gru.pt'))
disable_grad(self.gru)
self.cnn = SentimentCNN(text.vocab.vectors, padding_idx=text.vocab.stoi[text.pad_token], batch_size=batch_size ).cuda()
self.cnn.load_state_dict(torch.load('cnn.pt'))
disable_grad(self.cnn)
self.base = SentimentBase().cuda()
self.base.load_state_dict(torch.load('base.pt'))
disable_grad(self.base)
self.l_in = nn.Linear(4,1024,bias=False)
self.l_out = nn.Linear(1024,1,bias=False)
def forward(self,x):
o1 = self.lstm(x)
o2 = self.gru(x)
o3 = self.cnn(x)
o4 = self.base(x)
out = torch.cat([o1,o2,o3,o4],1)
return self.l_out(F.relu(self.l_in(out)))
model = Ensemble(text.vocab.vectors, padding_idx=text.vocab.stoi[text.pad_token], batch_size=batch_size ).cuda()
print(model)
print('-'*80)
train(model,'ensemble.pt', 3)
print('-'*80 )<load_pretrained> | Sex_female_Survived = X_dropna_categor.loc[(X_dropna_categor.Sex == 0)&(X_dropna_categor.Survived == 1)]
Sex_female_NoSurvived = X_dropna_categor.loc[(X_dropna_categor.Sex == 0)&(X_dropna_categor.Survived == 0)]
X_Sex_male_Survived = X_dropna_categor.loc[(X_dropna_categor.Sex == 1)&(X_dropna_categor.Survived == 1)]
X_Sex_male_NoSurvived = X_dropna_categor.loc[(X_dropna_categor.Sex == 1)&(X_dropna_categor.Survived == 0)]
X_test_male = Xtest_dropna_categor.loc[Xtest_dropna_categor.Sex == 1]
X_test_female = Xtest_dropna_categor.loc[Xtest_dropna_categor.Sex == 0]
female_Survived_mean, female_NoSurvived_mean = Sex_female_Survived['Age'].mean() , Sex_female_NoSurvived['Age'].mean()
male_Survived_mean, male_NoSurvived_mean = X_Sex_male_Survived['Age'].mean() , X_Sex_male_NoSurvived['Age'].mean()
female_Survived_std, female_NoSurvived_std = Sex_female_Survived['Age'].std() , Sex_female_NoSurvived['Age'].std()
male_Survived_std, male_NoSurvived_std = X_Sex_male_Survived['Age'].std() , X_Sex_male_NoSurvived['Age'].std()
female_std, female_mean = X_test_female['Age'].std() , X_test_female['Age'].mean()
male_std, male_mean = X_test_male['Age'].std() , X_test_male['Age'].mean()
X_train['Survived'] = X_train['Survived'].astype(int ) | Titanic - Machine Learning from Disaster |
6,678,936 | print(os.listdir())
model = Ensemble(text.vocab.vectors, padding_idx=text.vocab.stoi[text.pad_token], batch_size=batch_size ).cuda()
model.load_state_dict(torch.load('ensemble.pt'))
<init_hyperparams> | def derf(sample, mean, std):
age_shape = sample['Age'].shape[0]
if age_shape > 0:
standard_error_ofthe_mean = std / math.sqrt(age_shape)
random_mean = round(random.uniform(mean-(1.96*standard_error_ofthe_mean), mean+(1.96*standard_error_ofthe_mean)) , 2)
else: random_mean = 0
return random_mean | Titanic - Machine Learning from Disaster |
6,678,936 | print('Threshold:',search_result['threshold'])
submission_list = list(torchtext.data.BucketIterator(dataset=submission_x,
batch_size=batch_size,
sort=False,
train=False))
pred = []
with torch.no_grad() :
for submission_batch in submission_list:
model.eval()
x = submission_batch.text.cuda()
pred += torch.sigmoid(model(x ).squeeze(1)).cpu().data.numpy().tolist()
pred = np.array(pred)
df_subm = pd.DataFrame()
df_subm['qid'] = [qid.vocab.itos[j] for i in submission_list for j in i.qid.view(-1 ).numpy() ]
df_subm['prediction'] =(pred > search_result['threshold'] ).astype(int)
print(df_subm.head())
df_subm.to_csv('submission.csv', index=False )<import_modules> | for i in X_train.loc[(X_train['Sex']==0)&(X_train['Survived']==1)&(X_train['Age'].isnull())].index:
X_train.at[i, 'Age'] = derf(Sex_female_Survived, female_Survived_mean, female_Survived_std)
for h in X_train.loc[(X_train['Sex']==0)&(X_train['Survived']==0)&(X_train['Age'].isnull())].index:
X_train.at[h, 'Age'] = derf(Sex_female_NoSurvived, female_NoSurvived_mean, female_NoSurvived_std)
for l in X_train.loc[(X_train['Sex']==1)&(X_train['Survived']==1)&(X_train['Age'].isnull())].index:
X_train.at[l, 'Age'] = derf(X_Sex_male_Survived, male_Survived_mean, male_Survived_std)
for b in X_train.loc[(X_train['Sex']==1)&(X_train['Survived']==0)&(X_train['Age'].isnull())].index:
X_train.at[b, 'Age'] = derf(X_Sex_male_NoSurvived, male_NoSurvived_mean, male_NoSurvived_std)
for p in X_test.loc[(X_test['Sex']==1)&(X_test['Age'].isnull())].index:
X_test.at[p, 'Age'] = derf(X_test_male, male_mean, male_std)
for y in X_test.loc[(X_test['Sex']==0)&(X_test['Age'].isnull())].index:
X_test.at[y, 'Age'] = derf(X_test_female, female_mean, female_std ) | Titanic - Machine Learning from Disaster |
6,678,936 | tqdm.pandas(desc='Progress')
<define_variables> | X_train = X_train.drop(['Survived'], axis = 1 ) | Titanic - Machine Learning from Disaster |
6,678,936 | embed_size = 300
max_features = 120000
maxlen = 70
batch_size = 512
n_epochs = 5
n_splits = 5
SEED = 1029<set_options> | print(X_train.isnull().sum())
print(X_test.isnull().sum() ) | Titanic - Machine Learning from Disaster |
6,678,936 | def seed_everything(seed=1029):
random.seed(seed)
os.environ['PYTHONHASHSEED'] = str(seed)
np.random.seed(seed)
torch.manual_seed(seed)
torch.cuda.manual_seed(seed)
torch.backends.cudnn.deterministic = True
seed_everything()<features_selection> | def fe_creation(df):
df['Age2'] = df['Age']//10
df['Fare2'] = df['Fare']//10
for i in ['Sex', 'Family_Size', 'Fare2','Alone', 'famous_cabin']:
for j in ['Age2','Title', 'Embarked', 'Deck']:
df[i + "_" + j] = df[i].astype('str')+ "_" + df[j].astype('str')
return df
X_train = fe_creation(X_train)
X_test = fe_creation(X_test ) | Titanic - Machine Learning from Disaster |
6,678,936 | def load_glove(word_index):
EMBEDDING_FILE = '.. /input/embeddings/glove.840B.300d/glove.840B.300d.txt'
def get_coefs(word,*arr): return word, np.asarray(arr, dtype='float32')[:300]
embeddings_index = dict(get_coefs(*o.split(" ")) for o in open(EMBEDDING_FILE))
all_embs = np.stack(embeddings_index.values())
emb_mean,emb_std = -0.005838499,0.48782197
embed_size = all_embs.shape[1]
nb_words = min(max_features, len(word_index))
embedding_matrix = np.random.normal(emb_mean, emb_std,(nb_words, embed_size))
for word, i in word_index.items() :
if i >= max_features: continue
embedding_vector = embeddings_index.get(word)
if embedding_vector is not None: embedding_matrix[i] = embedding_vector
return embedding_matrix
def load_fasttext(word_index):
EMBEDDING_FILE = '.. /input/embeddings/wiki-news-300d-1M/wiki-news-300d-1M.vec'
def get_coefs(word,*arr): return word, np.asarray(arr, dtype='float32')
embeddings_index = dict(get_coefs(*o.split(" ")) for o in open(EMBEDDING_FILE)if len(o)>100)
all_embs = np.stack(embeddings_index.values())
emb_mean,emb_std = all_embs.mean() , all_embs.std()
embed_size = all_embs.shape[1]
nb_words = min(max_features, len(word_index))
embedding_matrix = np.random.normal(emb_mean, emb_std,(nb_words, embed_size))
for word, i in word_index.items() :
if i >= max_features: continue
embedding_vector = embeddings_index.get(word)
if embedding_vector is not None: embedding_matrix[i] = embedding_vector
return embedding_matrix
def load_para(word_index):
EMBEDDING_FILE = '.. /input/embeddings/paragram_300_sl999/paragram_300_sl999.txt'
def get_coefs(word,*arr): return word, np.asarray(arr, dtype='float32')
embeddings_index = dict(get_coefs(*o.split(" ")) for o in open(EMBEDDING_FILE, encoding="utf8", errors='ignore')if len(o)>100)
all_embs = np.stack(embeddings_index.values())
emb_mean,emb_std = -0.0053247833,0.49346462
embed_size = all_embs.shape[1]
nb_words = min(max_features, len(word_index))
embedding_matrix = np.random.normal(emb_mean, emb_std,(nb_words, embed_size))
for word, i in word_index.items() :
if i >= max_features: continue
embedding_vector = embeddings_index.get(word)
if embedding_vector is not None: embedding_matrix[i] = embedding_vector
return embedding_matrix<load_from_csv> | categorical_columns = []
features = X_train.columns.values.tolist()
for col in features:
if X_train[col].dtype in numerics: continue
categorical_columns.append(col)
categorical_columns | Titanic - Machine Learning from Disaster |
6,678,936 | df_train = pd.read_csv(".. /input/train.csv")
df_test = pd.read_csv(".. /input/test.csv")
df = pd.concat([df_train ,df_test],sort=True )<feature_engineering> | for col in categorical_columns:
if col in X_train.columns:
le = LabelEncoder()
le.fit(list(X_train[col].astype(str ).values)+ list(X_test[col].astype(str ).values))
X_train[col] = le.transform(list(X_train[col].astype(str ).values))
X_test[col] = le.transform(list(X_test[col].astype(str ).values)) | Titanic - Machine Learning from Disaster |
6,678,936 | def build_vocab(texts):
sentences = texts.apply(lambda x: x.split() ).values
vocab = {}
for sentence in sentences:
for word in sentence:
try:
vocab[word] += 1
except KeyError:
vocab[word] = 1
return vocab
vocab = build_vocab(df['question_text'] )<define_variables> | train0, test0 = X_train, X_test
target0 = Y | Titanic - Machine Learning from Disaster |
6,678,936 | sin = len(df_train[df_train["target"]==0])
insin = len(df_train[df_train["target"]==1])
persin =(sin/(sin+insin)) *100
perinsin =(insin/(sin+insin)) *100
print("
print("<feature_engineering> | scaler = StandardScaler()
train0 = pd.DataFrame(scaler.fit_transform(train0), columns = train0.columns)
test0 = pd.DataFrame(scaler.transform(test0), columns = test0.columns)
train0b = train0.copy()
test0b = test0.copy()
trainb, testb, targetb, target_testb = train_test_split(train0b, target0, test_size=test_train_split_part, random_state=random_state ) | Titanic - Machine Learning from Disaster |
6,678,936 | def build_vocab(texts):
sentences = texts.apply(lambda x: x.split() ).values
vocab = {}
for sentence in sentences:
for word in sentence:
try:
vocab[word] += 1
except KeyError:
vocab[word] = 1
return vocab
def known_contractions(embed):
known = []
for contract in contraction_mapping:
if contract in embed:
known.append(contract)
return known
def clean_contractions(text, mapping):
specials = ["’", "‘", "´", "`"]
for s in specials:
text = text.replace(s, "'")
text = ' '.join([mapping[t] if t in mapping else t for t in text.split(" ")])
return text
def correct_spelling(x, dic):
for word in dic.keys() :
x = x.replace(word, dic[word])
return x
def unknown_punct(embed, punct):
unknown = ''
for p in punct:
if p not in embed:
unknown += p
unknown += ' '
return unknown
def clean_numbers(x):
x = re.sub('[0-9]{5,}', '
x = re.sub('[0-9]{4}', '
x = re.sub('[0-9]{3}', '
x = re.sub('[0-9]{2}', '
return x
def clean_special_chars(text, punct, mapping):
for p in mapping:
text = text.replace(p, mapping[p])
for p in punct:
text = text.replace(p, f' {p} ')
specials = {'\u200b': ' ', '…': '...', '\ufeff': '', 'करना': '', 'है': ''}
for s in specials:
text = text.replace(s, specials[s])
return text
def add_lower(embedding, vocab):
count = 0
for word in vocab:
if word in embedding and word.lower() not in embedding:
embedding[word.lower() ] = embedding[word]
count += 1
print(f"Added {count} words to embedding" )<define_variables> | scaler = MinMaxScaler()
train0 = pd.DataFrame(scaler.fit_transform(train0), columns = train0.columns)
test0 = pd.DataFrame(scaler.fit_transform(test0), columns = test0.columns ) | Titanic - Machine Learning from Disaster |
6,678,936 | puncts = [',', '.', '"', ':', ')', '(', '-', '!', '?', '|', ';', "'", '$', '&', '/', '[', ']', '>', '%', '=', '
'·', '_', '{', '}', '©', '^', '®', '`', '<', '→', '°', '€', '™', '›', '♥', '←', '×', '§', '″', '′', 'Â', '█', '½', 'à', '…',
'“', '★', '”', '–', '●', 'â', '►', '−', '¢', '²', '¬', '░', '¶', '↑', '±', '¿', '▾', '═', '¦', '║', '―', '¥', '▓', '—', '‹', '─',
'▒', '：', '¼', '⊕', '▼', '▪', '†', '■', '’', '▀', '¨', '▄', '♫', '☆', 'é', '¯', '♦', '¤', '▲', 'è', '¸', '¾', 'Ã', '⋅', '‘', '∞',
'∙', '）', '↓', '、', '│', '（', '»', '，', '♪', '╩', '╚', '³', '・', '╦', '╣', '╔', '╗', '▬', '❤', 'ï', 'Ø', '¹', '≤', '‡', '√', ]
def clean_text(x):
x = str(x)
for punct in puncts:
x = x.replace(punct, f' {punct} ')
return x
def clean_numbers(x):
return x
mispell_dict = {"ain't": "is not", "aren't": "are not","can't": "cannot", "'cause": "because", "could've": "could have", "couldn't": "could not", "didn't": "did not", "doesn't": "does not", "don't": "do not", "hadn't": "had not", "hasn't": "has not", "haven't": "have not", "he'd": "he would","he'll": "he will", "he's": "he is", "how'd": "how did", "how'd'y": "how do you", "how'll": "how will", "how's": "how is", "I'd": "I would", "I'd've": "I would have", "I'll": "I will", "I'll've": "I will have","I'm": "I am", "I've": "I have", "i'd": "i would", "i'd've": "i would have", "i'll": "i will", "i'll've": "i will have","i'm": "i am", "i've": "i have", "isn't": "is not", "it'd": "it would", "it'd've": "it would have", "it'll": "it will", "it'll've": "it will have","it's": "it is", "let's": "let us", "ma'am": "madam", "mayn't": "may not", "might've": "might have","mightn't": "might not","mightn't've": "might not have", "must've": "must have", "mustn't": "must not", "mustn't've": "must not have", "needn't": "need not", "needn't've": "need not have","o'clock": "of the clock", "oughtn't": "ought not", "oughtn't've": "ought not have", "shan't": "shall not", "sha'n't": "shall not", "shan't've": "shall not have", "she'd": "she would", "she'd've": "she would have", "she'll": "she will", "she'll've": "she will have", "she's": "she is", "should've": "should have", "shouldn't": "should not", "shouldn't've": "should not have", "so've": "so have","so's": "so as", "this's": "this is","that'd": "that would", "that'd've": "that would have", "that's": "that is", "there'd": "there would", "there'd've": "there would have", "there's": "there is", "here's": "here is","they'd": "they would", "they'd've": "they would have", "they'll": "they will", "they'll've": "they will have", "they're": "they are", "they've": "they have", "to've": "to have", "wasn't": "was not", "we'd": "we would", "we'd've": "we would have", "we'll": "we will", "we'll've": "we will have", "we're": "we are", "we've": "we have", "weren't": "were not", "what'll": "what will", "what'll've": "what will have", "what're": "what are", "what's": "what is", "what've": "what have", "when's": "when is", "when've": "when have", "where'd": "where did", "where's": "where is", "where've": "where have", "who'll": "who will", "who'll've": "who will have", "who's": "who is", "who've": "who have", "why's": "why is", "why've": "why have", "will've": "will have", "won't": "will not", "won't've": "will not have", "would've": "would have", "wouldn't": "would not", "wouldn't've": "would not have", "y'all": "you all", "y'all'd": "you all would","y'all'd've": "you all would have","y'all're": "you all are","y'all've": "you all have","you'd": "you would", "you'd've": "you would have", "you'll": "you will", "you'll've": "you will have", "you're": "you are", "you've": "you have", 'colour': 'color', 'centre': 'center', 'favourite': 'favorite', 'travelling': 'traveling', 'counselling': 'counseling', 'theatre': 'theater', 'cancelled': 'canceled', 'labour': 'labor', 'organisation': 'organization', 'wwii': 'world war 2', 'citicise': 'criticize', 'youtu ': 'youtube ', 'Qoura': 'Quora', 'sallary': 'salary', 'Whta': 'What', 'narcisist': 'narcissist', 'howdo': 'how do', 'whatare': 'what are', 'howcan': 'how can', 'howmuch': 'how much', 'howmany': 'how many', 'whydo': 'why do', 'doI': 'do I', 'theBest': 'the best', 'howdoes': 'how does', 'mastrubation': 'masturbation', 'mastrubate': 'masturbate', "mastrubating": 'masturbating', 'pennis': 'penis', 'Etherium': 'Ethereum', 'narcissit': 'narcissist', 'bigdata': 'big data', '2k17': '2017', '2k18': '2018', 'qouta': 'quota', 'exboyfriend': 'ex boyfriend', 'airhostess': 'air hostess', "whst": 'what', 'watsapp': 'whatsapp', 'demonitisation': 'demonetization', 'demonitization': 'demonetization', 'demonetisation': 'demonetization'}
def _get_mispell(mispell_dict):
mispell_re = re.compile('(%s)' % '|'.join(mispell_dict.keys()))
return mispell_dict, mispell_re
mispellings, mispellings_re = _get_mispell(mispell_dict)
def replace_typical_misspell(text):
def replace(match):
return mispellings[match.group(0)]
return mispellings_re.sub(replace, text )<feature_engineering> | train, test, target, target_test = train_test_split(train0, target0, test_size=test_train_split_part, random_state=random_state ) | Titanic - Machine Learning from Disaster |
6,678,936 | def add_features(df):
df['question_text'] = df['question_text'].progress_apply(lambda x:str(x))
df['total_length'] = df['question_text'].progress_apply(len)
df['capitals'] = df['question_text'].progress_apply(lambda comment: sum(1 for c in comment if c.isupper()))
df['caps_vs_length'] = df.progress_apply(lambda row: float(row['capitals'])/float(row['total_length']),
axis=1)
df['num_words'] = df.question_text.str.count('\S+')
df['num_unique_words'] = df['question_text'].progress_apply(lambda comment: len(set(w for w in comment.split())))
df['words_vs_unique'] = df['num_unique_words'] / df['num_words']
return df
def load_and_prec() :
train_df = pd.read_csv(".. /input/train.csv")
test_df = pd.read_csv(".. /input/test.csv")
print("Train shape : ",train_df.shape)
print("Test shape : ",test_df.shape)
train_df["question_text"] = train_df["question_text"].apply(lambda x: x.lower())
test_df["question_text"] = test_df["question_text"].apply(lambda x: x.lower())
train_df["question_text"] = train_df["question_text"].progress_apply(lambda x: clean_text(x))
test_df["question_text"] = test_df["question_text"].apply(lambda x: clean_text(x))
train_df["question_text"] = train_df["question_text"].progress_apply(lambda x: clean_numbers(x))
test_df["question_text"] = test_df["question_text"].apply(lambda x: clean_numbers(x))
train_df["question_text"] = train_df["question_text"].progress_apply(lambda x: replace_typical_misspell(x))
test_df["question_text"] = test_df["question_text"].apply(lambda x: replace_typical_misspell(x))
train_X = train_df["question_text"].fillna("_
test_X = test_df["question_text"].fillna("_
train = add_features(train_df)
test = add_features(test_df)
features = train[['caps_vs_length', 'words_vs_unique']].fillna(0)
test_features = test[['caps_vs_length', 'words_vs_unique']].fillna(0)
ss = StandardScaler()
ss.fit(np.vstack(( features, test_features)))
features = ss.transform(features)
test_features = ss.transform(test_features)
tokenizer = Tokenizer(num_words=max_features)
tokenizer.fit_on_texts(list(train_X))
train_X = tokenizer.texts_to_sequences(train_X)
test_X = tokenizer.texts_to_sequences(test_X)
train_X = pad_sequences(train_X, maxlen=maxlen)
test_X = pad_sequences(test_X, maxlen=maxlen)
train_y = train_df['target'].values
np.random.seed(SEED)
trn_idx = np.random.permutation(len(train_X))
train_X = train_X[trn_idx]
train_y = train_y[trn_idx]
features = features[trn_idx]
return train_X, test_X, train_y, features, test_features, tokenizer.word_index
<train_model> | num_models = 20
acc_train = []
acc_test = []
acc_all = np.empty(( len(metrics_now)*2, 0)).tolist()
acc_all | Titanic - Machine Learning from Disaster |
6,678,936 | x_train, x_test, y_train, features, test_features, word_index = load_and_prec()
<save_model> | acc_all_pred = np.empty(( len(metrics_now), 0)).tolist()
acc_all_pred | Titanic - Machine Learning from Disaster |
6,678,936 | np.save("x_train",x_train)
np.save("x_test",x_test)
np.save("y_train",y_train)
np.save("features",features)
np.save("test_features",test_features)
np.save("word_index.npy",word_index )<load_pretrained> | cv_train = ShuffleSplit(n_splits=cv_n_split, test_size=test_train_split_part, random_state=random_state ) | Titanic - Machine Learning from Disaster |
6,678,936 | x_train = np.load("x_train.npy")
x_test = np.load("x_test.npy")
y_train = np.load("y_train.npy")
features = np.load("features.npy")
test_features = np.load("test_features.npy")
word_index = np.load("word_index.npy" ).item()<normalization> | def acc_d(y_meas, y_pred):
return mean_absolute_error(y_meas, y_pred)*len(y_meas)/sum(abs(y_meas))
def acc_rmse(y_meas, y_pred):
return(mean_squared_error(y_meas, y_pred)) **0.5 | Titanic - Machine Learning from Disaster |
6,678,936 | seed_everything()
glove_embeddings = load_glove(word_index)
paragram_embeddings = load_para(word_index)
embedding_matrix = np.mean([glove_embeddings, paragram_embeddings, paragram_embeddings], axis=0)
del glove_embeddings, paragram_embeddings
gc.collect()
np.shape(embedding_matrix )<split> | def acc_metrics_calc(num,model,train,test,target,target_test):
global acc_all
ytrain = model.predict(train ).astype(int)
ytest = model.predict(test ).astype(int)
print('target = ', target[:5].values)
print('ytrain = ', ytrain[:5])
print('target_test =', target_test[:5].values)
print('ytest =', ytest[:5])
num_acc = 0
for x in metrics_now:
if x == 1:
acc_train = round(r2_score(target, ytrain)* 100, 2)
acc_test = round(r2_score(target_test, ytest)* 100, 2)
elif x == 2:
acc_train = round(metrics.accuracy_score(target, ytrain)* 100, 2)
acc_test = round(metrics.accuracy_score(target_test, ytest)* 100, 2)
elif x == 3:
acc_train = round(acc_rmse(target, ytrain)* 100, 2)
acc_test = round(acc_rmse(target_test, ytest)* 100, 2)
elif x == 4:
acc_train = round(acc_d(target, ytrain)* 100, 2)
acc_test = round(acc_d(target_test, ytest)* 100, 2)
print('acc of', metrics_all[x], 'for train =', acc_train)
print('acc of', metrics_all[x], 'for test =', acc_test)
acc_all[num_acc].append(acc_train)
acc_all[num_acc+1].append(acc_test)
num_acc += 2 | Titanic - Machine Learning from Disaster |
6,678,936 | splits = list(StratifiedKFold(n_splits=n_splits, shuffle=True, random_state=SEED ).split(x_train, y_train))
splits[:3]<choose_model_class> | def acc_metrics_calc_pred(num,model,name_model,train,test,target):
global acc_all_pred
ytrain = model.predict(train ).astype(int)
ytest = model.predict(test ).astype(int)
print('**********')
print(name_model)
print('target = ', target[:15].values)
print('ytrain = ', ytrain[:15])
print('ytest =', ytest[:15])
num_acc = 0
for x in metrics_now:
if x == 1:
acc_train = round(r2_score(target, ytrain)* 100, 2)
elif x == 2:
acc_train = round(metrics.accuracy_score(target, ytrain)* 100, 2)
elif x == 3:
acc_train = round(acc_rmse(target, ytrain)* 100, 2)
elif x == 4:
acc_train = round(acc_d(target, ytrain)* 100, 2)
print('acc of', metrics_all[x], 'for train =', acc_train)
acc_all_pred[num_acc].append(acc_train)
num_acc += 1
submission[target_name] = ytest
submission.to_csv('submission_' + name_model + '.csv', index=False ) | Titanic - Machine Learning from Disaster |
6,678,936 | class CyclicLR(object):
def __init__(self, optimizer, base_lr=1e-3, max_lr=6e-3,
step_size=2000, mode='triangular', gamma=1.,
scale_fn=None, scale_mode='cycle', last_batch_iteration=-1):
if not isinstance(optimizer, Optimizer):
raise TypeError('{} is not an Optimizer'.format(
type(optimizer ).__name__))
self.optimizer = optimizer
if isinstance(base_lr, list)or isinstance(base_lr, tuple):
if len(base_lr)!= len(optimizer.param_groups):
raise ValueError("expected {} base_lr, got {}".format(
len(optimizer.param_groups), len(base_lr)))
self.base_lrs = list(base_lr)
else:
self.base_lrs = [base_lr] * len(optimizer.param_groups)
if isinstance(max_lr, list)or isinstance(max_lr, tuple):
if len(max_lr)!= len(optimizer.param_groups):
raise ValueError("expected {} max_lr, got {}".format(
len(optimizer.param_groups), len(max_lr)))
self.max_lrs = list(max_lr)
else:
self.max_lrs = [max_lr] * len(optimizer.param_groups)
self.step_size = step_size
if mode not in ['triangular', 'triangular2', 'exp_range'] \
and scale_fn is None:
raise ValueError('mode is invalid and scale_fn is None')
self.mode = mode
self.gamma = gamma
if scale_fn is None:
if self.mode == 'triangular':
self.scale_fn = self._triangular_scale_fn
self.scale_mode = 'cycle'
elif self.mode == 'triangular2':
self.scale_fn = self._triangular2_scale_fn
self.scale_mode = 'cycle'
elif self.mode == 'exp_range':
self.scale_fn = self._exp_range_scale_fn
self.scale_mode = 'iterations'
else:
self.scale_fn = scale_fn
self.scale_mode = scale_mode
self.batch_step(last_batch_iteration + 1)
self.last_batch_iteration = last_batch_iteration
def batch_step(self, batch_iteration=None):
if batch_iteration is None:
batch_iteration = self.last_batch_iteration + 1
self.last_batch_iteration = batch_iteration
for param_group, lr in zip(self.optimizer.param_groups, self.get_lr()):
param_group['lr'] = lr
def _triangular_scale_fn(self, x):
return 1.
def _triangular2_scale_fn(self, x):
return 1 /(2.**(x - 1))
def _exp_range_scale_fn(self, x):
return self.gamma**(x)
def get_lr(self):
step_size = float(self.step_size)
cycle = np.floor(1 + self.last_batch_iteration /(2 * step_size))
x = np.abs(self.last_batch_iteration / step_size - 2 * cycle + 1)
lrs = []
param_lrs = zip(self.optimizer.param_groups, self.base_lrs, self.max_lrs)
for param_group, base_lr, max_lr in param_lrs:
base_height =(max_lr - base_lr)* np.maximum(0,(1 - x))
if self.scale_mode == 'cycle':
lr = base_lr + base_height * self.scale_fn(cycle)
else:
lr = base_lr + base_height * self.scale_fn(self.last_batch_iteration)
lrs.append(lr)
return lrs
<choose_model_class> | linreg = LinearRegression()
linreg_CV = GridSearchCV(linreg, param_grid={}, cv=cv_train, verbose=False)
linreg_CV.fit(train, target)
print(linreg_CV.best_params_)
acc_metrics_calc(0,linreg_CV,train,test,target,target_test ) | Titanic - Machine Learning from Disaster |
6,678,936 | embedding_dim = 300
embedding_path = '.. /save/embedding_matrix.npy'
use_pretrained_embedding = True
hidden_size = 60
gru_len = hidden_size
Routings = 4
Num_capsule = 5
Dim_capsule = 5
dropout_p = 0.25
rate_drop_dense = 0.28
LR = 0.001
T_epsilon = 1e-9
num_classes = 30
class Embed_Layer(nn.Module):
def __init__(self, embedding_matrix=None, vocab_size=None, embedding_dim=300):
super(Embed_Layer, self ).__init__()
self.encoder = nn.Embedding(vocab_size + 1, embedding_dim)
if use_pretrained_embedding:
self.encoder.weight.data.copy_(t.from_numpy(embedding_matrix))
def forward(self, x, dropout_p=0.25):
return nn.Dropout(p=dropout_p )(self.encoder(x))
class GRU_Layer(nn.Module):
def __init__(self):
super(GRU_Layer, self ).__init__()
self.gru = nn.GRU(input_size=300,
hidden_size=gru_len,
bidirectional=True)
def init_weights(self):
ih =(param.data for name, param in self.named_parameters() if 'weight_ih' in name)
hh =(param.data for name, param in self.named_parameters() if 'weight_hh' in name)
b =(param.data for name, param in self.named_parameters() if 'bias' in name)
for k in ih:
nn.init.xavier_uniform_(k)
for k in hh:
nn.init.orthogonal_(k)
for k in b:
nn.init.constant_(k, 0)
def forward(self, x):
return self.gru(x)
class Caps_Layer(nn.Module):
def __init__(self, input_dim_capsule=gru_len * 2, num_capsule=Num_capsule, dim_capsule=Dim_capsule, \
routings=Routings, kernel_size=(9, 1), share_weights=True,
activation='default', **kwargs):
super(Caps_Layer, self ).__init__(**kwargs)
self.num_capsule = num_capsule
self.dim_capsule = dim_capsule
self.routings = routings
self.kernel_size = kernel_size
self.share_weights = share_weights
if activation == 'default':
self.activation = self.squash
else:
self.activation = nn.ReLU(inplace=True)
if self.share_weights:
self.W = nn.Parameter(
nn.init.xavier_normal_(t.empty(1, input_dim_capsule, self.num_capsule * self.dim_capsule)))
else:
self.W = nn.Parameter(
t.randn(BATCH_SIZE, input_dim_capsule, self.num_capsule * self.dim_capsule))
def forward(self, x):
if self.share_weights:
u_hat_vecs = t.matmul(x, self.W)
else:
print('add later')
batch_size = x.size(0)
input_num_capsule = x.size(1)
u_hat_vecs = u_hat_vecs.view(( batch_size, input_num_capsule,
self.num_capsule, self.dim_capsule))
u_hat_vecs = u_hat_vecs.permute(0, 2, 1, 3)
b = t.zeros_like(u_hat_vecs[:, :, :, 0])
for i in range(self.routings):
b = b.permute(0, 2, 1)
c = F.softmax(b, dim=2)
c = c.permute(0, 2, 1)
b = b.permute(0, 2, 1)
outputs = self.activation(t.einsum('bij,bijk->bik',(c, u_hat_vecs)))
if i < self.routings - 1:
b = t.einsum('bik,bijk->bij',(outputs, u_hat_vecs))
return outputs
def squash(self, x, axis=-1):
s_squared_norm =(x ** 2 ).sum(axis, keepdim=True)
scale = t.sqrt(s_squared_norm + T_epsilon)
return x / scale
class Capsule_Main(nn.Module):
def __init__(self, embedding_matrix=None, vocab_size=None):
super(Capsule_Main, self ).__init__()
self.embed_layer = Embed_Layer(embedding_matrix, vocab_size)
self.gru_layer = GRU_Layer()
self.gru_layer.init_weights()
self.caps_layer = Caps_Layer()
self.dense_layer = Dense_Layer()
def forward(self, content):
content1 = self.embed_layer(content)
content2, _ = self.gru_layer(
content1)
content3 = self.caps_layer(content2)
output = self.dense_layer(content3)
return output
<normalization> | svr = SVC()
svr_CV = GridSearchCV(svr, param_grid={'kernel': ['linear', 'poly', 'rbf', 'sigmoid'],
'tol': [1e-4]},
cv=cv_train, verbose=False)
svr_CV.fit(train, target)
print(svr_CV.best_params_)
acc_metrics_calc(1,svr_CV,train,test,target,target_test ) | Titanic - Machine Learning from Disaster |
6,678,936 | class Attention(nn.Module):
def __init__(self, feature_dim, step_dim, bias=True, **kwargs):
super(Attention, self ).__init__(**kwargs)
self.supports_masking = True
self.bias = bias
self.feature_dim = feature_dim
self.step_dim = step_dim
self.features_dim = 0
weight = torch.zeros(feature_dim, 1)
nn.init.xavier_uniform_(weight)
self.weight = nn.Parameter(weight)
if bias:
self.b = nn.Parameter(torch.zeros(step_dim))
def forward(self, x, mask=None):
feature_dim = self.feature_dim
step_dim = self.step_dim
eij = torch.mm(
x.contiguous().view(-1, feature_dim),
self.weight
).view(-1, step_dim)
if self.bias:
eij = eij + self.b
eij = torch.tanh(eij)
a = torch.exp(eij)
if mask is not None:
a = a * mask
a = a / torch.sum(a, 1, keepdim=True)+ 1e-10
weighted_input = x * torch.unsqueeze(a, -1)
return torch.sum(weighted_input, 1)
class NeuralNet(nn.Module):
def __init__(self):
super(NeuralNet, self ).__init__()
fc_layer = 16
fc_layer1 = 16
self.embedding = nn.Embedding(max_features, embed_size)
self.embedding.weight = nn.Parameter(torch.tensor(embedding_matrix, dtype=torch.float32))
self.embedding.weight.requires_grad = False
self.embedding_dropout = nn.Dropout2d(0.1)
self.lstm = nn.LSTM(embed_size, hidden_size, bidirectional=True, batch_first=True)
self.gru = nn.GRU(hidden_size * 2, hidden_size, bidirectional=True, batch_first=True)
self.lstm2 = nn.LSTM(hidden_size * 2, hidden_size, bidirectional=True, batch_first=True)
self.lstm_attention = Attention(hidden_size * 2, maxlen)
self.gru_attention = Attention(hidden_size * 2, maxlen)
self.linear = nn.Linear(hidden_size*8+3, fc_layer1)
self.relu = nn.ReLU()
self.dropout = nn.Dropout(0.1)
self.fc = nn.Linear(fc_layer**2,fc_layer)
self.out = nn.Linear(fc_layer, 1)
self.lincaps = nn.Linear(Num_capsule * Dim_capsule, 1)
self.caps_layer = Caps_Layer()
def forward(self, x):
h_embedding = self.embedding(x[0])
h_embedding = torch.squeeze(
self.embedding_dropout(torch.unsqueeze(h_embedding, 0)))
h_lstm, _ = self.lstm(h_embedding)
h_gru, _ = self.gru(h_lstm)
content3 = self.caps_layer(h_gru)
content3 = self.dropout(content3)
batch_size = content3.size(0)
content3 = content3.view(batch_size, -1)
content3 = self.relu(self.lincaps(content3))
h_lstm_atten = self.lstm_attention(h_lstm)
h_gru_atten = self.gru_attention(h_gru)
avg_pool = torch.mean(h_gru, 1)
max_pool, _ = torch.max(h_gru, 1)
f = torch.tensor(x[1], dtype=torch.float ).cuda()
conc = torch.cat(( h_lstm_atten, h_gru_atten,content3, avg_pool, max_pool,f), 1)
conc = self.relu(self.linear(conc))
conc = self.dropout(conc)
out = self.out(conc)
return out<categorify> | linear_svc = LinearSVC()
param_grid = {'dual':[False],
'C': np.linspace(1, 15, 15)}
linear_svc_CV = GridSearchCV(linear_svc, param_grid=param_grid, cv=cv_train, verbose=False)
linear_svc_CV.fit(train, target)
print(linear_svc_CV.best_params_)
acc_metrics_calc(2,linear_svc_CV,train,test,target,target_test ) | Titanic - Machine Learning from Disaster |
6,678,936 | class MyDataset(Dataset):
def __init__(self,dataset):
self.dataset = dataset
def __getitem__(self, index):
data, target = self.dataset[index]
return data, target, index
def __len__(self):
return len(self.dataset )<define_variables> | %%time
mlp = MLPClassifier()
param_grid = {'hidden_layer_sizes': [i for i in range(2,5)],
'solver': ['sgd'],
'learning_rate': ['adaptive'],
'max_iter': [1000]
}
mlp_GS = GridSearchCV(mlp, param_grid=param_grid, cv=cv_train, verbose=False)
mlp_GS.fit(train, target)
print(mlp_GS.best_params_)
acc_metrics_calc(3,mlp_GS,train,test,target,target_test ) | Titanic - Machine Learning from Disaster |
6,678,936 | def sigmoid(x):
return 1 /(1 + np.exp(-x))
train_preds = np.zeros(( len(x_train)))
test_preds = np.zeros(( len(df_test)))
seed_everything()
x_test_cuda = torch.tensor(x_test, dtype=torch.long ).cuda()
test = torch.utils.data.TensorDataset(x_test_cuda)
test_loader = torch.utils.data.DataLoader(test, batch_size=batch_size, shuffle=False)
avg_losses_f = []
avg_val_losses_f = []<data_type_conversions> | sgd = SGDClassifier(early_stopping=True)
param_grid = {'alpha': [0.0001, 0.001, 0.01, 0.1, 1]}
sgd_CV = GridSearchCV(sgd, param_grid=param_grid, cv=cv_train, verbose=False)
sgd_CV.fit(train, target)
print(sgd_CV.best_params_)
acc_metrics_calc(4,sgd_CV,train,test,target,target_test ) | Titanic - Machine Learning from Disaster |
6,678,936 | for i,(train_idx, valid_idx)in enumerate(splits):
x_train = np.array(x_train)
y_train = np.array(y_train)
features = np.array(features)
x_train_fold = torch.tensor(x_train[train_idx.astype(int)], dtype=torch.long ).cuda()
y_train_fold = torch.tensor(y_train[train_idx.astype(int), np.newaxis], dtype=torch.float32 ).cuda()
kfold_X_features = features[train_idx.astype(int)]
kfold_X_valid_features = features[valid_idx.astype(int)]
x_val_fold = torch.tensor(x_train[valid_idx.astype(int)], dtype=torch.long ).cuda()
y_val_fold = torch.tensor(y_train[valid_idx.astype(int), np.newaxis], dtype=torch.float32 ).cuda()
model = NeuralNet()
model.cuda()
loss_fn = torch.nn.BCEWithLogitsLoss(reduction='sum')
step_size = 300
base_lr, max_lr = 0.001, 0.003
optimizer = torch.optim.Adam(filter(lambda p: p.requires_grad, model.parameters()),
lr=max_lr)
scheduler = CyclicLR(optimizer, base_lr=base_lr, max_lr=max_lr,
step_size=step_size, mode='exp_range',
gamma=0.99994)
train = torch.utils.data.TensorDataset(x_train_fold, y_train_fold)
valid = torch.utils.data.TensorDataset(x_val_fold, y_val_fold)
train = MyDataset(train)
valid = MyDataset(valid)
train_loader = torch.utils.data.DataLoader(train, batch_size=batch_size, shuffle=True)
valid_loader = torch.utils.data.DataLoader(valid, batch_size=batch_size, shuffle=False)
print(f'Fold {i + 1}')
for epoch in range(n_epochs):
start_time = time.time()
model.train()
avg_loss = 0.
for i,(x_batch, y_batch, index)in enumerate(train_loader):
f = kfold_X_features[index]
y_pred = model([x_batch,f])
if scheduler:
scheduler.batch_step()
loss = loss_fn(y_pred, y_batch)
optimizer.zero_grad()
loss.backward()
optimizer.step()
avg_loss += loss.item() / len(train_loader)
model.eval()
valid_preds_fold = np.zeros(( x_val_fold.size(0)))
test_preds_fold = np.zeros(( len(df_test)))
avg_val_loss = 0.
for i,(x_batch, y_batch, index)in enumerate(valid_loader):
f = kfold_X_valid_features[index]
y_pred = model([x_batch,f] ).detach()
avg_val_loss += loss_fn(y_pred, y_batch ).item() / len(valid_loader)
valid_preds_fold[i * batch_size:(i+1)* batch_size] = sigmoid(y_pred.cpu().numpy())[:, 0]
elapsed_time = time.time() - start_time
print('Epoch {}/{} \t loss={:.4f} \t val_loss={:.4f} \t time={:.2f}s'.format(
epoch + 1, n_epochs, avg_loss, avg_val_loss, elapsed_time))
avg_losses_f.append(avg_loss)
avg_val_losses_f.append(avg_val_loss)
for i,(x_batch,)in enumerate(test_loader):
f = test_features[i * batch_size:(i+1)* batch_size]
y_pred = model([x_batch,f] ).detach()
test_preds_fold[i * batch_size:(i+1)* batch_size] = sigmoid(y_pred.cpu().numpy())[:, 0]
train_preds[valid_idx] = valid_preds_fold
test_preds += test_preds_fold / len(splits)
print('All \t loss={:.4f} \t val_loss={:.4f} \t '.format(np.average(avg_losses_f),np.average(avg_val_losses_f)))
<compute_test_metric> | decision_tree = DecisionTreeClassifier()
param_grid = {'min_samples_leaf': [i for i in range(2,10)]}
decision_tree_CV = GridSearchCV(decision_tree, param_grid=param_grid, cv=cv_train, verbose=False)
decision_tree_CV.fit(train, target)
print(decision_tree_CV.best_params_)
acc_metrics_calc(5,decision_tree_CV,train,test,target,target_test ) | Titanic - Machine Learning from Disaster |
6,678,936 | def bestThresshold(y_train,train_preds):
tmp = [0,0,0]
delta = 0
for tmp[0] in tqdm(np.arange(0.1, 0.501, 0.01)) :
tmp[1] = f1_score(y_train, np.array(train_preds)>tmp[0])
if tmp[1] > tmp[2]:
delta = tmp[0]
tmp[2] = tmp[1]
print('best threshold is {:.4f} with F1 score: {:.4f}'.format(delta, tmp[2]))
return delta
delta = bestThresshold(y_train,train_preds )<save_to_csv> | %%time
random_forest = RandomForestClassifier()
param_grid = {'n_estimators': [300, 400, 500, 600], 'min_samples_split': [60], 'min_samples_leaf': [20, 25, 30, 35, 40],
'max_features': ['auto'], 'max_depth': [5, 6, 7, 8, 9, 10], 'criterion': ['gini'], 'bootstrap': [False]}
random_forest_CV = GridSearchCV(estimator=random_forest, param_grid=param_grid,
cv=cv_train, verbose=False)
random_forest_CV.fit(train, target)
print(random_forest_CV.best_params_)
acc_metrics_calc(6,random_forest_CV,train,test,target,target_test ) | Titanic - Machine Learning from Disaster |
6,678,936 | submission = df_test[['qid']].copy()
submission['prediction'] =(test_preds > delta ).astype(int)
submission.to_csv('submission.csv', index=False )<import_modules> | %%time
xgb_clf = xgb.XGBClassifier(objective='reg:squarederror')
parameters = {'n_estimators': [200, 300, 400],
'learning_rate': [0.001, 0.003, 0.005, 0.006, 0.01],
'max_depth': [4, 5, 6]}
xgb_reg = GridSearchCV(estimator=xgb_clf, param_grid=parameters, cv=cv_train ).fit(trainb, targetb)
print("Best score: %0.3f" % xgb_reg.best_score_)
print("Best parameters set:", xgb_reg.best_params_)
acc_metrics_calc(7,xgb_reg,trainb,testb,targetb,target_testb ) | Titanic - Machine Learning from Disaster |
6,678,936 | tqdm.pandas(desc='Progress')
<define_variables> | Xtrain, Xval, Ztrain, Zval = train_test_split(trainb, targetb, test_size=test_train_split_part, random_state=random_state)
modelL = lgb.LGBMClassifier(n_estimators=1000, num_leaves=50)
modelL.fit(Xtrain, Ztrain, eval_set=[(Xval, Zval)], early_stopping_rounds=50, verbose=True ) | Titanic - Machine Learning from Disaster |
6,678,936 | embed_size = 300
max_features = 120000
maxlen = 70
batch_size = 512
n_epochs = 5
n_splits = 5
SEED = 1029<set_options> | acc_metrics_calc(8,modelL,trainb,testb,targetb,target_testb ) | Titanic - Machine Learning from Disaster |
6,678,936 | def seed_everything(seed=1029):
random.seed(seed)
os.environ['PYTHONHASHSEED'] = str(seed)
np.random.seed(seed)
torch.manual_seed(seed)
torch.cuda.manual_seed(seed)
torch.backends.cudnn.deterministic = True
seed_everything()<features_selection> | gradient_boosting = GradientBoostingClassifier()
param_grid = {'learning_rate' : [0.001, 0.01, 0.1],
'max_depth': [i for i in range(2,5)],
'min_samples_leaf': [i for i in range(2,5)]}
gradient_boosting_CV = GridSearchCV(estimator=gradient_boosting, param_grid=param_grid,
cv=cv_train, verbose=False)
gradient_boosting_CV.fit(train, target)
print(gradient_boosting_CV.best_params_)
acc_metrics_calc(9,gradient_boosting_CV,train,test,target,target_test ) | Titanic - Machine Learning from Disaster |
6,678,936 | def load_glove(word_index):
EMBEDDING_FILE = '.. /input/embeddings/glove.840B.300d/glove.840B.300d.txt'
def get_coefs(word,*arr): return word, np.asarray(arr, dtype='float32')[:300]
embeddings_index = dict(get_coefs(*o.split(" ")) for o in open(EMBEDDING_FILE))
all_embs = np.stack(embeddings_index.values())
emb_mean,emb_std = -0.005838499,0.48782197
embed_size = all_embs.shape[1]
nb_words = min(max_features, len(word_index))
embedding_matrix = np.random.normal(emb_mean, emb_std,(nb_words, embed_size))
for word, i in word_index.items() :
if i >= max_features: continue
embedding_vector = embeddings_index.get(word)
if embedding_vector is not None: embedding_matrix[i] = embedding_vector
return embedding_matrix
def load_fasttext(word_index):
EMBEDDING_FILE = '.. /input/embeddings/wiki-news-300d-1M/wiki-news-300d-1M.vec'
def get_coefs(word,*arr): return word, np.asarray(arr, dtype='float32')
embeddings_index = dict(get_coefs(*o.split(" ")) for o in open(EMBEDDING_FILE)if len(o)>100)
all_embs = np.stack(embeddings_index.values())
emb_mean,emb_std = all_embs.mean() , all_embs.std()
embed_size = all_embs.shape[1]
nb_words = min(max_features, len(word_index))
embedding_matrix = np.random.normal(emb_mean, emb_std,(nb_words, embed_size))
for word, i in word_index.items() :
if i >= max_features: continue
embedding_vector = embeddings_index.get(word)
if embedding_vector is not None: embedding_matrix[i] = embedding_vector
return embedding_matrix
def load_para(word_index):
EMBEDDING_FILE = '.. /input/embeddings/paragram_300_sl999/paragram_300_sl999.txt'
def get_coefs(word,*arr): return word, np.asarray(arr, dtype='float32')
embeddings_index = dict(get_coefs(*o.split(" ")) for o in open(EMBEDDING_FILE, encoding="utf8", errors='ignore')if len(o)>100)
all_embs = np.stack(embeddings_index.values())
emb_mean,emb_std = -0.0053247833,0.49346462
embed_size = all_embs.shape[1]
nb_words = min(max_features, len(word_index))
embedding_matrix = np.random.normal(emb_mean, emb_std,(nb_words, embed_size))
for word, i in word_index.items() :
if i >= max_features: continue
embedding_vector = embeddings_index.get(word)
if embedding_vector is not None: embedding_matrix[i] = embedding_vector
return embedding_matrix<load_from_csv> | ridge = RidgeClassifier()
ridge_CV = GridSearchCV(estimator=ridge, param_grid={'alpha': np.linspace (.1, 1.5, 15)}, cv=cv_train, verbose=False)
ridge_CV.fit(train, target)
print(ridge_CV.best_params_)
acc_metrics_calc(10,ridge_CV,train,test,target,target_test ) | Titanic - Machine Learning from Disaster |
6,678,936 | df_train = pd.read_csv(".. /input/train.csv")
df_test = pd.read_csv(".. /input/test.csv")
df = pd.concat([df_train ,df_test],sort=True )<feature_engineering> | %%time
bagging = BaggingClassifier(base_estimator=linear_svc_CV)
param_grid={'max_features': [0.5, 0.6, 0.7, 0.8, 0.9, 1.0],
'n_estimators': [3, 5, 10],
'warm_start' : [True],
'random_state': [random_state]}
bagging_CV = GridSearchCV(estimator=bagging, param_grid=param_grid, cv=cv_train, verbose=False)
bagging_CV.fit(train, target)
print(bagging_CV.best_params_)
acc_metrics_calc(11,bagging_CV,train,test,target,target_test ) | Titanic - Machine Learning from Disaster |
6,678,936 | def build_vocab(texts):
sentences = texts.apply(lambda x: x.split() ).values
vocab = {}
for sentence in sentences:
for word in sentence:
try:
vocab[word] += 1
except KeyError:
vocab[word] = 1
return vocab
vocab = build_vocab(df['question_text'] )<define_variables> | etr = ExtraTreesClassifier()
etr_CV = GridSearchCV(estimator=etr, param_grid={'min_samples_leaf' : [10, 20, 30, 40, 50]}, cv=cv_train, verbose=False)
etr_CV.fit(train, target)
acc_metrics_calc(12,etr_CV,train,test,target,target_test ) | Titanic - Machine Learning from Disaster |
6,678,936 | sin = len(df_train[df_train["target"]==0])
insin = len(df_train[df_train["target"]==1])
persin =(sin/(sin+insin)) *100
perinsin =(insin/(sin+insin)) *100
print("
print("<feature_engineering> | Ada_Boost = AdaBoostClassifier()
Ada_Boost_CV = GridSearchCV(estimator=Ada_Boost, param_grid={'learning_rate' : [.01,.1,.5, 1]}, cv=cv_train, verbose=False)
Ada_Boost_CV.fit(train, target)
acc_metrics_calc(13,Ada_Boost_CV,train,test,target,target_test ) | Titanic - Machine Learning from Disaster |
6,678,936 | def build_vocab(texts):
sentences = texts.apply(lambda x: x.split() ).values
vocab = {}
for sentence in sentences:
for word in sentence:
try:
vocab[word] += 1
except KeyError:
vocab[word] = 1
return vocab
def known_contractions(embed):
known = []
for contract in contraction_mapping:
if contract in embed:
known.append(contract)
return known
def clean_contractions(text, mapping):
specials = ["’", "‘", "´", "`"]
for s in specials:
text = text.replace(s, "'")
text = ' '.join([mapping[t] if t in mapping else t for t in text.split(" ")])
return text
def correct_spelling(x, dic):
for word in dic.keys() :
x = x.replace(word, dic[word])
return x
def unknown_punct(embed, punct):
unknown = ''
for p in punct:
if p not in embed:
unknown += p
unknown += ' '
return unknown
def clean_numbers(x):
x = re.sub('[0-9]{5,}', '
x = re.sub('[0-9]{4}', '
x = re.sub('[0-9]{3}', '
x = re.sub('[0-9]{2}', '
return x
def clean_special_chars(text, punct, mapping):
for p in mapping:
text = text.replace(p, mapping[p])
for p in punct:
text = text.replace(p, f' {p} ')
specials = {'\u200b': ' ', '…': '...', '\ufeff': '', 'करना': '', 'है': ''}
for s in specials:
text = text.replace(s, specials[s])
return text
def add_lower(embedding, vocab):
count = 0
for word in vocab:
if word in embedding and word.lower() not in embedding:
embedding[word.lower() ] = embedding[word]
count += 1
print(f"Added {count} words to embedding" )<define_variables> | logreg = LogisticRegression()
logreg_CV = GridSearchCV(estimator=logreg, param_grid={'C' : [.1,.3,.5,.7, 1]}, cv=cv_train, verbose=False)
logreg_CV.fit(train, target)
acc_metrics_calc(14,logreg_CV,train,test,target,target_test ) | Titanic - Machine Learning from Disaster |
6,678,936 | puncts = [',', '.', '"', ':', ')', '(', '-', '!', '?', '|', ';', "'", '$', '&', '/', '[', ']', '>', '%', '=', '
'·', '_', '{', '}', '©', '^', '®', '`', '<', '→', '°', '€', '™', '›', '♥', '←', '×', '§', '″', '′', 'Â', '█', '½', 'à', '…',
'“', '★', '”', '–', '●', 'â', '►', '−', '¢', '²', '¬', '░', '¶', '↑', '±', '¿', '▾', '═', '¦', '║', '―', '¥', '▓', '—', '‹', '─',
'▒', '：', '¼', '⊕', '▼', '▪', '†', '■', '’', '▀', '¨', '▄', '♫', '☆', 'é', '¯', '♦', '¤', '▲', 'è', '¸', '¾', 'Ã', '⋅', '‘', '∞',
'∙', '）', '↓', '、', '│', '（', '»', '，', '♪', '╩', '╚', '³', '・', '╦', '╣', '╔', '╗', '▬', '❤', 'ï', 'Ø', '¹', '≤', '‡', '√', ]
def clean_text(x):
x = str(x)
for punct in puncts:
x = x.replace(punct, f' {punct} ')
return x
def clean_numbers(x):
x = re.sub('[0-9]{5,}', '
x = re.sub('[0-9]{4}', '
x = re.sub('[0-9]{3}', '
x = re.sub('[0-9]{2}', '
return x
mispell_dict = {"ain't": "is not", "aren't": "are not","can't": "cannot", "'cause": "because", "could've": "could have", "couldn't": "could not", "didn't": "did not", "doesn't": "does not", "don't": "do not", "hadn't": "had not", "hasn't": "has not", "haven't": "have not", "he'd": "he would","he'll": "he will", "he's": "he is", "how'd": "how did", "how'd'y": "how do you", "how'll": "how will", "how's": "how is", "I'd": "I would", "I'd've": "I would have", "I'll": "I will", "I'll've": "I will have","I'm": "I am", "I've": "I have", "i'd": "i would", "i'd've": "i would have", "i'll": "i will", "i'll've": "i will have","i'm": "i am", "i've": "i have", "isn't": "is not", "it'd": "it would", "it'd've": "it would have", "it'll": "it will", "it'll've": "it will have","it's": "it is", "let's": "let us", "ma'am": "madam", "mayn't": "may not", "might've": "might have","mightn't": "might not","mightn't've": "might not have", "must've": "must have", "mustn't": "must not", "mustn't've": "must not have", "needn't": "need not", "needn't've": "need not have","o'clock": "of the clock", "oughtn't": "ought not", "oughtn't've": "ought not have", "shan't": "shall not", "sha'n't": "shall not", "shan't've": "shall not have", "she'd": "she would", "she'd've": "she would have", "she'll": "she will", "she'll've": "she will have", "she's": "she is", "should've": "should have", "shouldn't": "should not", "shouldn't've": "should not have", "so've": "so have","so's": "so as", "this's": "this is","that'd": "that would", "that'd've": "that would have", "that's": "that is", "there'd": "there would", "there'd've": "there would have", "there's": "there is", "here's": "here is","they'd": "they would", "they'd've": "they would have", "they'll": "they will", "they'll've": "they will have", "they're": "they are", "they've": "they have", "to've": "to have", "wasn't": "was not", "we'd": "we would", "we'd've": "we would have", "we'll": "we will", "we'll've": "we will have", "we're": "we are", "we've": "we have", "weren't": "were not", "what'll": "what will", "what'll've": "what will have", "what're": "what are", "what's": "what is", "what've": "what have", "when's": "when is", "when've": "when have", "where'd": "where did", "where's": "where is", "where've": "where have", "who'll": "who will", "who'll've": "who will have", "who's": "who is", "who've": "who have", "why's": "why is", "why've": "why have", "will've": "will have", "won't": "will not", "won't've": "will not have", "would've": "would have", "wouldn't": "would not", "wouldn't've": "would not have", "y'all": "you all", "y'all'd": "you all would","y'all'd've": "you all would have","y'all're": "you all are","y'all've": "you all have","you'd": "you would", "you'd've": "you would have", "you'll": "you will", "you'll've": "you will have", "you're": "you are", "you've": "you have", 'colour': 'color', 'centre': 'center', 'favourite': 'favorite', 'travelling': 'traveling', 'counselling': 'counseling', 'theatre': 'theater', 'cancelled': 'canceled', 'labour': 'labor', 'organisation': 'organization', 'wwii': 'world war 2', 'citicise': 'criticize', 'youtu ': 'youtube ', 'Qoura': 'Quora', 'sallary': 'salary', 'Whta': 'What', 'narcisist': 'narcissist', 'howdo': 'how do', 'whatare': 'what are', 'howcan': 'how can', 'howmuch': 'how much', 'howmany': 'how many', 'whydo': 'why do', 'doI': 'do I', 'theBest': 'the best', 'howdoes': 'how does', 'mastrubation': 'masturbation', 'mastrubate': 'masturbate', "mastrubating": 'masturbating', 'pennis': 'penis', 'Etherium': 'Ethereum', 'narcissit': 'narcissist', 'bigdata': 'big data', '2k17': '2017', '2k18': '2018', 'qouta': 'quota', 'exboyfriend': 'ex boyfriend', 'airhostess': 'air hostess', "whst": 'what', 'watsapp': 'whatsapp', 'demonitisation': 'demonetization', 'demonitization': 'demonetization', 'demonetisation': 'demonetization'}
def _get_mispell(mispell_dict):
mispell_re = re.compile('(%s)' % '|'.join(mispell_dict.keys()))
return mispell_dict, mispell_re
mispellings, mispellings_re = _get_mispell(mispell_dict)
def replace_typical_misspell(text):
def replace(match):
return mispellings[match.group(0)]
return mispellings_re.sub(replace, text )<feature_engineering> | knn = KNeighborsClassifier()
param_grid={'n_neighbors': range(2, 7)}
knn_CV = GridSearchCV(estimator=knn, param_grid=param_grid,
cv=cv_train, verbose=False ).fit(train, target)
print(knn_CV.best_params_)
acc_metrics_calc(15,knn_CV,train,test,target,target_test ) | Titanic - Machine Learning from Disaster |
6,678,936 | def add_features(df):
df['question_text'] = df['question_text'].progress_apply(lambda x:str(x))
df['total_length'] = df['question_text'].progress_apply(len)
df['capitals'] = df['question_text'].progress_apply(lambda comment: sum(1 for c in comment if c.isupper()))
df['caps_vs_length'] = df.progress_apply(lambda row: float(row['capitals'])/float(row['total_length']),
axis=1)
df['num_words'] = df.question_text.str.count('\S+')
df['num_unique_words'] = df['question_text'].progress_apply(lambda comment: len(set(w for w in comment.split())))
df['words_vs_unique'] = df['num_unique_words'] / df['num_words']
return df
def load_and_prec() :
train_df = pd.read_csv(".. /input/train.csv")
test_df = pd.read_csv(".. /input/test.csv")
print("Train shape : ",train_df.shape)
print("Test shape : ",test_df.shape)
train_df["question_text"] = train_df["question_text"].apply(lambda x: x.lower())
test_df["question_text"] = test_df["question_text"].apply(lambda x: x.lower())
train_df["question_text"] = train_df["question_text"].progress_apply(lambda x: clean_text(x))
test_df["question_text"] = test_df["question_text"].apply(lambda x: clean_text(x))
train_df["question_text"] = train_df["question_text"].progress_apply(lambda x: clean_numbers(x))
test_df["question_text"] = test_df["question_text"].apply(lambda x: clean_numbers(x))
train_df["question_text"] = train_df["question_text"].progress_apply(lambda x: replace_typical_misspell(x))
test_df["question_text"] = test_df["question_text"].apply(lambda x: replace_typical_misspell(x))
train_X = train_df["question_text"].fillna("_
test_X = test_df["question_text"].fillna("_
train = add_features(train_df)
test = add_features(test_df)
features = train[['caps_vs_length', 'words_vs_unique']].fillna(0)
test_features = test[['caps_vs_length', 'words_vs_unique']].fillna(0)
ss = StandardScaler()
ss.fit(np.vstack(( features, test_features)))
features = ss.transform(features)
test_features = ss.transform(test_features)
tokenizer = Tokenizer(num_words=max_features)
tokenizer.fit_on_texts(list(train_X))
train_X = tokenizer.texts_to_sequences(train_X)
test_X = tokenizer.texts_to_sequences(test_X)
train_X = pad_sequences(train_X, maxlen=maxlen)
test_X = pad_sequences(test_X, maxlen=maxlen)
train_y = train_df['target'].values
np.random.seed(SEED)
trn_idx = np.random.permutation(len(train_X))
train_X = train_X[trn_idx]
train_y = train_y[trn_idx]
features = features[trn_idx]
return train_X, test_X, train_y, features, test_features, tokenizer.word_index
<train_model> | gaussian = GaussianNB()
param_grid={'var_smoothing': [1e-8, 1e-9, 1e-10]}
gaussian_CV = GridSearchCV(estimator=gaussian, param_grid=param_grid, cv=cv_train, verbose=False)
gaussian_CV.fit(train, target)
print(gaussian_CV.best_params_)
acc_metrics_calc(16,gaussian_CV,train,test,target,target_test ) | Titanic - Machine Learning from Disaster |
6,678,936 | x_train, x_test, y_train, features, test_features, word_index = load_and_prec()
<save_model> | perceptron = Perceptron()
param_grid = {'penalty': [None, 'l2', 'l1', 'elasticnet']}
perceptron_CV = GridSearchCV(estimator=perceptron, param_grid=param_grid, cv=cv_train, verbose=False)
perceptron_CV.fit(train, target)
print(perceptron_CV.best_params_)
acc_metrics_calc(17,perceptron_CV,train,test,target,target_test ) | Titanic - Machine Learning from Disaster |
6,678,936 | np.save("x_train",x_train)
np.save("x_test",x_test)
np.save("y_train",y_train)
np.save("features",features)
np.save("test_features",test_features)
np.save("word_index.npy",word_index )<load_pretrained> | gpc = GaussianProcessClassifier()
param_grid = {'max_iter_predict': [100, 200],
'warm_start': [True, False],
'n_restarts_optimizer': range(3)}
gpc_CV = GridSearchCV(estimator=gpc, param_grid=param_grid, cv=cv_train, verbose=False)
gpc_CV.fit(train, target)
print(gpc_CV.best_params_)
acc_metrics_calc(18,gpc_CV,train,test,target,target_test ) | Titanic - Machine Learning from Disaster |
6,678,936 | x_train = np.load("x_train.npy")
x_test = np.load("x_test.npy")
y_train = np.load("y_train.npy")
features = np.load("features.npy")
test_features = np.load("test_features.npy")
word_index = np.load("word_index.npy" ).item()<normalization> | Voting_ens = VotingClassifier(estimators=[('log', logreg_CV),('mlp', mlp_GS),('svc', linear_svc_CV)])
Voting_ens.fit(train, target)
acc_metrics_calc(19,Voting_ens,train,test,target,target_test ) | Titanic - Machine Learning from Disaster |
6,678,936 | seed_everything()
glove_embeddings = load_glove(word_index)
paragram_embeddings = load_para(word_index)
embedding_matrix = np.mean([glove_embeddings, paragram_embeddings], axis=0)
del glove_embeddings, paragram_embeddings
gc.collect()
np.shape(embedding_matrix )<split> | models = pd.DataFrame({
'Model': ['Linear Regression', 'Support Vector Machines', 'Linear SVC',
'MLPClassifier', 'Stochastic Gradient Decent',
'Decision Tree Classifier', 'Random Forest', 'XGBClassifier', 'LGBMClassifier',
'GradientBoostingClassifier', 'RidgeClassifier', 'BaggingClassifier', 'ExtraTreesClassifier',
'AdaBoostClassifier', 'Logistic Regression',
'KNN', 'Naive Bayes', 'Perceptron', 'Gaussian Process Classification',
'VotingClassifier']} ) | Titanic - Machine Learning from Disaster |
6,678,936 | splits = list(StratifiedKFold(n_splits=n_splits, shuffle=True, random_state=SEED ).split(x_train, y_train))
splits[:3]<choose_model_class> | for x in metrics_now:
xs = metrics_all[x]
models[xs + '_train'] = acc_all[(x-1)*2]
models[xs + '_test'] = acc_all[(x-1)*2+1]
if xs == "acc":
models[xs + '_diff'] = models[xs + '_train'] - models[xs + '_test']
models | Titanic - Machine Learning from Disaster |
6,678,936 | class CyclicLR(object):
def __init__(self, optimizer, base_lr=1e-3, max_lr=6e-3,
step_size=2000, mode='triangular', gamma=1.,
scale_fn=None, scale_mode='cycle', last_batch_iteration=-1):
if not isinstance(optimizer, Optimizer):
raise TypeError('{} is not an Optimizer'.format(
type(optimizer ).__name__))
self.optimizer = optimizer
if isinstance(base_lr, list)or isinstance(base_lr, tuple):
if len(base_lr)!= len(optimizer.param_groups):
raise ValueError("expected {} base_lr, got {}".format(
len(optimizer.param_groups), len(base_lr)))
self.base_lrs = list(base_lr)
else:
self.base_lrs = [base_lr] * len(optimizer.param_groups)
if isinstance(max_lr, list)or isinstance(max_lr, tuple):
if len(max_lr)!= len(optimizer.param_groups):
raise ValueError("expected {} max_lr, got {}".format(
len(optimizer.param_groups), len(max_lr)))
self.max_lrs = list(max_lr)
else:
self.max_lrs = [max_lr] * len(optimizer.param_groups)
self.step_size = step_size
if mode not in ['triangular', 'triangular2', 'exp_range'] \
and scale_fn is None:
raise ValueError('mode is invalid and scale_fn is None')
self.mode = mode
self.gamma = gamma
if scale_fn is None:
if self.mode == 'triangular':
self.scale_fn = self._triangular_scale_fn
self.scale_mode = 'cycle'
elif self.mode == 'triangular2':
self.scale_fn = self._triangular2_scale_fn
self.scale_mode = 'cycle'
elif self.mode == 'exp_range':
self.scale_fn = self._exp_range_scale_fn
self.scale_mode = 'iterations'
else:
self.scale_fn = scale_fn
self.scale_mode = scale_mode
self.batch_step(last_batch_iteration + 1)
self.last_batch_iteration = last_batch_iteration
def batch_step(self, batch_iteration=None):
if batch_iteration is None:
batch_iteration = self.last_batch_iteration + 1
self.last_batch_iteration = batch_iteration
for param_group, lr in zip(self.optimizer.param_groups, self.get_lr()):
param_group['lr'] = lr
def _triangular_scale_fn(self, x):
return 1.
def _triangular2_scale_fn(self, x):
return 1 /(2.**(x - 1))
def _exp_range_scale_fn(self, x):
return self.gamma**(x)
def get_lr(self):
step_size = float(self.step_size)
cycle = np.floor(1 + self.last_batch_iteration /(2 * step_size))
x = np.abs(self.last_batch_iteration / step_size - 2 * cycle + 1)
lrs = []
param_lrs = zip(self.optimizer.param_groups, self.base_lrs, self.max_lrs)
for param_group, base_lr, max_lr in param_lrs:
base_height =(max_lr - base_lr)* np.maximum(0,(1 - x))
if self.scale_mode == 'cycle':
lr = base_lr + base_height * self.scale_fn(cycle)
else:
lr = base_lr + base_height * self.scale_fn(self.last_batch_iteration)
lrs.append(lr)
return lrs
<choose_model_class> | print('Prediction accuracy for models')
ms = metrics_all[metrics_now[1]]
models.sort_values(by=[(ms + '_test'),(ms + '_train')], ascending=False ) | Titanic - Machine Learning from Disaster |
6,678,936 | embedding_dim = 300
embedding_path = '.. /save/embedding_matrix.npy'
use_pretrained_embedding = True
hidden_size = 60
gru_len = hidden_size
Routings = 4
Num_capsule = 5
Dim_capsule = 5
dropout_p = 0.25
rate_drop_dense = 0.28
LR = 0.001
T_epsilon = 1e-7
num_classes = 30
class Embed_Layer(nn.Module):
def __init__(self, embedding_matrix=None, vocab_size=None, embedding_dim=300):
super(Embed_Layer, self ).__init__()
self.encoder = nn.Embedding(vocab_size + 1, embedding_dim)
if use_pretrained_embedding:
self.encoder.weight.data.copy_(t.from_numpy(embedding_matrix))
def forward(self, x, dropout_p=0.25):
return nn.Dropout(p=dropout_p )(self.encoder(x))
class GRU_Layer(nn.Module):
def __init__(self):
super(GRU_Layer, self ).__init__()
self.gru = nn.GRU(input_size=300,
hidden_size=gru_len,
bidirectional=True)
def init_weights(self):
ih =(param.data for name, param in self.named_parameters() if 'weight_ih' in name)
hh =(param.data for name, param in self.named_parameters() if 'weight_hh' in name)
b =(param.data for name, param in self.named_parameters() if 'bias' in name)
for k in ih:
nn.init.xavier_uniform_(k)
for k in hh:
nn.init.orthogonal_(k)
for k in b:
nn.init.constant_(k, 0)
def forward(self, x):
return self.gru(x)
class Caps_Layer(nn.Module):
def __init__(self, input_dim_capsule=gru_len * 2, num_capsule=Num_capsule, dim_capsule=Dim_capsule, \
routings=Routings, kernel_size=(9, 1), share_weights=True,
activation='default', **kwargs):
super(Caps_Layer, self ).__init__(**kwargs)
self.num_capsule = num_capsule
self.dim_capsule = dim_capsule
self.routings = routings
self.kernel_size = kernel_size
self.share_weights = share_weights
if activation == 'default':
self.activation = self.squash
else:
self.activation = nn.ReLU(inplace=True)
if self.share_weights:
self.W = nn.Parameter(
nn.init.xavier_normal_(t.empty(1, input_dim_capsule, self.num_capsule * self.dim_capsule)))
else:
self.W = nn.Parameter(
t.randn(BATCH_SIZE, input_dim_capsule, self.num_capsule * self.dim_capsule))
def forward(self, x):
if self.share_weights:
u_hat_vecs = t.matmul(x, self.W)
else:
print('add later')
batch_size = x.size(0)
input_num_capsule = x.size(1)
u_hat_vecs = u_hat_vecs.view(( batch_size, input_num_capsule,
self.num_capsule, self.dim_capsule))
u_hat_vecs = u_hat_vecs.permute(0, 2, 1, 3)
b = t.zeros_like(u_hat_vecs[:, :, :, 0])
for i in range(self.routings):
b = b.permute(0, 2, 1)
c = F.softmax(b, dim=2)
c = c.permute(0, 2, 1)
b = b.permute(0, 2, 1)
outputs = self.activation(t.einsum('bij,bijk->bik',(c, u_hat_vecs)))
if i < self.routings - 1:
b = t.einsum('bik,bijk->bij',(outputs, u_hat_vecs))
return outputs
def squash(self, x, axis=-1):
s_squared_norm =(x ** 2 ).sum(axis, keepdim=True)
scale = t.sqrt(s_squared_norm + T_epsilon)
return x / scale
class Capsule_Main(nn.Module):
def __init__(self, embedding_matrix=None, vocab_size=None):
super(Capsule_Main, self ).__init__()
self.embed_layer = Embed_Layer(embedding_matrix, vocab_size)
self.gru_layer = GRU_Layer()
self.gru_layer.init_weights()
self.caps_layer = Caps_Layer()
self.dense_layer = Dense_Layer()
def forward(self, content):
content1 = self.embed_layer(content)
content2, _ = self.gru_layer(
content1)
content3 = self.caps_layer(content2)
output = self.dense_layer(content3)
return output
<normalization> | pd.options.display.float_format = '{:,.2f}'.format | Titanic - Machine Learning from Disaster |
6,678,936 | class Attention(nn.Module):
def __init__(self, feature_dim, step_dim, bias=True, **kwargs):
super(Attention, self ).__init__(**kwargs)
self.supports_masking = True
self.bias = bias
self.feature_dim = feature_dim
self.step_dim = step_dim
self.features_dim = 0
weight = torch.zeros(feature_dim, 1)
nn.init.xavier_uniform_(weight)
self.weight = nn.Parameter(weight)
if bias:
self.b = nn.Parameter(torch.zeros(step_dim))
def forward(self, x, mask=None):
feature_dim = self.feature_dim
step_dim = self.step_dim
eij = torch.mm(
x.contiguous().view(-1, feature_dim),
self.weight
).view(-1, step_dim)
if self.bias:
eij = eij + self.b
eij = torch.tanh(eij)
a = torch.exp(eij)
if mask is not None:
a = a * mask
a = a / torch.sum(a, 1, keepdim=True)+ 1e-10
weighted_input = x * torch.unsqueeze(a, -1)
return torch.sum(weighted_input, 1)
class NeuralNet(nn.Module):
def __init__(self):
super(NeuralNet, self ).__init__()
fc_layer = 16
fc_layer1 = 16
self.embedding = nn.Embedding(max_features, embed_size)
self.embedding.weight = nn.Parameter(torch.tensor(embedding_matrix, dtype=torch.float32))
self.embedding.weight.requires_grad = False
self.embedding_dropout = nn.Dropout2d(0.1)
self.lstm = nn.LSTM(embed_size, hidden_size, bidirectional=True, batch_first=True)
self.gru = nn.GRU(hidden_size * 2, hidden_size, bidirectional=True, batch_first=True)
self.lstm2 = nn.LSTM(hidden_size * 2, hidden_size, bidirectional=True, batch_first=True)
self.lstm_attention = Attention(hidden_size * 2, maxlen)
self.gru_attention = Attention(hidden_size * 2, maxlen)
self.bn = nn.BatchNorm1d(16, momentum=0.5)
self.linear = nn.Linear(hidden_size*8+3, fc_layer1)
self.relu = nn.ReLU()
self.dropout = nn.Dropout(0.1)
self.fc = nn.Linear(fc_layer**2,fc_layer)
self.out = nn.Linear(fc_layer, 1)
self.lincaps = nn.Linear(Num_capsule * Dim_capsule, 1)
self.caps_layer = Caps_Layer()
def forward(self, x):
h_embedding = self.embedding(x[0])
h_embedding = torch.squeeze(
self.embedding_dropout(torch.unsqueeze(h_embedding, 0)))
h_lstm, _ = self.lstm(h_embedding)
h_gru, _ = self.gru(h_lstm)
content3 = self.caps_layer(h_gru)
content3 = self.dropout(content3)
batch_size = content3.size(0)
content3 = content3.view(batch_size, -1)
content3 = self.relu(self.lincaps(content3))
h_lstm_atten = self.lstm_attention(h_lstm)
h_gru_atten = self.gru_attention(h_gru)
avg_pool = torch.mean(h_gru, 1)
max_pool, _ = torch.max(h_gru, 1)
f = torch.tensor(x[1], dtype=torch.float ).cuda()
conc = torch.cat(( h_lstm_atten, h_gru_atten,content3, avg_pool, max_pool,f), 1)
conc = self.relu(self.linear(conc))
conc = self.bn(conc)
conc = self.dropout(conc)
out = self.out(conc)
return out<categorify> | metrics_main = 2
xs = metrics_all[metrics_main]
xs_train = metrics_all[metrics_main] + '_train'
xs_test = metrics_all[metrics_main] + '_test'
print('The best models by the',xs,'criterion:')
direct_sort = False if(metrics_main >= 2)else True
models_sort = models.sort_values(by=[xs_test, xs_train], ascending=direct_sort ) | Titanic - Machine Learning from Disaster |
6,678,936 | class MyDataset(Dataset):
def __init__(self,dataset):
self.dataset = dataset
def __getitem__(self, index):
data, target = self.dataset[index]
return data, target, index
def __len__(self):
return len(self.dataset )<define_variables> | models_sort = models_sort[models_sort.Model != 'VotingClassifier']
models_best = models_sort[(models_sort.acc_diff < 5)&(models_sort.acc_train > 90)]
models_best[['Model', ms + '_train', ms + '_test', 'acc_diff']].sort_values(by=['acc_test'], ascending=False ) | Titanic - Machine Learning from Disaster |
6,678,936 | def sigmoid(x):
return 1 /(1 + np.exp(-x))
train_preds = np.zeros(( len(x_train)))
test_preds = np.zeros(( len(df_test)))
seed_everything()
x_test_cuda = torch.tensor(x_test, dtype=torch.long ).cuda()
test = torch.utils.data.TensorDataset(x_test_cuda)
test_loader = torch.utils.data.DataLoader(test, batch_size=batch_size, shuffle=False)
avg_losses_f = []
avg_val_losses_f = []<data_type_conversions> | models_pred = pd.DataFrame(models_best.Model, columns = ['Model'])
N_best_models = len(models_best.Model ) | Titanic - Machine Learning from Disaster |
6,678,936 | for i,(train_idx, valid_idx)in enumerate(splits):
x_train = np.array(x_train)
y_train = np.array(y_train)
features = np.array(features)
x_train_fold = torch.tensor(x_train[train_idx.astype(int)], dtype=torch.long ).cuda()
y_train_fold = torch.tensor(y_train[train_idx.astype(int), np.newaxis], dtype=torch.float32 ).cuda()
kfold_X_features = features[train_idx.astype(int)]
kfold_X_valid_features = features[valid_idx.astype(int)]
x_val_fold = torch.tensor(x_train[valid_idx.astype(int)], dtype=torch.long ).cuda()
y_val_fold = torch.tensor(y_train[valid_idx.astype(int), np.newaxis], dtype=torch.float32 ).cuda()
model = NeuralNet()
model.cuda()
loss_fn = torch.nn.BCEWithLogitsLoss(reduction='sum')
step_size = 300
base_lr, max_lr = 0.001, 0.003
optimizer = torch.optim.Adam(filter(lambda p: p.requires_grad, model.parameters()),
lr=max_lr)
scheduler = CyclicLR(optimizer, base_lr=base_lr, max_lr=max_lr,
step_size=step_size, mode='exp_range',
gamma=0.99994)
train = torch.utils.data.TensorDataset(x_train_fold, y_train_fold)
valid = torch.utils.data.TensorDataset(x_val_fold, y_val_fold)
train = MyDataset(train)
valid = MyDataset(valid)
train_loader = torch.utils.data.DataLoader(train, batch_size=batch_size, shuffle=True)
valid_loader = torch.utils.data.DataLoader(valid, batch_size=batch_size, shuffle=False)
print(f'Fold {i + 1}')
for epoch in range(n_epochs):
start_time = time.time()
model.train()
avg_loss = 0.
for i,(x_batch, y_batch, index)in enumerate(train_loader):
f = kfold_X_features[index]
y_pred = model([x_batch,f])
if scheduler:
scheduler.batch_step()
loss = loss_fn(y_pred, y_batch)
optimizer.zero_grad()
loss.backward()
optimizer.step()
avg_loss += loss.item() / len(train_loader)
model.eval()
valid_preds_fold = np.zeros(( x_val_fold.size(0)))
test_preds_fold = np.zeros(( len(df_test)))
avg_val_loss = 0.
for i,(x_batch, y_batch, index)in enumerate(valid_loader):
f = kfold_X_valid_features[index]
y_pred = model([x_batch,f] ).detach()
avg_val_loss += loss_fn(y_pred, y_batch ).item() / len(valid_loader)
valid_preds_fold[i * batch_size:(i+1)* batch_size] = sigmoid(y_pred.cpu().numpy())[:, 0]
elapsed_time = time.time() - start_time
print('Epoch {}/{} \t loss={:.4f} \t val_loss={:.4f} \t time={:.2f}s'.format(
epoch + 1, n_epochs, avg_loss, avg_val_loss, elapsed_time))
avg_losses_f.append(avg_loss)
avg_val_losses_f.append(avg_val_loss)
for i,(x_batch,)in enumerate(test_loader):
f = test_features[i * batch_size:(i+1)* batch_size]
y_pred = model([x_batch,f] ).detach()
test_preds_fold[i * batch_size:(i+1)* batch_size] = sigmoid(y_pred.cpu().numpy())[:, 0]
train_preds[valid_idx] = valid_preds_fold
test_preds += test_preds_fold / len(splits)
print('All \t loss={:.4f} \t val_loss={:.4f} \t '.format(np.average(avg_losses_f),np.average(avg_val_losses_f)))
<compute_test_metric> | def model_fit(name_model,train,target):
if name_model == 'LGBMClassifier':
Xtrain, Xval, Ztrain, Zval = train_test_split(train, target, test_size=test_train_split_part, random_state=random_state)
model = lgb.LGBMClassifier(n_estimators=1000)
model.fit(Xtrain, Ztrain, eval_set=[(Xval, Zval)], early_stopping_rounds=50, verbose=False)
else:
param_grid={}
if name_model == 'Linear Regression':
model_clf = LinearRegression()
elif name_model == 'Support Vector Machines':
model_clf = SVC()
param_grid = {'kernel': ['linear', 'poly', 'rbf', 'sigmoid'], 'tol': [1e-4]}
elif name_model == 'Linear SVC':
model_clf = LinearSVC()
param_grid = {'dual':[False],
'C': np.linspace(1, 15, 15)}
elif name_model == 'MLPClassifier':
model_clf = MLPClassifier()
param_grid = {'hidden_layer_sizes': [i for i in range(2,5)],
'solver': ['sgd'],
'learning_rate': ['adaptive'],
'max_iter': [1000]
}
elif name_model == 'Stochastic Gradient Decent':
model_clf = SGDClassifier(early_stopping=True)
param_grid = {'alpha': [0.0001, 0.001, 0.01, 0.1, 1]}
elif name_model == 'Decision Tree Classifier':
model_clf = DecisionTreeClassifier()
param_grid = {'min_samples_leaf': [i for i in range(2,10)]}
elif name_model == 'Random Forest':
model_clf = RandomForestClassifier()
param_grid = {'n_estimators': [300, 400, 500, 600], 'min_samples_split': [60], 'min_samples_leaf': [20, 25, 30, 35, 40],
'max_features': ['auto'], 'max_depth': [5, 6, 7, 8, 9, 10], 'criterion': ['gini'], 'bootstrap': [False]}
elif name_model == 'XGBClassifier':
model_clf = xgb.XGBClassifier(objective='reg:squarederror')
param_grid = {'n_estimators': [200, 300, 400],
'learning_rate': [0.001, 0.003, 0.005, 0.006, 0.01],
'max_depth': [4, 5, 6]}
elif name_model == 'GradientBoostingClassifier':
model_clf = GradientBoostingClassifier()
param_grid = {'learning_rate' : [0.001, 0.01, 0.1],
'max_depth': [i for i in range(2,5)],
'min_samples_leaf': [i for i in range(2,5)]}
elif name_model == 'RidgeClassifier':
model_clf = RidgeClassifier()
param_grid={'alpha': np.linspace (.1, 1.5, 15)}
elif name_model == 'BaggingClassifier':
model_base_estimator = GridSearchCV(LinearSVC() , param_grid={'dual':[False], 'C': np.linspace(1, 15, 15)},
cv=cv_train, verbose=False)
model_base_estimator.fit(train, target)
model_clf = BaggingClassifier(base_estimator=model_base_estimator)
param_grid={'max_features': [0.5, 0.6, 0.7, 0.8, 0.9, 1.0],
'n_estimators': [3, 5, 10],
'warm_start' : [True],
'random_state': [random_state]}
elif name_model == 'ExtraTreesClassifier':
model_clf = ExtraTreesClassifier()
param_grid={'min_samples_leaf' : [10, 20, 30, 40, 50]}
elif name_model == 'AdaBoostClassifier':
model_clf = AdaBoostClassifier()
param_grid={'learning_rate' : [.01,.1,.5, 1]}
elif name_model == 'Logistic Regression':
model_clf = LogisticRegression()
param_grid={'C' : [.1,.3,.5,.7, 1]}
elif name_model == 'KNN':
model_clf = KNeighborsClassifier()
param_grid={'n_neighbors': range(2, 7)}
elif name_model == 'Naive Bayes':
model_clf = GaussianNB()
param_grid={'var_smoothing': [1e-8, 1e-9, 1e-10]}
elif name_model == 'Perceptron':
model_clf = Perceptron()
param_grid = {'penalty': [None, 'l2', 'l1', 'elasticnet']}
elif name_model == 'Gaussian Process Classification':
model_clf = GaussianProcessClassifier()
param_grid = {'max_iter_predict': [100, 200],
'warm_start': [True, False],
'n_restarts_optimizer': range(3)}
model = GridSearchCV(model_clf, param_grid=param_grid, cv=cv_train, verbose=False)
model.fit(train, target)
return model | Titanic - Machine Learning from Disaster |
6,678,936 | def bestThresshold(y_train,train_preds):
tmp = [0,0,0]
delta = 0
for tmp[0] in tqdm(np.arange(0.1, 0.501, 0.01)) :
tmp[1] = f1_score(y_train, np.array(train_preds)>tmp[0])
if tmp[1] > tmp[2]:
delta = tmp[0]
tmp[2] = tmp[1]
print('best threshold is {:.4f} with F1 score: {:.4f}'.format(delta, tmp[2]))
return delta
delta = bestThresshold(y_train,train_preds )<save_to_csv> | for i in range(N_best_models):
name_model = models_best.iloc[i]['Model']
if(name_model == 'LGBMClassifier')or(name_model == 'XGBClassifier'):
model = model_fit(name_model,train0b,target0)
acc_metrics_calc_pred(i,model,name_model,train0b,test0b,target0)
else:
model = model_fit(name_model,train0,target0)
acc_metrics_calc_pred(i,model,name_model,train0,test0,target0 ) | Titanic - Machine Learning from Disaster |
6,678,936 | <set_options><EOS> | for x in metrics_now:
xs = metrics_all[x]
models_pred[xs + '_train'] = acc_all_pred[(x-1)]
models_pred[['Model', 'acc_train']].sort_values(by=['acc_train'], ascending=False ) | Titanic - Machine Learning from Disaster |
965,330 | <SOS> metric: categorizationaccuracy Kaggle data source: titanic-machine-learning-from-disaster<load_from_csv> | import numpy as np
import pandas as pd
from catboost import CatBoostClassifier, Pool, cv
import hyperopt | Titanic - Machine Learning from Disaster |
965,330 | EMBEDDING_FILE = '.. /input/embeddings/glove.840B.300d/glove.840B.300d.txt'
train = pd.read_csv('.. /input/train.csv')
test = pd.read_csv('.. /input/test.csv')
submission = pd.read_csv('.. /input/sample_submission.csv' )<string_transform> | train = pd.read_csv('.. /input/train.csv')
test = pd.read_csv('.. /input/test.csv')
train_size = train.shape[0]
test_size = test.shape[0]
data = pd.concat([train, test] ) | Titanic - Machine Learning from Disaster |
965,330 | X_train = train["question_text"].fillna("fillna" ).values
y_train = train["target"].values
X_test = test["question_text"].fillna("fillna" ).values
max_features = 40000
maxlen = 50
embed_size = 300
tokenizer = text.Tokenizer(num_words=max_features)
tokenizer.fit_on_texts(list(X_train)+ list(X_test))
X_train = tokenizer.texts_to_sequences(X_train)
X_test = tokenizer.texts_to_sequences(X_test)
x_train = sequence.pad_sequences(X_train, maxlen=maxlen)
x_test = sequence.pad_sequences(X_test, maxlen=maxlen )<feature_engineering> | data['Title'] = data['Name'].str.extract('([A-Za-z]+)\.', expand=False ) | Titanic - Machine Learning from Disaster |
965,330 | def get_coefs(word, *arr): return word, np.asarray(arr, dtype='float32')
embeddings_index = dict(get_coefs(*o.rstrip().rsplit(' ')) for o in open(EMBEDDING_FILE))
word_index = tokenizer.word_index
nb_words = min(max_features, len(word_index))
embedding_matrix = np.zeros(( nb_words, embed_size))
for word, i in word_index.items() :
if i >= max_features: continue
embedding_vector = embeddings_index.get(word)
if embedding_vector is not None: embedding_matrix[i] = embedding_vector<predict_on_test> | age_ref = data.groupby('Title' ).Age.mean()
data['Age'] = data.apply(lambda r: r.Age if pd.notnull(r.Age)else age_ref[r.Title] , axis=1)
del age_ref | Titanic - Machine Learning from Disaster |
965,330 | class F1Evaluation(Callback):
def __init__(self, validation_data=() , interval=1):
super(Callback, self ).__init__()
self.interval = interval
self.X_val, self.y_val = validation_data
def on_epoch_end(self, epoch, logs={}):
if epoch % self.interval == 0:
y_pred = self.model.predict(self.X_val, verbose=0)
y_pred =(y_pred > 0.35 ).astype(int)
score = f1_score(self.y_val, y_pred)
print("
F1 Score - epoch: %d - score: %.6f
" %(epoch+1, score))<choose_model_class> | data.loc[(data.PassengerId==1044, 'Fare')] = 14.43 | Titanic - Machine Learning from Disaster |
965,330 | filter_sizes = [1,2,3,5]
num_filters = 36
def get_model() :
inp = Input(shape=(maxlen,))
x = Embedding(max_features, embed_size, weights=[embedding_matrix] )(inp)
x = SpatialDropout1D(0.4 )(x)
conv_0 = Conv1D(num_filters, kernel_size=(filter_sizes[0]),
kernel_initializer='he_normal', activation='elu' )(x)
conv_1 = Conv1D(num_filters, kernel_size=(filter_sizes[1]),
kernel_initializer='he_normal', activation='elu' )(x)
conv_2 = Conv1D(num_filters, kernel_size=(filter_sizes[2]),
kernel_initializer='he_normal', activation='elu' )(x)
conv_3 = Conv1D(num_filters, kernel_size=(filter_sizes[3]),
kernel_initializer='he_normal', activation='elu' )(x)
maxpool_0 = MaxPool1D(pool_size=(maxlen - filter_sizes[0] + 1))(conv_0)
maxpool_1 = MaxPool1D(pool_size=(maxlen - filter_sizes[1] + 1))(conv_1)
maxpool_2 = MaxPool1D(pool_size=(maxlen - filter_sizes[2] + 1))(conv_2)
maxpool_3 = MaxPool1D(pool_size=(maxlen - filter_sizes[3] + 1))(conv_3)
z = Concatenate(axis=1 )([maxpool_0, maxpool_1, maxpool_2, maxpool_3])
z = Flatten()(z)
z = BatchNormalization()(z)
outp = Dense(1, activation="sigmoid" )(z)
model = Model(inputs=inp, outputs=outp)
model.compile(loss='binary_crossentropy',
optimizer='adam',
metrics=['accuracy'])
return model
model = get_model()<train_model> | data['Embarked'] = data['Embarked'].fillna('S')
data['Cabin'] = data['Cabin'].fillna('Undefined' ) | Titanic - Machine Learning from Disaster |
965,330 | batch_size = 1024
epochs = 4
X_tra, X_val, y_tra, y_val = train_test_split(x_train, y_train, train_size=0.95,
random_state=233)
F1_Score = F1Evaluation(validation_data=(X_val, y_val), interval=1)
hist = model.fit(X_tra, y_tra, batch_size=batch_size, epochs=epochs,
validation_data=(X_val, y_val),
callbacks=[F1_Score], verbose=True )<choose_model_class> | cols = [
'Pclass',
'Name',
'Sex',
'Age',
'SibSp',
'Parch',
'Ticket',
'Fare',
'Cabin',
'Embarked'
]
X_train = data[:train_size][cols]
Y_train = data[:train_size]['Survived'].astype(int)
X_test = data[train_size:][cols]
categorical_features_indices = [0,1,2,6,8,9]
X_train.head() | Titanic - Machine Learning from Disaster |
965,330 | filter_sizes = [1,2,3,5]
num_filters = 36
def get_model() :
inp = Input(shape=(maxlen,))
x = Lambda(lambda x: K.reverse(x,axes=-1))(inp)
x = Embedding(max_features, embed_size, weights=[embedding_matrix] )(x)
x = SpatialDropout1D(0.4 )(x)
conv_0 = Conv1D(num_filters, kernel_size=(filter_sizes[0]),
kernel_initializer='he_normal', activation='elu' )(x)
conv_1 = Conv1D(num_filters, kernel_size=(filter_sizes[1]),
kernel_initializer='he_normal', activation='elu' )(x)
conv_2 = Conv1D(num_filters, kernel_size=(filter_sizes[2]),
kernel_initializer='he_normal', activation='elu' )(x)
conv_3 = Conv1D(num_filters, kernel_size=(filter_sizes[3]),
kernel_initializer='he_normal', activation='elu' )(x)
maxpool_0 = MaxPool1D(pool_size=(maxlen - filter_sizes[0] + 1))(conv_0)
maxpool_1 = MaxPool1D(pool_size=(maxlen - filter_sizes[1] + 1))(conv_1)
maxpool_2 = MaxPool1D(pool_size=(maxlen - filter_sizes[2] + 1))(conv_2)
maxpool_3 = MaxPool1D(pool_size=(maxlen - filter_sizes[3] + 1))(conv_3)
z = Concatenate(axis=1 )([maxpool_0, maxpool_1, maxpool_2, maxpool_3])
z = Flatten()(z)
z = BatchNormalization()(z)
outp = Dense(1, activation="sigmoid" )(z)
model = Model(inputs=inp, outputs=outp)
model.compile(loss='binary_crossentropy',
optimizer='adam',
metrics=['accuracy'])
return model
model_flip = get_model()
<train_model> | train_pool = Pool(X_train, Y_train, cat_features=categorical_features_indices ) | Titanic - Machine Learning from Disaster |
965,330 | hist_flip = model_flip.fit(X_tra, y_tra, batch_size=batch_size, epochs=epochs,
validation_data=(X_val, y_val),
callbacks=[F1_Score], verbose=True )<predict_on_test> | Titanic - Machine Learning from Disaster | |
965,330 | val_y_pred1 = model.predict(X_val, batch_size=1024, verbose = True)
val_y_pred2 = model_flip.predict(X_val, batch_size=1024, verbose = True)
<compute_test_metric> | model = CatBoostClassifier(
depth=3,
iterations=300,
eval_metric='Accuracy',
random_seed=42,
logging_level='Silent',
allow_writing_files=False
)
cv_data = cv(
train_pool,
model.get_params() ,
fold_count=5
)
print('Best validation accuracy score: {:.2f}±{:.2f} on step {}'.format(
np.max(cv_data['test-Accuracy-mean']),
cv_data['test-Accuracy-std'][cv_data['test-Accuracy-mean'].idxmax(axis=0)],
cv_data['test-Accuracy-mean'].idxmax(axis=0)
))
print('Precise validation accuracy score: {}'.format(np.max(cv_data['test-Accuracy-mean'])))
model.fit(train_pool);
model.score(X_train, Y_train ) | Titanic - Machine Learning from Disaster |
965,330 | val_y_pred = np.mean([val_y_pred1,val_y_pred2],axis = 0 )<predict_on_test> | feature_importances = model.get_feature_importance(train_pool)
feature_names = X_train.columns
for score, name in sorted(zip(feature_importances, feature_names), reverse=True):
print('{}: {}'.format(name, score)) | Titanic - Machine Learning from Disaster |
965,330 | <compute_test_metric><EOS> | Y_pred = model.predict(X_test)
submission = pd.DataFrame({
"PassengerId": data[train_size:]["PassengerId"],
"Survived": Y_pred.astype(int)
})
submission.to_csv('submission.csv', index=False ) | Titanic - Machine Learning from Disaster |
737,908 | <SOS> metric: categorizationaccuracy Kaggle data source: titanic-machine-learning-from-disaster<compute_test_metric> | warnings.filterwarnings('ignore' ) | Titanic - Machine Learning from Disaster |
737,908 | best_threshold = 0.01
best_score = 0.0
for threshold in range(1, 100):
threshold = threshold / 100
score = f1_score(y_val, val_y_pred > threshold)
if score > best_score:
best_threshold = threshold
best_score = score
print("Score at threshold=0.5 is {}".format(f1_score(y_val, val_y_pred > 0.5)))
print("Optimal threshold is {} with a score of {}".format(best_threshold, best_score))<save_to_csv> | train_set = pd.read_csv('.. /input/train.csv')
test_set = pd.read_csv('.. /input/test.csv')
train_set.shape, test_set.shape | Titanic - Machine Learning from Disaster |
737,908 | y_pred =(y_pred > best_threshold ).astype(int)
submission['prediction'] = y_pred
submission.to_csv('submission.csv', index=False )<set_options> | full_set = pd.concat([train_set, test_set])
full_set.head() | Titanic - Machine Learning from Disaster |
737,908 | start = time.time()
seed = 32
os.environ['PYTHONHASHSEED'] = str(seed)
os.environ['OMP_NUM_THREADS'] = '4'
np.random.seed(seed)
rn.seed(seed)
session_conf = tf.ConfigProto(intra_op_parallelism_threads = 1,
inter_op_parallelism_threads = 1)
tf.set_random_seed(seed)
sess = tf.Session(graph = tf.get_default_graph() , config = session_conf)
K.set_session(sess)
train = pd.read_csv(".. /input/train.csv" ).fillna("missing")
test = pd.read_csv(".. /input/test.csv" ).fillna("missing")
embedding_file1 = ".. /input/embeddings/glove.840B.300d/glove.840B.300d.txt"
embedding_file2 = ".. /input/embeddings/paragram_300_sl999/paragram_300_sl999.txt"
embed_size = 300
max_features = 100000
max_len = 60<define_variables> | full_set['Age'][full_set['Age'].isnull() ] = full_set['Age'].median()
full_set['Age'] = full_set['Age'].astype(int ) | Titanic - Machine Learning from Disaster |
737,908 | puncts = [',', '.', '"', ':', ')', '(', '-', '!', '?', '|', ';', "'", '$', '&',
'/', '[', ']', '>', '%', '=', '
'·', '_', '{', '}', '©', '^', '®', '`', '<', '→', '°', '€', '™', '›',
'♥', '←', '×', '§', '″', '′', 'Â', '█', '½', 'à', '…', '“', '★', '”',
'–', '●', 'â', '►', '−', '¢', '²', '¬', '░', '¶', '↑', '±', '¿', '▾',
'═', '¦', '║', '―', '¥', '▓', '—', '‹', '─', '▒', '：', '¼', '⊕', '▼',
'▪', '†', '■', '’', '▀', '¨', '▄', '♫', '☆', 'é', '¯', '♦', '¤', '▲',
'è', '¸', '¾', 'Ã', '⋅', '‘', '∞', '∙', '）', '↓', '、', '│', '（', '»',
'，', '♪', '╩', '╚', '³', '・', '╦', '╣', '╔', '╗', '▬', '❤', 'ï', 'Ø',
'¹', '≤', '‡', '√', 'β', 'α', '∅', 'θ', '÷', '₹']
def clean_punct(x):
x = str(x)
for punct in puncts:
if punct in x:
x = x.replace(punct, f' {punct} ')
return x
train["question_text"] = train["question_text"].apply(lambda x: clean_punct(x))
test["question_text"] = test["question_text"].apply(lambda x: clean_punct(x))<define_variables> | full_set['HasCabin'] = full_set['Cabin'].apply(lambda x: 0 if isinstance(x, float)else 1 ) | Titanic - Machine Learning from Disaster |
737,908 | sincere = train[train["target"] == 0]
insincere = train[train["target"] == 1]
print("Sincere questions {}; Insincere questions {}".format(sincere.shape[0], insincere.shape[0]))<compute_train_metric> | full_set['Embarked'][full_set['Embarked'].isnull() ] = 'S'
full_set['Embarked'] = full_set['Embarked'].map({'S': 0, 'C': 1, 'Q': 2} ).astype(int ) | Titanic - Machine Learning from Disaster |
737,908 | def get_glove(embedding_file):
def get_coefs(word,*arr): return word, np.asarray(arr, dtype='float32')
embeddings_index = dict(get_coefs(*o.split(" ")) for o in open(embedding_file))
all_embs = np.stack(embeddings_index.values())
emb_mean, emb_std = all_embs.mean() , all_embs.std()
return embeddings_index, emb_mean, emb_std
def get_para(embedding_file):
def get_coefs(word,*arr): return word, np.asarray(arr, dtype='float32')
embeddings_index = dict(get_coefs(*o.split(" ")) for o in open(embedding_file,
encoding="utf8",
errors='ignore')if len(o)>100)
all_embs = np.stack(embeddings_index.values())
emb_mean, emb_std = all_embs.mean() , all_embs.std()
return embeddings_index, emb_mean, emb_std
glove_index, glove_mean, glove_std = get_glove(embedding_file1)
para_index, para_mean, para_std = get_para(embedding_file2 )<categorify> | full_set[ full_set['Fare'].isnull() ] | Titanic - Machine Learning from Disaster |
737,908 | def get_embed(tokenizer = None, embeddings_index = None, emb_mean = None, emb_std = None):
word_index = tokenizer.word_index
nb_words = min(max_features, len(word_index))
embedding_matrix = np.random.normal(emb_mean, emb_std,(nb_words, embed_size))
for word, i in word_index.items() :
if i >= max_features: continue
embedding_vector = embeddings_index.get(word)
if embedding_vector is not None: embedding_matrix[i] = embedding_vector
return nb_words, embedding_matrix<string_transform> | full_set['Fare'][ full_set['Fare'].isnull() ] = 0 . | Titanic - Machine Learning from Disaster |
737,908 | tokenizer = Tokenizer(num_words = max_features, lower = True)
tokenizer.fit_on_texts(train["question_text"])
train_token = tokenizer.texts_to_sequences(train["question_text"])
fake_test_token = tokenizer.texts_to_sequences(fake_test["question_text"])
test_token = tokenizer.texts_to_sequences(test["question_text"])
train_seq = pad_sequences(train_token, maxlen = max_len)
fake_test_seq = pad_sequences(fake_test_token, maxlen = max_len)
X_test = pad_sequences(test_token, maxlen = max_len)
del train_token, fake_test_token, test_token; gc.collect()<load_pretrained> | full_set['Title'] = full_set['Name'].apply(get_title)
print(full_set['Title'].unique() ) | Titanic - Machine Learning from Disaster |
737,908 | nb_words, embedding_matrix1 = get_embed(tokenizer = tokenizer, embeddings_index = glove_index,
emb_mean = glove_mean,
emb_std = glove_std)
nb_words, embedding_matrix2 = get_embed(tokenizer = tokenizer, embeddings_index = para_index,
emb_mean = para_mean,
emb_std = para_std)
embedding_matrix = np.mean([embedding_matrix1, embedding_matrix2], axis = 0)
del embedding_matrix1, embedding_matrix2; gc.collect()
print("Embedding matrix completed!" )<choose_model_class> | commons = ['Mr','Mrs','Miss','Mme','Ms','Mlle']
rares = list(set(full_set['Title'].unique())- set(commons))
full_set['Title'] = full_set['Title'].replace('Ms','Miss')
full_set['Title'] = full_set['Title'].replace('Mlle','Miss')
full_set['Title'] = full_set['Title'].replace('Mme','Mrs')
full_set['Title'][full_set['Title'].isin(rares)] = 'Rare'
full_set['Title'] = full_set['Title'].map({'Mr':0, 'Mrs':1, 'Miss':2, 'Rare':3} ) | Titanic - Machine Learning from Disaster |
737,908 | class Attention(Layer):
def __init__(self, step_dim,
W_regularizer=None, b_regularizer=None,
W_constraint=None, b_constraint=None,
bias=True, **kwargs):
self.supports_masking = True
self.init = initializers.get('glorot_uniform')
self.W_regularizer = regularizers.get(W_regularizer)
self.b_regularizer = regularizers.get(b_regularizer)
self.W_constraint = constraints.get(W_constraint)
self.b_constraint = constraints.get(b_constraint)
self.bias = bias
self.step_dim = step_dim
self.features_dim = 0
super(Attention, self ).__init__(**kwargs)
def build(self, input_shape):
assert len(input_shape)== 3
self.W = self.add_weight(( input_shape[-1],),
initializer=self.init,
name='{}_W'.format(self.name),
regularizer=self.W_regularizer,
constraint=self.W_constraint)
self.features_dim = input_shape[-1]
if self.bias:
self.b = self.add_weight(( input_shape[1],),
initializer='zero',
name='{}_b'.format(self.name),
regularizer=self.b_regularizer,
constraint=self.b_constraint)
else:
self.b = None
self.built = True
def compute_mask(self, input, input_mask=None):
return None
def call(self, x, mask=None):
features_dim = self.features_dim
step_dim = self.step_dim
eij = K.reshape(K.dot(K.reshape(x,(-1, features_dim)) ,
K.reshape(self.W,(features_dim, 1))),(-1, step_dim))
if self.bias:
eij += self.b
eij = K.tanh(eij)
a = K.exp(eij)
if mask is not None:
a *= K.cast(mask, K.floatx())
a /= K.cast(K.sum(a, axis=1, keepdims=True)+ K.epsilon() , K.floatx())
a = K.expand_dims(a)
weighted_input = x * a
return K.sum(weighted_input, axis=1)
def compute_output_shape(self, input_shape):
return input_shape[0], self.features_dim<find_best_params> | full_set['FamSize'] = full_set['Parch'] + full_set['SibSp'] + 1
| Titanic - Machine Learning from Disaster |
737,908 | def get_f1(true, val):
precision, recall, thresholds = precision_recall_curve(true, val)
thresholds = np.append(thresholds, 1.001)
F = 2 /(1/precision + 1/recall)
best_score = np.max(F)
best_threshold = thresholds[np.argmax(F)]
return best_threshold, best_score<choose_model_class> | full_set['Sex'] = full_set['Sex'].map({'male':0, 'female':1} ) | Titanic - Machine Learning from Disaster |
737,908 | def build_model(units = 40, dr = 0.3):
inp = Input(shape =(max_len,))
embed_layer = Embedding(nb_words, embed_size, input_length = max_len,
weights = [embedding_matrix], trainable = False )(inp)
x = SpatialDropout1D(dr, seed = seed )(embed_layer)
x = Bidirectional(CuDNNLSTM(units, kernel_initializer = glorot_normal(seed = seed),
recurrent_initializer = orthogonal(gain = 1.0, seed = seed),
return_sequences = True))(x)
x = Bidirectional(CuDNNGRU(units, kernel_initializer = glorot_normal(seed = seed),
recurrent_initializer = orthogonal(gain = 1.0, seed = seed),
return_sequences = True))(x)
att = Attention(max_len )(x)
avg_pool = GlobalAveragePooling1D()(x)
max_pool = GlobalMaxPooling1D()(x)
main = concatenate([att, avg_pool, max_pool])
main = Dense(64, kernel_initializer = glorot_normal(seed = seed))(main)
main = Activation("relu" )(main)
main = Dropout(0.1, seed = seed )(main)
out = Dense(1, activation = "sigmoid",
kernel_initializer = glorot_normal(seed = seed))(main)
model = Model(inputs = inp, outputs = out)
model.compile(loss = "binary_crossentropy",
optimizer = Adam() ,
metrics = None)
return model<split> | full_set.drop(['Cabin','Name','Parch','PassengerId','SibSp','Ticket'], axis=1, inplace=True ) | Titanic - Machine Learning from Disaster |
737,908 | fold = 5
batch_size = 1024
epochs = 5
oof_pred = np.zeros(( train.shape[0], 1))
pred = np.zeros(( test_shape[0], 1))
fake_pred = np.zeros(( test_shape[0], 1))
thresholds = []
k_fold = StratifiedKFold(n_splits = fold, random_state = seed, shuffle = True)
for i,(train_idx, val_idx)in enumerate(k_fold.split(train_seq, target)) :
print("-"*50)
print("Trainging fold {}/{}".format(i+1, fold))
X_train, y_train = train_seq[train_idx], target[train_idx]
X_val, y_val = train_seq[val_idx], target[val_idx]
K.clear_session()
model = build_model(units = 60)
model.fit(X_train, y_train, batch_size = batch_size, epochs = epochs,
validation_data =(X_val, y_val), verbose = 2)
val_pred = model.predict(X_val, batch_size = batch_size)
oof_pred[val_idx] = val_pred
fake_pred += model.predict(fake_test_seq, batch_size = batch_size)/fold
pred += model.predict(X_test, batch_size = batch_size)/fold
threshold, score = get_f1(y_val, val_pred)
print("F1 score at threshold {} is {}".format(threshold, score))<compute_test_metric> | def fare_bin(fare):
if fare <= 7.8958:
return 0.
elif 7.8958 < fare <= 14.4542:
return 1.
elif 14.4542 < fare <= 31.2750:
return 2.
else:
return 3 . | Titanic - Machine Learning from Disaster |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.