Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Added dataset in JSON format #6

Open
wants to merge 4 commits into
base: master
Choose a base branch
from
Open

Added dataset in JSON format #6

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
7ac5941
Added dataset in JSON format
Jun 3, 2020
9ce6419
Fixed indentation for generate_input_long_history2 and various refact…
Jun 8, 2020
63cffc8
Merge pull request #1 from gunarto90/py3
Jun 8, 2020
d79592a
Modified for compatibility with Python 3
Jun 9, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
5 changes: 4 additions & 1 deletion codes/main.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -91,7 +91,10 @@ def run(args):
len([y for x in test_idx for y in test_idx[x]])))
SAVE_PATH = args.save_path
tmp_path = 'checkpoint/'
os.mkdir(SAVE_PATH + tmp_path)
try:
os.mkdir(SAVE_PATH + tmp_path)
except:
pass
for epoch in range(parameters.epoch):
st = time.time()
if args.pretrain == 0:
Expand Down
28 changes: 14 additions & 14 deletions codes/model.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,11 +47,11 @@ def init_weights(self):
b = (param.data for name, param in self.named_parameters() if 'bias' in name)

for t in ih:
nn.init.xavier_uniform(t)
nn.init.xavier_uniform_(t)
for t in hh:
nn.init.orthogonal(t)
nn.init.orthogonal_(t)
for t in b:
nn.init.constant(t, 0)
nn.init.constant_(t, 0)

def forward(self, loc, tim):
h1 = Variable(torch.zeros(1, 1, self.hidden_size))
Expand All @@ -74,7 +74,7 @@ def forward(self, loc, tim):
out = self.dropout(out)

y = self.fc(out)
score = F.log_softmax(y) # calculate loss by NLLoss
score = F.log_softmax(y, dim=1) # calculate loss by NLLoss
return score


Expand Down Expand Up @@ -102,7 +102,7 @@ def forward(self, out_state, history):
for i in range(state_len):
for j in range(seq_len):
attn_energies[i, j] = self.score(out_state[i], history[j])
return F.softmax(attn_energies)
return F.softmax(attn_energies, dim=1)

def score(self, hidden, encoder_output):
if self.method == 'dot':
Expand Down Expand Up @@ -163,11 +163,11 @@ def init_weights(self):
b = (param.data for name, param in self.named_parameters() if 'bias' in name)

for t in ih:
nn.init.xavier_uniform(t)
nn.init.xavier_uniform_(t)
for t in hh:
nn.init.orthogonal(t)
nn.init.orthogonal_(t)
for t in b:
nn.init.constant(t, 0)
nn.init.constant_(t, 0)

def forward(self, loc, tim, history_loc, history_tim, history_count, uid, target_len):
h1 = Variable(torch.zeros(1, 1, self.hidden_size))
Expand Down Expand Up @@ -196,7 +196,7 @@ def forward(self, loc, tim, history_loc, history_tim, history_count, uid, target
count += c

history = torch.cat((loc_emb_history2, tim_emb_history2), 1)
history = F.tanh(self.fc_attn(history))
history = torch.tanh(self.fc_attn(history))

if self.rnn_type == 'GRU' or self.rnn_type == 'RNN':
out_state, h1 = self.rnn(x, h1)
Expand All @@ -214,7 +214,7 @@ def forward(self, loc, tim, history_loc, history_tim, history_count, uid, target
out = self.dropout(out)

y = self.fc_final(out)
score = F.log_softmax(y)
score = F.log_softmax(y, dim=1)

return score

Expand Down Expand Up @@ -263,11 +263,11 @@ def init_weights(self):
b = (param.data for name, param in self.named_parameters() if 'bias' in name)

for t in ih:
nn.init.xavier_uniform(t)
nn.init.xavier_uniform_(t)
for t in hh:
nn.init.orthogonal(t)
nn.init.orthogonal_(t)
for t in b:
nn.init.constant(t, 0)
nn.init.constant_(t, 0)

def forward(self, loc, tim, target_len):
h1 = Variable(torch.zeros(1, 1, self.hidden_size))
Expand Down Expand Up @@ -300,6 +300,6 @@ def forward(self, loc, tim, target_len):
out = self.dropout(out)

y = self.fc_final(out)
score = F.log_softmax(y)
score = F.log_softmax(y, dim=1)

return score