01
import torch
02
from torch import nn
03
model = ResNet(blocks=4)
04
loss = criterion(logits, y)
05
loss.backward()
06
optimizer.step()
07
model.fit(X_train, y_train)
08
preds = model.predict(X_test)
09
X = np.array([[...], [...], [...]])
10
git commit -m "improve baseline"
11
acc = accuracy_score(y, preds)
12
with torch.no_grad():
01
import torch
02
from torch import nn
03
model = ResNet(blocks=4)
04
loss = criterion(logits, y)
05
loss.backward()
06
optimizer.step()
07
model.fit(X_train, y_train)
08
preds = model.predict(X_test)
09
X = np.array([[...], [...], [...]])
10
git commit -m "improve baseline"
11
acc = accuracy_score(y, preds)
12
with torch.no_grad():
01
import torch
02
from torch import nn
03
model = ResNet(blocks=4)
04
loss = criterion(logits, y)
05
loss.backward()
06
optimizer.step()
07
model.fit(X_train, y_train)
08
preds = model.predict(X_test)
09
X = np.array([[...], [...], [...]])
10
git commit -m "improve baseline"
11
acc = accuracy_score(y, preds)
12
with torch.no_grad():
01
import torch
02
from torch import nn
03
model = ResNet(blocks=4)
04
loss = criterion(logits, y)
05
loss.backward()
06
optimizer.step()
07
model.fit(X_train, y_train)
08
preds = model.predict(X_test)
09
X = np.array([[...], [...], [...]])
10
git commit -m "improve baseline"
11
acc = accuracy_score(y, preds)
12
with torch.no_grad():
{ }
< />
( )
[ ]
=>
#
;
0 1
∑
∇
λ
θ
∂
σ
λθ
01
PB
Prachya Biswas
Home
Experience
Achievements
Skills
Projects
More
Talks and Features
[event name]
[host]
Back to Talks