Spaces:
Sleeping
Sleeping
Update preprocess_test.py
Browse files- preprocess_test.py +7 -3
preprocess_test.py
CHANGED
|
@@ -136,7 +136,8 @@ class Preprocess_Test:
|
|
| 136 |
BATCH_SIZE = 256
|
| 137 |
correct = 0
|
| 138 |
total = 0
|
| 139 |
-
|
|
|
|
| 140 |
with torch.no_grad():
|
| 141 |
for i in range(0, len(X_test_t), BATCH_SIZE):
|
| 142 |
batch_x = X_test_t[i:i + BATCH_SIZE].to(self.device)
|
|
@@ -146,12 +147,15 @@ class Preprocess_Test:
|
|
| 146 |
predicted = torch.argmax(outputs, dim=1)
|
| 147 |
total += batch_y.size(0)
|
| 148 |
correct += (predicted == batch_y).sum().item()
|
| 149 |
-
|
|
|
|
| 150 |
if i == 0:
|
| 151 |
print(f"Test batch - Predicted: {predicted.cpu().numpy()[:10]}")
|
| 152 |
print(f"Test batch - Actual: {batch_y.cpu().numpy()[:10]}")
|
| 153 |
|
| 154 |
|
| 155 |
-
|
|
|
|
|
|
|
| 156 |
|
| 157 |
|
|
|
|
| 136 |
BATCH_SIZE = 256
|
| 137 |
correct = 0
|
| 138 |
total = 0
|
| 139 |
+
all_predictions = []
|
| 140 |
+
|
| 141 |
with torch.no_grad():
|
| 142 |
for i in range(0, len(X_test_t), BATCH_SIZE):
|
| 143 |
batch_x = X_test_t[i:i + BATCH_SIZE].to(self.device)
|
|
|
|
| 147 |
predicted = torch.argmax(outputs, dim=1)
|
| 148 |
total += batch_y.size(0)
|
| 149 |
correct += (predicted == batch_y).sum().item()
|
| 150 |
+
all_predictions.extend(predicted.cpu().numpy().tolist())
|
| 151 |
+
|
| 152 |
if i == 0:
|
| 153 |
print(f"Test batch - Predicted: {predicted.cpu().numpy()[:10]}")
|
| 154 |
print(f"Test batch - Actual: {batch_y.cpu().numpy()[:10]}")
|
| 155 |
|
| 156 |
|
| 157 |
+
return {
|
| 158 |
+
"predictions": all_predictions}
|
| 159 |
+
|
| 160 |
|
| 161 |
|