Why add coding at train.py it cannot run?

[m012110024]

When adding new coding inside the train.py. The new coding I add is needed to print out the confusion matrix, but after that, the new coding inside the train.py didn’t show anything.
The coding i add inside as shown below:

%%

import torch
import torchmetrics

def get_Accuracy(num_class):

accuracy = torchmetrics.Accuracy(
    num_classes=num_class,
    average='micro',
    multiclass=False
    # top_k=1
)

return accuracy

def get_Precision(num_class):

precision = torchmetrics.Precision(
    num_classes=num_class,
    average='micro',
    multiclass=False,
)

return precision

def get_Dice():
dice = torchmetrics.Dice(
num_classes=2,
average=‘micro’,

)
return dice

def get_Average_precision(num_class):

average_precision = torchmetrics.AveragePrecision(

# num_classes=num_class,

pos_label=0,

average=None

)

return average_precision

def get_Precision_Recall():
precision_recall = torchmetrics.PrecisionRecallCurve(
num_classes=2,
pos_label=1
)

return precision_recall

def get_AUC():
AUC = torchmetrics.AUC(
reorder=True
)
return AUC

def get_F1Score():
F1_Score = torchmetrics.F1Score(
num_classes=2,

)
return F1_Score

def get_Confusion_Matrix():
confusion_matrix = torchmetrics.ConfusionMatrix(
num_classes=2,
multilabel=False
)

return confusion_matrix

%%

preds = torch.tensor([1, 0.4, 0.8, 0.4])

target = torch.tensor([1, 1, 0, 0])

dice = get_Dice()

accuracy = get_Accuracy(1)

average_precision = get_Average_precision(1)

precision = get_Precision(1)

confusion_matrix = get_Confusion_Matrix()

# result = accuracy(preds, target)

acc = accuracy(preds, target)

confusion = confusion_matrix(preds, target)

result = precision(preds, target)

print(acc, result, confusion)

%%

[AastaLLL]

Hi,

Except confusion matrix, do you get acc and result output log?

Thanks.

[m012110024]

No

[m012110024]

Only show this result

[AastaLLL]

Hi,

Which train.py are you using?
Is it the below one? But the source looks very different compared to the content you shared.

dusty-nv/pytorch-classification/blob/master/train.py

Thanks.

[m012110024]

I using the jetson inference train.py

[AastaLLL]

There is no update from you for a period, assuming this is not an issue any more.
Hence we are closing this topic. If need further support, please open a new one.
Thanks

Hi,

Could you share which script you are using with us?
Thanks.