Medicalai ErrorStoppping Not Working

[zeynepkoyun]

Hi @aibharata

When we want to stop the training early with medicalai via a callback called EarlyStopping, it gives an error at line 209 in the file "medicalai/chief/core.py".
How should the data be defined here, because it is desired to behave like a string first and then a dictionary.
Can you share sample data?

Error line

if bestModelCond is not None and bestModelCond.lower() != 'default':
print('\n[INFO]: ENGINE INITIALIZED WITH "SAVING BEST MODEL" and Early Stopping MODE\n')
earlystop_callback = tf.keras.callbacks.EarlyStopping(
monitor=bestModelCond['monitor'],
min_delta=bestModelCond['min_delta'],
patience=bestModelCond['patience']
)
if callbacks is None:
callbacks=[earlystop_callback]
else:
callbacks.append(earlystop_callback)

[zeynepkoyun]

Hi @aibharata

When we want to stop the training early with medicalai via a callback called EarlyStopping, it gives an error at line 209 in the file "medicalai/chief/core.py". How should the data be defined here, because it is desired to behave like a string first and then a dictionary. Can you share sample data?

Error line

if bestModelCond is not None and bestModelCond.lower() != 'default': print('\n[INFO]: ENGINE INITIALIZED WITH "SAVING BEST MODEL" and Early Stopping MODE\n') earlystop_callback = tf.keras.callbacks.EarlyStopping( monitor=bestModelCond['monitor'], min_delta=bestModelCond['min_delta'], patience=bestModelCond['patience'] ) if callbacks is None: callbacks=[earlystop_callback] else: callbacks.append(earlystop_callback)

Hi @aibharata

is there any progress on this?

[aibharata]

Hi @zeynepkoyun,

Thank you for bringing this to our notice. There are a few bugs here. bestModelCond = 'default' should have initialized the conditions, but it doesn't as of now.

Currently, it expects a dictionary. You can initialize and pass bestModelCond like this :

bestModelCond  = {
    monitor='val_accuracy', 
    patience=8, 
    min_delta=0.001, 
}

We will issue a bug fix in a few days to support more conditions.

By the way, which version are you using currently? 1.1.59 or 1.2.9.2rc0

[zeynepkoyun]

Merhaba @zeynepkoyun ,

Bunu dikkatimize sunduğunuz için teşekkür ederiz. Burada birkaç bug var. bestModelCond = 'default' koşulları başlatmalıydı, ama şu an için değil.

Şu anda, bir sözlük bekliyor. bestModelCond'u şu şekilde başlatabilir ve iletebilirsiniz:

bestModelCond  = {
    monitor='val_accuracy', 
    patience=8, 
    min_delta=0.001, 
}

Daha fazla koşulu desteklemek için birkaç gün içinde bir hata düzeltmesi yayınlayacağız.

Bu arada, şu anda hangi sürümü kullanıyorsunuz? 1.1.59 or 1.2.9.2rc0

Hi @aibharata
I'll try and let you know.
The version I am currently using is 1.1.59.
I will be waiting for the new version.