You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Select your parameters and paths
2.1. Setting main training parameters
2.2. Data augmentation
Here I disable Use_Data_augmentation
2.3. Using weights from a pre-trained model as initial weights
Using the trained model
5.1 Generate prediction(s) from unseen dataset
Here in setion 5.1 is where I get the error.
Describe the bug
See beginning of the message
Expected behavior
Get model predictions
Screenshots
Desktop (please complete the following information):
OS: macOS Monterey
Browser: chrome
Version: chrome version 127.0.6533.122
My version of the notebook is:
6. Version log
v2.1.2:
Correct for data loading to avoid .DS_Store or similar
v2.1.1:
Replaced all absolute pathing with relative pathing
v2.1:
Updated to TensorFlow 2.11
Updated to fpdf2 and add lines to ensure a proper format. Correct keras package version parsing.
Smartphone (please complete the following information):
Does not apply
The text was updated successfully, but these errors were encountered:
Sorry for the delay, it has been a really busy week 😅
Seems like the code it's failing when getting the input size, I will take a look to the notebook in ZeroCostDL4Mic.
I will see if I can reproduce the error and fix it, when I manage to solve the issue I will let you know here with another comment 🤗
One quick question to try to reproduce the error, when using section 2.3. Using weights from a pre-trained model as initial weights. Did you used any previous weights?
Thanks for coming back to me.
To your question: when using section 2.3. Using weights from a pre-trained model as initial weights. Did you used any previous weights?
In section 2.3 I have:
Then I go straight to 5.1 Generate prediction(s) from unseen dataset where I have:
Inside the results folder I have unet_2d_multlilabel/weights_best.hdf5 so like this: results/unet_2d_multlilabel/weights_best.hdf5
Then I comment out the line to avoid the error:
print('Model input size: '+str(Input_size[0])+'x'+str(Input_size[1]))
I get the following error in
section 5.1 Generate prediction(s) from unseen dataset
when running the notebookU-Net_2D_Multilabel_ZeroCostDL4Mic
.IndexError: list index out of range
The error happens in
function_29
, in the following line:print('Model input size: '+str(Input_size[0])+'x'+str(Input_size[1]))
I believe what is happening here is that
Input_size
is[(None, 256, 256, 1)]
, whereInput_size[0]
is theNone
(the batch size)If comment the
Input_size
everywhere in the function I am able to get the predictions:To Reproduce
I have run the following sections of the notebook:
1.1. Load key dependencies
2.1. Setting main training parameters
2.2. Data augmentation
Here I disable
Use_Data_augmentation
2.3. Using weights from a pre-trained model as initial weights
5.1 Generate prediction(s) from unseen dataset
Here in setion 5.1 is where I get the error.
Describe the bug
See beginning of the message
Expected behavior
Get model predictions
Screenshots
Desktop (please complete the following information):
My version of the notebook is:
Smartphone (please complete the following information):
Does not apply
The text was updated successfully, but these errors were encountered: