C2W1 Assignment Error while grading but everything works as expected in the lab

[Deepti_Prasad]

Hi @NicoleChJ

As commented previously, (YOU HAVE HARD CODED THE PATH, and instructions in the grade cell mentions
Don’t hardcode the paths. Use os.path.join to append the new directories to the root_path parameter

you need to do correction for GRADED FUNCTION: create_train_val_dirs
def create_train_val_dirs(root_path) and GRADED FUNCTION: split_data
def split_data(SOURCE_DIR, TRAINING_DIR, VALIDATION_DIR, SPLIT_SIZE)

Use the below comment link to make the correction for these two grade cells. Your output does not match for the split data grade cell. below is your output with the expected output

C2W1_Assignment_Cats vs Dogs

Hello Viktor, Thank you for sharing the screenshot of your error. based on your screenshot your this grader cell has incorrect code [Screenshot 1945-06-03 at 10.52.31 PM] As explained in the previous shared link in this post where it explains according to the hint given please follow these step to make the corrections Hints tells you to use os.makedirs to create your intermediate subdirectories not direct directories for your root path. So as I already have mentioned your first line of c…

3. In GRADED FUNCTION: create_model
   def create_model():
   In model.compile statement, optimizer need to be recalled as ‘rmsprop’ without adding any learning_rate

Regards
DP