At the testing accuracy, I get 0.92.
When I run “Test your function” I get this error.
IndexError: single positional indexer is out-of-bounds
At the end of the get_accuracy function, I get i = 4950 and m = 4951.
I checked to see what i was on the last go around and it was 4950.
So I am lost as to the cause of the out of bounds.
I am using city1 = data.iloc[i][‘city1’] to get city1
That is what the error points to.
1442 len_axis = len(self.obj._get_axis(axis))
1443 if key >= len_axis or key < -len_axis:
→ 1444 raise IndexError(“single positional indexer is out-of-bounds”)
All tests pass up to this point.
Any idea as where I might check?
Thanks
Okay, I kept checking and changed iloc to loc.
FAIL on my part.
It works now.
Thanks
1 Like
Glad you got it. I noticed that if you print out the string returned for city1 in the graded function the last element is Zagreb, but in the unit test it goes back to Kabul. which is where the exception is thrown. I didn’t track down why yet and it seems overtaken by events so I won’t bother.
Not sure how you are implementing the guidance
# loop through the rows of the dataframe
One way is using pandas.DataFrame.iterrows() which returns two objects. First is the index, i, and second is the data of the row as a Series object. From which you can access a column as my_row[‘my_column’]. Seems more efficient to use the row directly provided by the iterator rather than do another lookup using the index, but I didn’t benchmark the two approaches. Certainly is easier to write and read. Cheers
I made it past this original post but got stuck at the same place as several others did, 4 passing and 2 failing tests with the wrong values.
I went to the thread where the others posted their same issue.
Should I create a new post or post to the thread?
https://community.deeplearning.ai/t/stuck-with-dimensions-on-compute-pca/68696/13
Thanks
Upon further reading in the Pandas doc, it looks to me like itertuples is the right choice…
https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.itertuples.html