C1 W1 Assignment - Uncaught (in promise) TypeError: Cannot convert undefined or null to object

Course Q&A TensorFlow: Data and Deployment Specialization Browser-based Models with TensorFlow.js
1

I have problem with the Course 1 / Week 1 Assignment. I am delayed with the delivery date for this assignment, I have been studyng javascript callbacks and promises trying to understand the problem, the code seems fine but when the program runs it reports errors.

The program needs to read 2 datasets, ‘wdbc-train.csv’ and ‘wdbc-test.csv’, the same code works fine with the first dataset and fail processing the second dataset. Firstable, I believe the second dataset, ‘wdbc-test.csv’ had wrong data, but it is fine. When I swith the order, processing first ‘wdbc-test.csv’ and after that ‘wdbc-train.csv’ I receive the error message processing the second dataset, always with the second dataset. Please help me to fix the problem.

The portion of code with the problem is:
const convertedTestingData =
await testingData.map(
({xs, ys}) => {
console.log(‘xs:’, xs, ‘ys:’, ys); // Check values here
return ({ xs: Object.values(xs), ys: Object.values(ys) })
}
).batch(10);

The error message:
C1_W1_Assignment.html:63 xs: undefined ys: undefined
C1_W1_Assignment.html:64 Uncaught (in promise) TypeError: Cannot convert undefined or null to object
at Function.values ()
at C1_W1_Assignment.html:64:46
at tfjs@latest:17:942499
at tfjs@latest:17:220203
at e.value (tfjs@latest:17:220344)
at e.value (tfjs@latest:17:220100)
at X5 (tfjs@latest:17:243540)
at r.transform (tfjs@latest:17:942477)
at r. (tfjs@latest:17:931911)
at h (tfjs@latest:17:2100)

1 Like
2

Make sure your assignment notebook and other necessary files are in same folder and is able to read your data from the assignment notebook.

3

Thank you for the response.
The ‘data’ folder is in the same folder with the assignment code.
The .csv files are fine, if I switch the files, reading the ‘wdbc-train.csv’ for testingUrl, the program fails reading this file. Always the programs process correctly the first file and fails in the second file, no matter which file it is.
Also, I can display the ‘testingData’, the problem occurs in the .map process.

The code switching the file:

        const testingUrl = '/data/wdbc-train.csv';   
        const testingData = await tf.data.csv(testingUrl, {
            columnConfigs: {
                diagnosis: {islabel: true}
            }                    
        });   
        console.log('testingData=', await testingData); 
        const arrayTestingData = await testingData.toArray();
        console.log('arrayTestingData=', await arrayTestingData)           
        // Convert the testing data into arrays in the space below.
        // Note: In this case, the labels are integers, not strings.
        // Therefore, there is no need to convert string labels into
        // a one-hot encoded array of label values like we did in the
        // Iris dataset example.  
        const convertedTestingData = 
            await testingData.map(
                ({xs, ys}) => { 
                    console.log('xs:', xs, 'ys:', ys); // Check values here  
                    return ({ xs: Object.values(xs), ys: Object.values(ys) })   
                }
            ).batch(10);
        console.log('convertedTestingData=', await convertedTestingData);

        const arrayconvertedTestingData = await convertedTestingData.toArray();
        console.log('arrayconvertedTestingData=', await arrayconvertedTestingData);

The output and error message:

Navigated to http://127.0.0.1:8087/C1_W1_Assignment.html
C1_W1_Assignment.html:17 trainingData= s {size: null, input: r, hasHeader: true, fullColumnNames: undefined, columnNamesValidated: false, …}
C1_W1_Assignment.html:19 arrayTrainingData= (114) [{…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, …]
C1_W1_Assignment.html:31 convertedTrainingData= i {size: null}
index-4ab31049.js:9 SyntaxError: Unexpected token ‘b’, “b6c0b0d1-9”… is not valid JSON
at JSON.parse ()
at Mi.parseValue (index-4ab31049.js:9:26389)
at Mi.get (index-4ab31049.js:9:26074)
at async aa (quill.tsx-efc8bc9b.js:2:42085)
(anonymous) @ index-4ab31049.js:9
(anonymous) @ index-4ab31049.js:9
await in (anonymous) (async)
aa @ quill.tsx-efc8bc9b.js:2
await in aa (async)
(anonymous) @ quill.tsx-efc8bc9b.js:19
Show 4 more frames
Show less
C1_W1_Assignment.html:33 arrayconvertedTrainingData= (12) [{…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}]
C1_W1_Assignment.html:53 testingData= s {size: null, input: r, hasHeader: true, fullColumnNames: undefined, columnNamesValidated: false, …}
C1_W1_Assignment.html:55 arrayTestingData= (455) [{…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, {…}, …]
C1_W1_Assignment.html:68 convertedTestingData= i {size: null}
C1_W1_Assignment.html:64 xs: undefined ys: undefined
C1_W1_Assignment.html:65 Uncaught (in promise) TypeError: Cannot convert undefined or null to object
at Function.values ()
at C1_W1_Assignment.html:65:46
at tfjs@latest:17:942499
at tfjs@latest:17:220203
at e.value (tfjs@latest:17:220344)
at e.value (tfjs@latest:17:220100)
at X5 (tfjs@latest:17:243540)
at r.transform (tfjs@latest:17:942477)
at r. (tfjs@latest:17:931911)
at h (tfjs@latest:17:2100)
(anonymous) @ C1_W1_Assignment.html:65
(anonymous) @ tfjs@latest:17
(anonymous) @ tfjs@latest:17
value @ tfjs@latest:17
value @ tfjs@latest:17
X5 @ tfjs@latest:17
(anonymous) @ tfjs@latest:17
(anonymous) @ tfjs@latest:17
h @ tfjs@latest:17
(anonymous) @ tfjs@latest:17
(anonymous) @ tfjs@latest:17
u @ tfjs@latest:17
o @ tfjs@latest:17
await in o (async)
(anonymous) @ C1_W1_Assignment.html:128

1 Like
4

Hello @Juan_J_Moreno

When I said all file same folder I also meant, it should be located where your environment assignment is looking for those files, per se in my case it was local c drive location where I placed my folder but if I remove the assignment book from that folder and directly place in local c drive, I was getting still the error.

Did you get what I am saying?

5

Thank you very much.
Yes I understand, both files are in local drive in my PC.

I have resolved the problem deleting all the code for the testing-data that looks identical to the code for training-data. Then, I made a copy of the training-data code, after that I rename the variables and the program finally works. I can’t find exactly where was the problem.

I’d submitted the assignment and passed.

Thanks.

2 Likes
6

I made sure the test and training files have same number of columns and do not have nulls, outliers etc.