Model training failed with an error

vaishnavi
vaishnavi Registered Posts: 40 ✭✭✭✭

I have 6 inputs to my model where 5 inputs are dummy encoded and 1 input is vectorized thru CountVectorization. I have selected RandomForest, LogisticRegression, SVM & Neural network in the "design" section of the lab. When I click on "Train" am getting the below error and model training fails.

Error:-

----------------------------

Training failed

Failed to train : <class 'TypeError'> : an integer is required (got type bytes)
When I choose only one feature with "TfIDF vectorization" getting below error :-
--------------------------------------------------------------------------------------------------------------------------
Failed to train : <class 'FileNotFoundError'> : [Errno 2] No such file or directory: '/usr/src/app/dataiku-install/analysis-data/ADT_RETR/AAhDY96O/3NTdaWlj/sessions/s4/pp5/tfidfvec_Cleansed_Failure/Error.pkl.gz'
When I choose only one feature with "CountVectorizer", I am getting the below error :-
-------------------------------------------------------------------------------------------------------------------
Failed to train : <class 'FileNotFoundError'> : [Errno 2] No such file or directory: '/usr/src/app/dataiku-install/analysis-data/ADT_RETR/AAhDY96O/3NTdaWlj/sessions/s2/pp5/countvec_Cleansed_Failure/Error.pkl.gz'
I am using python 3.9

Answers

  • Georghios
    Georghios Dataiku DSS Core Designer, Dataiku DSS ML Practitioner, Dataiku DSS Adv Designer, Registered Posts: 15 ✭✭✭
    edited July 17

    I have an issue with RandomForest too but mine is:

    Failed to train : <class 'TypeError'> : Object of type int64 is not JSON serializable

Setup Info
    Tags
      Help me…