Help-Query for implementation of Fault Location(categorical data)prediction based on Binary Test Pattern and expected/actual output response (optional)

My name is Diptanshu Bagchi (India). I am a student of Electronics Communication Engineering. My MTech specialization is VLSI Design. I am currently working on a thesis on Fault Diagnosis of Digital Circuit (Part of Design for Testing subdomain in Digital VLSI).

There currently my job is to predict Fault location in a Digital Circuit using ML Model. I have generated a Fault table using Python to be used as feed for input Feature as part of Model Training to predict Fault Location, whereas my Training Model used here is Random Forest Classifier.

Here the generated dataset in the Fault Table has 10k rows containing Test Patterns, that is input in X axis and the binary pattern (like 00110101110…)of the input test data has approx 64 bits in each row and hence it’s encoded to be converted into a shorter length and then used. Next the Y axis i.e. the predicted value for each Faulty nodes in the digital circuit is also encoded in a numerical feature in the Y axis predicted value. Also in the same Fault table I have corresponding many targeted faulty nodes (data for training purpose) for each pattern with respective short numerical ids, that the test pattern can detect.

Next respective python code though is showing predicted value in encoded numerical value for each predicted faulty node or location, there is a problem with evaluation metrics like F1 Score, Recall, Sensitivity, Accuracy .

Here I am getting the value of each of these above parameters (used in Classifier model) is 0.0 in the output.

This part is very strange because I am completely new to this domain and as a novice I can’t guess accurately which factor is contributing to such problem of each parameter being generated.

My project is very crucial as part of final year thesis work and it completely depends on expected result otherwise I am stuck and cannot proceed further to show any considerable output for the ML model being used.

I am here to seek for help from this community fellow members.

Please anyone help me how to resolve such issue regarding the parameters I stated above.

It’s really urgent and please try to give any hint ASAP.

1 Like

Plz Fellow members, give me solution

1 Like