[thejbills]

You’re not giving the approrate arguments for the PDF functions that you are selecting. When you test it on your side make sure that you can execute the PDF method with the appropriate arguments.

[thejbills]

GOOD catch. We’re updating that.

[thejbills]

That is correct. The test and validation datasets both contain the same three assets.

[thejbills]

We’ve recieved it, and just replied. Thanks

[thejbills]

Running the max and min numbers.

Max score
Max regression score: 1.0
Max classification score: 1.0
Max total score: 1.0

Min score
Min regression score: 0
Min classification score: -5.0
Min total score: -2.5

[thejbills]

Use
[email protected]

[thejbills]

Please just email us the request with the new team memeber’s information.

[thejbills]

Hi Qin,

As stated in the main description “Classification and regression predictions will be scored separately, and the final score is the mean score of all predictions.”

[thejbills]

Design torque is not a measured variable, thus it’s not in the X_train.csv (features) file. And the goal is to predict the torque margin which is a function of the design torque. Torque margin is given in the y_train.csv (targets) file.

[thejbills]

Our scoring function actually uses the parameters you supply including the scale (I’ll update the description language). But the scoring functions corrects area and peak probability.

If the area is not 1 it’s normalized to 1.
Then
If the max probability is greater than 1, it’s normalized to one.

[thejbills]

From the data challenge problem descrition page

Each engine is instrumented to capture the outside air temperature, mean gas temperature, power available, indicated airspeed, net power, and compressor speed.

oat = outside air temperature
mgt = mean gas temperature
pa = power available
ias = indicated airspeed
np = net power
ng = compressor speed (that one’s not as obvious)

[thejbills]

FYI, the leader board is up. All participants should have recieved an email notifying them.

[thejbills]

There is. We originaly had 6 engines, but we were able to get another one. There are 4 engines in the training dataset.

[thejbills]

There is. We originaly had 6 engines, but we were able to get another one. There are 4 engines in the training dataset.

[thejbills]

You are correct!

The idea is to strongly penalize predicting that the engine is okay when infact it’s not. This is a false negative.

For the reference of others reading this.

Positive refers to the faulty state
Negative referes to the healty state

False positive = predicting faulty engine when it is healthy
False negative = predicting healthy engine when it is faulty

[Author]

Posts