Examinations Results: Not Qualified To Proceed

Tears in Rain

Hey, I've finished first year Theoretical Physics and under the examination results I'm in the section "Not Qualified to Proceed", and it says I need to sit a supplemental exam in Linear Algebra. OK fair enough, so what happens after I pass the supplemental exam? What will it say for my overall mark for the year depending on how well I do? And what's the difference between "Not Qualified to Proceed" and "Fail 1"?(there's a separate section for Fail 1.)

EuropeanSon

Tears in Rain wrote: »

Hey, I've finished first year Theoretical Physics and under the examination results I'm in the section "Not Qualified to Proceed", and it says I need to sit a supplemental exam in Linear Algebra. OK fair enough, so what happens after I pass the supplemental exam? What will it say for my overall mark for the year depending on how well I do? And what's the difference between "Not Qualified to Proceed" and "Fail 1"?(there's a separate section for Fail 1.)

I don't know about the first question as regards your grade. Possibly they'll give you a 3, but I don't know if they might give you higher if your average is high enough.
Fail 1 means their overall percentage score was failing (<40% or whatever the pass mark is). If you are NQTP you have an overall percentage above the pass mark.

dabh

Tears in Rain wrote: »

Hey, I've finished first year Theoretical Physics and under the examination results I'm in the section "Not Qualified to Proceed", and it says I need to sit a supplemental exam in Linear Algebra. OK fair enough, so what happens after I pass the supplemental exam? What will it say for my overall mark for the year depending on how well I do? And what's the difference between "Not Qualified to Proceed" and "Fail 1"?(there's a separate section for Fail 1.)

'Not Qualified to Proceed' means that your average was 40% or more, but you did not qualify to compensate. An F1 indicates an average of at least 30% but less than 40%.

Assuming you pass in Linear Algebra, the supplemental mark from the Linear Algebra will be averaged with marks from other modules from the annual (June) examination, and a new grade will be determined: I or II.1 or II.2 or III, depending on the overall mark. Your College record will however show that this grade was obtained at the supplemental examination. In this respect Maths and TP differ from Computer Science and other courses that (AFAIK) only record a grade of 'Clear' or 'Pass' for those passing at the supplemental examination.

Tears in Rain

dabh wrote: »

'Not Qualified to Proceed' means that your average was 40% or more, but you did not qualify to compensate. An F1 indicates an average of at least 30% but less than 40%.

Assuming you pass in Linear Algebra, the supplemental mark from the Linear Algebra will be averaged with marks from other modules from the annual (June) examination, and a new grade will be determined: I or II.1 or II.2 or III, depending on the overall mark. Your College record will however show that this grade was obtained at the supplemental examination. In this respect Maths and TP differ from Computer Science and other courses that (AFAIK) only record a grade of 'Clear' or 'Pass' for those passing at the supplemental examination.

Haha, brilliant, that's exactly wanted to hear, looks like a II.1 so I did hear that before from someone else, but my friends (mostly Computer Science) tried to convince me I was wrong, glad to hear I wasn't.

Maybe_Memories

Tears in Rain wrote: »

Hey, I've finished first year Theoretical Physics and under the examination results I'm in the section "Not Qualified to Proceed", and it says I need to sit a supplemental exam in Linear Algebra. OK fair enough, so what happens after I pass the supplemental exam? What will it say for my overall mark for the year depending on how well I do? And what's the difference between "Not Qualified to Proceed" and "Fail 1"?(there's a separate section for Fail 1.)

Out of curiousity, was it the Linear Algebra II?
Judging by the amount of people in maths who failed it (myself included) Vlad must have done the whole "right, one tiny mistake, whole thing is wrong!"

EuropeanSon

Maybe_Memories wrote: »

Out of curiousity, was it the Linear Algebra II?
Judging by the amount of people in maths who failed it (myself included) Vlad must have done the whole "right, one tiny mistake, whole thing is wrong!"

Ah now, that's hardly fair. It was a tricky paper, but I could have envisaged a scenario where I failed, and he still gave me 58%. I think he marked it fairly at worst.

PurpleFistMixer

Maybe_Memories wrote: »

Out of curiousity, was it the Linear Algebra II?
Judging by the amount of people in maths who failed it (myself included) Vlad must have done the whole "right, one tiny mistake, whole thing is wrong!"

I don't know if he said the same thing to you as he said to my class in first year, but he did warn us that calculation errors in certain (ie easy, calculation-based) questions would render the majority, if not the entire question wrong.

Tears in Rain

Maybe_Memories wrote: »

Out of curiousity, was it the Linear Algebra II?
Judging by the amount of people in maths who failed it (myself included) Vlad must have done the whole "right, one tiny mistake, whole thing is wrong!"

It was yeah and it'd be a bit weird if I failed I but not II wouldn't it? Ah well, just need to use the summer to turn into an Algebra god...

PurpleFistMixer wrote: »

I don't know if he said the same thing to you as he said to my class in first year, but he did warn us that calculation errors in certain (ie easy, calculation-based) questions would render the majority, if not the entire question wrong.

Yeah, heard the same thing from last year too...

Maybe_Memories

PurpleFistMixer wrote: »

I don't know if he said the same thing to you as he said to my class in first year, but he did warn us that calculation errors in certain (ie easy, calculation-based) questions would render the majority, if not the entire question wrong.

I'd assume Jordan Normal Form falls under this category?

Also, you know how on the front page of the exam he's like "you can use any statement proved in class or homework, but you most formulate it clearly, example, in class we proved that if blah blah blah then blah blah blah"

How strict is he on this? I just threw down the related stuff that we were given in class/homework and used it to answer the question, but never actually said where I got it from.

Tears in Rain

Maybe_Memories wrote: »

I'd assume Jordan Normal Form falls under this category?

Also, you know how on the front page of the exam he's like "you can use any statement proved in class or homework, but you most formulate it clearly, example, in class we proved that if blah blah blah then blah blah blah"

How strict is he on this? I just threw down the related stuff that we were given in class/homework and used it to answer the question, but never actually said where I got it from.

JNF counts as easy now? I still haven't a clue how it works

amacachi

EuropeanSon wrote: »

I don't know about the first question as regards your grade. Possibly they'll give you a 3, but I don't know if they might give you higher if your average is high enough.
Fail 1 means their overall percentage score was failing (<40% or whatever the pass mark is). If you are NQTP you have an overall percentage above the pass mark.

What's the difference between NQTP and Ungraded?

Maybe_Memories

Tears in Rain wrote: »

JNF counts as easy now? I still haven't a clue how it works

Well getting the JNF is grand.
Getting the Jordan Basis on the other hand...

I figured out a pattern by looking through all the problems he did, and came up with this:

Get the eigenvalues of your matrix A, and for each one get A-tI where t is the eigenvalue. Let B = A-tI

Get rk(B), rk(B^2) and so on until you get a rank that was the same as the previous one.

If rk(B) = rk(B^2), just solve Bx=0 and x is part of your jordan basis

If rk(B) =/= rk(B^2) but rk(B^2) = 0, solve Bx=0 and the vector you use is the missing leading one.

If rk(B) =/= rk(B^2) and rk(B^2) =/= 0, and say rk(B^k-1) = rk(B^k), you reduce the kernel of B^k using the kernel of B^k-1 and you get the basis vector.


Now, it's very possible I'm completely wrong. None of the books Vlad suggested have anything about Jordan Basis.

PurpleFistMixer

Maybe_Memories wrote: »

I'd assume Jordan Normal Form falls under this category?

Also, you know how on the front page of the exam he's like "you can use any statement proved in class or homework, but you most formulate it clearly, example, in class we proved that if blah blah blah then blah blah blah"

How strict is he on this? I just threw down the related stuff that we were given in class/homework and used it to answer the question, but never actually said where I got it from.

IIRC there were three question types. "Easy" straightforward calculations where it's simply applying a method (so if you make an arithmetic error you're punished), trickier questions that might require some though (I would guess that JNF falls under this because it's a "hard" calculation, even if it is just applying a method) and then proof-type questions where you haven't seen it before. He is more forgiving with the grading as the difficulty increases, or so he said.

Regarding using stuff from class, err, you sort of answered your own question I think. In his example he said "in class we proved", so you should include that. Or "from a theorem we proved in class" or "it is known that" or whatever. You just want to give a bit of an explanation of what you're doing rather than producing results from seemingly nowhere. Finding the right balance of explanation and maths depends a lot on the subject/lecturer/situation, though.

Tears in Rain

Maybe_Memories wrote: »

Well getting the JNF is grand.
Getting the Jordan Basis on the other hand...

I figured out a pattern by looking through all the problems he did, and came up with this:

Get the eigenvalues of your matrix A, and for each one get A-tI where t is the eigenvalue. Let B = A-tI

Get rk(B), rk(B^2) and so on until you get a rank that was the same as the previous one.

If rk(B) = rk(B^2), just solve Bx=0 and x is part of your jordan basis

If rk(B) =/= rk(B^2) but rk(B^2) = 0, solve Bx=0 and the vector you use is the missing leading one.

If rk(B) =/= rk(B^2) and rk(B^2) =/= 0, and say rk(B^k-1) = rk(B^k), you reduce the kernel of B^k using the kernel of B^k-1 and you get the basis vector.


Now, it's very possible I'm completely wrong. None of the books Vlad suggested have anything about Jordan Basis.

Haha, I'll give that a shot...hopefully it'll work out for some of the problems

Yeah it's tough going from notes when you don't have any external reference, I tend to follow Vlad's problem examples about half way through until BAM he seemingly pulls a vector out of nowhere, or someone infers the jordan block is of size two from divining sheep trails or some ****e...it's mental

Maybe_Memories

Tears in Rain wrote: »

Haha, I'll give that a shot...hopefully it'll work out for some of the problems

Yeah it's tough going from notes when you don't have any external reference, I tend to follow Vlad's problem examples about half way through until BAM he seemingly pulls a vector out of nowhere, or someone infers the jordan block is of size two from divining sheep trails or some ****e...it's mental

When you get the characteristic polynomial of the matrix, factorise it, the power of each factor is the size of the jordan block with the corresponding eigenvector on the diagonal.

Although sometimes he has extra numbers above the diagonal, which really confuses me.

Tears in Rain

Maybe_Memories wrote: »

When you get the characteristic polynomial of the matrix, factorise it, the power of each factor is the size of the jordan block with the corresponding eigenvector on the diagonal.

Although sometimes he has extra numbers above the diagonal, which really confuses me.

Ah deadly, that makes a lot more sense now