LLM Zoomcamp 2024

Homework 3: Vector Search Statistics

Distribution of scores and reported study time for this homework.

Submissions

354

Median total score

6

Average total score

6

Score distribution

All values are points.

Questions score

Min
2
Median
6.0
Max
6
Q1
5.0
Avg
5.5
Q3
6.0

Learning in public score

Min
-
Median
0.0
Max
7
Q1
0.0
Avg
0.7
Q3
0.0

Total score

Min
2
Median
6.0
Max
14
Q1
5.0
Avg
6.2
Q3
6.0

Time distribution

All values are hours reported by students.

Lectures

Min
0.1
Median
4.0
Max
180.0
Q1
3.0
Avg
6.0
Q3
6.0

Homework

Min
0.0
Median
3.0
Max
30.0
Q1
2.0
Avg
3.8
Q3
4.5

Question breakdown

Correctness and answer distribution per question.

1. Getting the embeddings model: First value

342 / 354 correct (96.6%)

1 -0.24 0 (0.0%)
2 -0.04 9 (2.5%)
3 0.07 342 (96.6%)
4 0.27 2 (0.6%)

2. Creating the embeddings: Shape

354 / 354 correct (100.0%)

Answer Count
(375, 768) 286
(948, 768) 26
(375,768) 13
375, 768 3
768 3
(375,) 2
[1] 1
375 1
'(375, 768)' 1
Shape of X: (375, 768) 1
(375, 384) 1
`(948, 768)` 1
'(948, 768)' 1
(375, 678) 1
(948,768) 1
la forma de X serĂ­a (2, 768). 1
(768,) 1
all course documents shape (948,768) | MLops-zoomcamp shape (375,768) 1
the shape of X is: (375, 768) 1
3.3 1
Shape of X: (0,) 1
Shape of X: (948, 768) 1

3. Search: Highest score

341 / 354 correct (96.3%)

1 65.0 5 (1.4%)
2 6.5 4 (1.1%)
3 0.65 341 (96.3%)
4 0.065 2 (0.6%)

4. Hit-rate for our search engine

310 / 354 correct (87.6%)

1 0.93 310 (87.6%)
2 0.73 18 (5.1%)
3 0.53 6 (1.7%)
4 0.33 12 (3.4%)

5. Indexing with Elasticsearch: ID with the highest score

354 / 354 correct (100.0%)

Answer Count
ee58a693 240
'ee58a693' 5
gpvNspABoFYOXg7pnNlp 4
c6a22665 4
0227b872 3
3 2
OeLsrZABtfWUogOrK7j4 2
0a278fb2 2
'id': 'ee58a693' 2
RDWukpABlXkXunfNd2VQ 1
plx6opABR4ummgtAZ4Nz 1
Mtrpy5ABx3F3ZuhZKuuP 1
xdfwy5AB4S9snf-nVKE8 1
c839b764 1
cXJRpZABEjKDLYmdNTzP 1
hcbOrJABqX7xbSVQ1e3N 1
ee58a693 - '_score': 0.82532895, 1
9yNGsZABz2SCwBl9VD6e 1
A6nysJABXTX41oljXCU5 1
ANS: ee58a693 1
id': '7842b56a' 1
Koxbs5ABY5NlRslY9JkC 1
165597 1
67e2fd13 1
636f55d5 1
QiSBxpABaW2SpJzCzgBx 1
0.7374862 1
Caq2tZABx8A_rsVbQeq3 1
HwjItZABC8E9VgRzHKkx 1
I322tpABIjR23DumhxaF 1
c02e79ef 1
jEoyt5ABxQSt3T_5s1ZU 1
6ba259b1 1
. 1
cc60f7bc 1
IvlAuZABTXt3LRFGCyYZ 1
AKLStZABCodpLxrpbUmE 1
0.98 1
41521c92 1
0.82 1
ee58a692 1
RkC7vJABhYzG9Dnw0EB8 1
"ee58a693" 1
vqeCvZABPFZCKH5adFcn 1
Ml4GwJAB6701txdGLBcz 1
Bm0IwZABupWVn6tlAet7 1
Mhh_uZABceSvweL8aVLZ 1
'FfWIxpABwNnko1CMGwt3' 1
7Me2u5ABZBz-jAg4lu9n 1
USAQxpABZTH14HzSeHn_ 1
I47nxZABQVi2iz-W811Z 1
I gave up 1
5170565b 1
5 1
321 1
npY3x5ABZaGhAutJyDCu 1
vhpfx5ABnQzzkW0aSNFI 1
0.82532895 1
0.7454634 1
0.76 1
7842b56a 1
RJd_zJABxMuSB7Nk15Ek 1

6. Hit-rate for Elasticsearch

252 / 354 correct (71.2%)

1 0.93 252 (71.2%)
2 0.73 66 (18.6%)
3 0.53 14 (4.0%)
4 0.33 10 (2.8%)

Calculated: 11 October 2024, 08:34