Optimize notebook validation #2053
Conversation
sacpis
commented
Aug 6, 2024
sacpis
commented
- Read the file content at once
- Encapsulate jupyter installation check in a function
- Uses a single regex search for all patterns
* Encapsulate jupyter installation check in a function * Uses a single regex search for all patterns
|
CUDA Quantum Docs Bot: A preview of the documentation can be found here. |
|
CUDA Quantum Docs Bot: A preview of the documentation can be found here. |
|
CUDA Quantum Docs Bot: A preview of the documentation can be found here. |
|
CUDA Quantum Docs Bot: A preview of the documentation can be found here. |
|
@bmhowe23 Would you be able to review this PR? |
|
CUDA Quantum Docs Bot: A preview of the documentation can be found here. |
docs/notebook_validation.py
Outdated
| if any( | ||
| re.search(pattern, lines) for pattern in | ||
| ['set_target[\\\s\(]+"(.+)\\\\"[)]', '--target ([^ ]+)']): | ||
| matches = re.findall( | ||
| 'set_target[\\\s\(]+"(.+)\\\\"[)]|--target ([^ ]+)', lines) |
There was a problem hiding this comment.
I'm having trouble figuring out what this code change is doing. It looks more confusing to me with the duplicate strings in there. What exactly is this optimizing?
There was a problem hiding this comment.
I'm having trouble figuring out what this code change is doing. It looks more confusing to me with the duplicate strings in there. What exactly is this optimizing?
And I'm also curious which notebooks (if any) show a measurable difference in runtime because of this optimization?
There was a problem hiding this comment.
Thanks @bmhowe23 for reviewing this change. I will put these patterns in a list instead.
This change avoids unnecessary computations by first ensuring that any pattern from the pattern list exists and if It it does not then it avoids further checks (if condition for a match).
If any pattern exists, then it uses re.findall to collect all matching substrings in a single operation.
Basically, the idea for this PR is to read the entire file at once (f.read()). This avoids reading the file using f.readlines() (as it creates a list of string (per line) in memory, which could use high memory).
* creating a common list for patterns
…pis/cuda-quantum into optimize_notebook_validation
|
CUDA Quantum Docs Bot: A preview of the documentation can be found here. |
