dearwolf/aider
1
0
Fork 0
mirror of https://github.com/Aider-AI/aider.git synced 2025-05-30 17:24:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 3

copy

Browse source
This commit is contained in:
Paul Gauthier 2023-06-30 13:48:01 -07:00
parent cd6a278684
commit c793511957
1 changed files with 6 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

12
docs/benchmarks.md
View file

@ -217,20 +217,21 @@ usually on the order of 3-6 different variations. This feels
like they are load balancing across a number of slightly different like they are load balancing across a number of slightly different
instances of the model. instances of the model.
For some exercises, some of this variable responses pass the unit tests and other For some exercises, some of these variable responses pass the unit tests while
responses do not. other variants do not. Whether the exercises passes is therefore
a bit random, depending on which variant OpenAI returns.
Given that, it would be ideal to run all 133 exercises many times for each Given that, it would be ideal to run all 133 exercises many times for each
model/edit-format combination and report an average performance. model/edit-format combination and report an average performance.
This would average away the effect of the API variance. This would average away the effect of the API variance.
That would also significantly increase the cost of this sort of benchmarking, It would also significantly increase the cost of this sort of benchmarking,
so I didn't do that. so I didn't do that.
Benchmarking against 133 exercises provides some robustness all by itself, since Benchmarking against 133 exercises provides some robustness all by itself, since
we are measuring the performance across many exercises. we are measuring the performance across many exercises.
But to get a sense of how much the API variance impacts the benchmark outcomes, But to get a sense of how much the API variance impacts the benchmark outcomes,
I ran the `gpt-3.5-turbo-0613 + whole` experiment 10 times. I ran the `gpt-3.5-turbo-0613 / whole` experiment 10 times.
You'll see one set of error bars in the graph, which demark You'll see one set of error bars in the graph, which demark
the range of results across those 10 runs. the range of results across those 10 runs.
@ -239,7 +240,7 @@ contribute to a large variance in the benchmark results.
## Conclusions ## Conclusions
Based on these benchmarking results, aider will continue to usea Based on these benchmarking results, aider will continue to use
`whole` for gpt-3.5 and `diff` for gpt-4. `whole` for gpt-3.5 and `diff` for gpt-4.
While `gpt-4` gets slightly better results with the `whole` edit format, While `gpt-4` gets slightly better results with the `whole` edit format,
it significantly increases costs and latency compared to `diff`. it significantly increases costs and latency compared to `diff`.
@ -247,4 +248,3 @@ Since `gpt-4` is already costly and slow, this seems like an acceptable
tradeoff. tradeoff.