This is a collection of benchmark I collected over the years. In this blog post here I describe what I did and how new benchmarks can be added. For more results of a benchmark, please have a look in the respective folder within benchmarks.
| DATE | TEST | COMMENT | BEST | TIME_FACTOR | BEST_RUNS | DETAILS | DURATION |
|---|---|---|---|---|---|---|---|
| 2024-02-08 20:01:58 | assign with <- or = | varying size of vector | arrow sign | 1.2% | 3/6 | link | 00:00:00 |
| 2024-02-08 20:02:01 | Accsess a colum in a data frame, table or tibble. | varying size of data | $ df | 76.6% | 4/4 | link | 00:00:02 |
| 2024-02-08 20:02:02 | commented-out code | no comments vs. lots of comments | No comments | 3% | 7/7 | link | 00:00:01 |
| 2024-02-08 20:11:29 | Calculation of the cross product | varying number of rows and colums | crossprod(S) | 16.1% | 7/10 | link | 00:01:02 |
| 2024-02-08 20:02:13 | Creating dummies out of factor variable. | changing size and number of unique values | lapply | 35.1% | 7/12 | link | 00:00:10 |
| 2024-02-08 20:03:02 | Selecting rows by a filter criterion. | changing the colum type and the number of unique values | DT == & | -205.2% | 22/40 | link | 00:00:48 |
| 2024-02-08 20:05:11 | Is it faster to intersect or use which. | changing the length and the percentage of fits | which | 31% | 39/40 | link | 00:02:08 |
| 2024-02-08 20:05:19 | get range of numeric vector | varying size | max(x) - min(x) | 50.1% | 6/6 | link | 00:00:07 |
| 2024-02-08 20:05:32 | Creating a sample index. | varying sample size and range | runif | 77.9% | 5/5 | link | 00:00:12 |
| 2024-02-08 20:05:46 | Subset between values of a vector | between vs. <= and >= | <=> | 41.7% | 47/64 | link | 00:00:13 |
| 2024-02-08 20:06:01 | Is it faster to sum over 0's or NA's. | changing the length and the percentage of NA's and 0's | sum with 0 | 43.9% | 4/5 | link | 00:00:14 |
| 2024-02-08 20:06:04 | get unique levels of factors | varying sample size and number of unique value | unique(x) | 67.8% | 16/16 | link | 00:00:02 |