FuzzBench: 2023-01-15-latest-bug report

experiment summary

We show two different aggregate (cross-benchmark) rankings of fuzzers. The first is based on the average of per-benchmarks scores, where the score represents the percentage of the highest reached median bug-coverage on a given benchmark (higher value is better). The second ranking shows the average rank of fuzzers, after we rank them on each benchmark according to their median reached bug-covereges (lower value is better).
By avg. score
average normalized score
fuzzer
afl 75.0
aflfast 75.0
eclipser 75.0
libafl 75.0
mopt 75.0
aflsmart 50.0
centipede 50.0
honggfuzz 50.0
libfuzzer 50.0
fairfuzz 25.0
By avg. rank
average rank
fuzzer
afl 1.2
aflfast 1.2
eclipser 1.2
libafl 1.2
mopt 1.2
aflsmart 2.4
libfuzzer 2.4
honggfuzz 2.6
centipede 3.6
fairfuzz 3.8
  • Critical difference diagram
    The diagram visualizes the average rank of fuzzers (second ranking above) while showing the significance of the differences as well. What is considered a "critical difference" (CD) is based on the Friedman/Nemenyi post-hoc test. See more in the documentation.
    Note: If a fuzzer does not support all benchmarks, its ranking as shown in this diagram can be lower than it should be. So please check the list of supported benchmarks for the fuzzer(s) of your interest. The list could be specified in the fuzzer's README.md like this.
  • Median relative code-coverages on each benchmark

    Note: The relative coverage summary table shows the median relative performance of each fuzzer to the experiment maximum. Thus the highest relative performance may not be 100%.
    trial_relative_coverage = trial_coverage / experiment_max_coverage

      aflsmart afl mopt eclipser honggfuzz libafl aflfast libfuzzer centipede fairfuzz
    FuzzerMedian 94.00 95.00 96.00 94.00 93.00 90.00 93.00 88.00 83.50 82.00
    FuzzerMean 93.33 89.60 89.60 89.40 89.20 88.80 88.40 84.80 82.50 81.80
    bloaty_fuzz_target_52948c 94.00 95.00 96.00 94.00 92.00 90.00 93.00 88.00 nan 80.00
    harfbuzz_hb-shape-fuzzer_17863b 96.00 96.00 96.00 96.00 96.00 99.00 95.00 95.00 87.00 89.00
    libxml2_xml_e85b9b nan 98.00 97.00 98.00 96.00 84.00 97.00 82.00 80.00 82.00
    mbedtls_fuzz_dtlsclient_7c6b0e nan 69.00 69.00 69.00 69.00 80.00 67.00 68.00 67.00 69.00
    php_php-fuzz-parser_0dbedb 90.00 90.00 90.00 90.00 93.00 91.00 90.00 91.00 96.00 89.00
    • Fuzzers are sorted by "FuzzerMean" (average median relative coverage), highest on the left.
    • Green background = highest relative median coverage.
    • Blue gradient background = greater than 95% relative median coverage.
  • Median relative bug-coverages on each benchmark

    Note: The relative coverage summary table shows the median relative performance of each fuzzer to the experiment maximum. Thus the highest relative performance may not be 100%.
    trial_relative_coverage = trial_coverage / experiment_max_coverage

      aflsmart afl aflfast eclipser libafl mopt centipede honggfuzz libfuzzer fairfuzz
    FuzzerMedian 100.00 100.00 100.00 100.00 100.00 100.00 50.00 0.00 0.00 0.00
    FuzzerMean 66.67 60.00 60.00 60.00 60.00 60.00 50.00 40.00 40.00 20.00
    bloaty_fuzz_target_52948c 100.00 100.00 100.00 100.00 100.00 100.00 nan 0.00 100.00 0.00
    harfbuzz_hb-shape-fuzzer_17863b 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00
    libxml2_xml_e85b9b nan 100.00 100.00 100.00 100.00 100.00 0.00 100.00 0.00 0.00
    mbedtls_fuzz_dtlsclient_7c6b0e nan 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
    php_php-fuzz-parser_0dbedb 0.00 0.00 0.00 0.00 0.00 0.00 100.00 0.00 0.00 0.00
    • Fuzzers are sorted by "FuzzerMean" (average median relative coverage), highest on the left.
    • Green background = highest relative median coverage.
    • Blue gradient background = greater than 95% relative median coverage.
  • Total unique bugs found on each benchmark
      Total libfuzzer afl mopt aflfast eclipser honggfuzz libafl centipede aflsmart fairfuzz
    FuzzerSum 12 11 10 10 9 9 8 8 7 5 5
    bloaty_fuzz_target_52948c 1 1 1 1 1 1 1 1 0 1 1
    harfbuzz_hb-shape-fuzzer_17863b 7 6 5 6 5 5 4 5 3 3 2
    libxml2_xml_e85b9b 2 2 2 2 2 2 2 2 2 nan 2
    mbedtls_fuzz_dtlsclient_7c6b0e 0 0 0 0 0 0 0 0 0 nan 0
    php_php-fuzz-parser_0dbedb 2 2 2 1 1 1 1 0 2 1 0
    • Fuzzers are sorted by "FuzzerSum", highest on the left.
    • Green background = most unique bugs found.
    • *note: This table represents unique bugs found across all trials.

bloaty_fuzz_target_52948c summary

Discovered bug coverage distribution
Reached code coverage distribution
Mean code coverage growth over time
Mean code coverage growth over time
Mean bug coverage growth over time
Mean bug coverage growth over time
* The error bands show the 95% confidence interval around the mean code coverage.
error
The following fuzzers do not have enough samples: mopt, libafl, libfuzzer, honggfuzz, eclipser, fairfuzz.
  • Sample statistics and statistical significance (bugs covered)
    Bug coverage sample statistics
    count mean std min 25% median 75% max
    fuzzer time
    afl 82800 19.0 1.000000 0.000000 1.0 1.0 1.0 1.00 1.0
    aflfast 82800 19.0 0.947368 0.229416 0.0 1.0 1.0 1.00 1.0
    aflsmart 82800 16.0 1.000000 0.000000 1.0 1.0 1.0 1.00 1.0
    eclipser 82800 11.0 1.000000 0.000000 1.0 1.0 1.0 1.00 1.0
    libafl 82800 13.0 0.923077 0.277350 0.0 1.0 1.0 1.00 1.0
    libfuzzer 82800 13.0 1.000000 0.000000 1.0 1.0 1.0 1.00 1.0
    mopt 82800 14.0 1.000000 0.000000 1.0 1.0 1.0 1.00 1.0
    fairfuzz 82800 11.0 0.090909 0.301511 0.0 0.0 0.0 0.00 1.0
    honggfuzz 82800 12.0 0.250000 0.452267 0.0 0.0 0.0 0.25 1.0

    Vargha-Delaney A12 measure
    The table summarizes the A12 values from the pairwise Vargha-Delaney A measure of effect size. Green cells indicate the probability the fuzzer in the row will outperform the fuzzer in the column.
    Mann-Whitney U test
    The table summarizes the p values of pairwise Mann-Whitney U tests. Green cells indicate that the reached coverage distribution of a given fuzzer pair is significantly different.
  • Sample statistics and statistical significance (code coverage)
    Code coverage sample statistics
    count mean std min 25% median 75% max
    fuzzer time
    mopt 82800 14.0 6024.928571 100.926069 5885.0 5949.0 6029.0 6085.0 6233.0
    afl 82800 19.0 5909.210526 104.304351 5723.0 5851.0 5935.0 5966.0 6159.0
    aflsmart 82800 16.0 5864.812500 118.706483 5641.0 5763.5 5896.5 5944.5 6011.0
    eclipser 82800 11.0 5869.363636 121.009316 5684.0 5800.5 5863.0 5907.5 6093.0
    aflfast 82800 19.0 5844.000000 127.906650 5563.0 5766.0 5849.0 5943.5 6047.0
    honggfuzz 82800 12.0 5776.666667 112.743743 5633.0 5691.5 5786.0 5815.0 6064.0
    libafl 82800 13.0 5558.538462 290.271601 5110.0 5307.0 5671.0 5747.0 5937.0
    libfuzzer 82800 13.0 5524.384615 94.910957 5357.0 5481.0 5517.0 5590.0 5654.0
    fairfuzz 82800 11.0 5079.545455 94.912974 4941.0 5023.0 5040.0 5146.5 5263.0

    Vargha-Delaney A12 measure
    The table summarizes the A12 values from the pairwise Vargha-Delaney A measure of effect size. Green cells indicate the probability the fuzzer in the row will outperform the fuzzer in the column.
    Mann-Whitney U test
    The table summarizes the p values of pairwise Mann-Whitney U tests. Green cells indicate that the reached coverage distribution of a given fuzzer pair is significantly different.
  • Unique code coverage plots
    Ranking by unique code branches covered
    Each bar shows the total number of code branches found by a given fuzzer. The colored area shows the number of unique code branches (i.e., branches that were not covered by any other fuzzers).
    Pairwise unique code coverage
    Each cell represents the number of code branches covered by the fuzzer of the column but not by the fuzzer of the row

harfbuzz_hb-shape-fuzzer_17863b summary

Discovered bug coverage distribution
Reached code coverage distribution
Mean code coverage growth over time
Mean code coverage growth over time
Mean bug coverage growth over time
Mean bug coverage growth over time
* The error bands show the 95% confidence interval around the mean code coverage.
error
The following fuzzers do not have enough samples: eclipser, fairfuzz, honggfuzz, libfuzzer, libafl, mopt, centipede.
  • Sample statistics and statistical significance (bugs covered)
    Bug coverage sample statistics
    count mean std min 25% median 75% max
    fuzzer time
    afl 82800 18.0 1.000000 0.000000 1.0 1.0 1.0 1.0 1.0
    aflfast 82800 17.0 1.000000 0.000000 1.0 1.0 1.0 1.0 1.0
    aflsmart 82800 19.0 1.000000 0.000000 1.0 1.0 1.0 1.0 1.0
    centipede 82800 5.0 1.000000 0.000000 1.0 1.0 1.0 1.0 1.0
    eclipser 82800 13.0 1.000000 0.000000 1.0 1.0 1.0 1.0 1.0
    fairfuzz 82800 13.0 0.769231 0.438529 0.0 1.0 1.0 1.0 1.0
    honggfuzz 82800 13.0 1.000000 0.000000 1.0 1.0 1.0 1.0 1.0
    libafl 82800 11.0 1.000000 0.000000 1.0 1.0 1.0 1.0 1.0
    libfuzzer 82800 12.0 1.000000 0.000000 1.0 1.0 1.0 1.0 1.0
    mopt 82800 11.0 1.000000 0.000000 1.0 1.0 1.0 1.0 1.0

    Vargha-Delaney A12 measure
    The table summarizes the A12 values from the pairwise Vargha-Delaney A measure of effect size. Green cells indicate the probability the fuzzer in the row will outperform the fuzzer in the column.
    Mann-Whitney U test
    The table summarizes the p values of pairwise Mann-Whitney U tests. Green cells indicate that the reached coverage distribution of a given fuzzer pair is significantly different.
  • Sample statistics and statistical significance (code coverage)
    Code coverage sample statistics
    count mean std min 25% median 75% max
    fuzzer time
    libafl 82800 11.0 10442.090909 41.466745 10365.0 10422.00 10462.0 10476.0 10484.0
    honggfuzz 82800 13.0 10155.538462 36.951805 10098.0 10145.00 10153.0 10164.0 10249.0
    mopt 82800 11.0 10109.636364 100.313780 9885.0 10065.50 10153.0 10181.5 10194.0
    aflsmart 82800 19.0 10133.157895 69.989097 10002.0 10097.50 10129.0 10175.5 10259.0
    eclipser 82800 13.0 10125.384615 48.560166 10023.0 10101.00 10125.0 10161.0 10185.0
    afl 82800 18.0 10108.000000 56.604926 9984.0 10076.25 10113.5 10140.5 10194.0
    aflfast 82800 17.0 10056.352941 48.827427 9989.0 10011.00 10051.0 10106.0 10130.0
    libfuzzer 82800 12.0 9952.916667 57.754116 9858.0 9904.25 9970.0 9998.0 10029.0
    fairfuzz 82800 13.0 9139.153846 480.566306 8251.0 8796.00 9394.0 9584.0 9675.0
    centipede 82800 5.0 9213.200000 113.018583 9028.0 9205.00 9225.0 9297.0 9311.0

    Vargha-Delaney A12 measure
    The table summarizes the A12 values from the pairwise Vargha-Delaney A measure of effect size. Green cells indicate the probability the fuzzer in the row will outperform the fuzzer in the column.
    Mann-Whitney U test
    The table summarizes the p values of pairwise Mann-Whitney U tests. Green cells indicate that the reached coverage distribution of a given fuzzer pair is significantly different.
  • Unique code coverage plots
    Ranking by unique code branches covered
    Each bar shows the total number of code branches found by a given fuzzer. The colored area shows the number of unique code branches (i.e., branches that were not covered by any other fuzzers).
    Pairwise unique code coverage
    Each cell represents the number of code branches covered by the fuzzer of the column but not by the fuzzer of the row

libxml2_xml_e85b9b summary

Discovered bug coverage distribution
Reached code coverage distribution
Mean code coverage growth over time
Mean code coverage growth over time
Mean bug coverage growth over time
Mean bug coverage growth over time
* The error bands show the 95% confidence interval around the mean code coverage.
error
The following fuzzers do not have enough samples: honggfuzz, eclipser, mopt, fairfuzz, centipede, libfuzzer.
  • Sample statistics and statistical significance (bugs covered)
    Bug coverage sample statistics
    count mean std min 25% median 75% max
    fuzzer time
    afl 82800 17.0 1.000000 0.000000 1.0 1.00 1.0 1.0 1.0
    aflfast 82800 18.0 0.888889 0.323381 0.0 1.00 1.0 1.0 1.0
    eclipser 82800 10.0 1.000000 0.000000 1.0 1.00 1.0 1.0 1.0
    honggfuzz 82800 12.0 0.750000 0.452267 0.0 0.75 1.0 1.0 1.0
    libafl 82800 16.0 0.750000 0.447214 0.0 0.75 1.0 1.0 1.0
    mopt 82800 10.0 1.000000 0.000000 1.0 1.00 1.0 1.0 1.0
    centipede 82800 8.0 0.375000 0.517549 0.0 0.00 0.0 1.0 1.0
    fairfuzz 82800 9.0 0.444444 0.527046 0.0 0.00 0.0 1.0 1.0
    libfuzzer 82800 7.0 0.428571 0.534522 0.0 0.00 0.0 1.0 1.0

    Vargha-Delaney A12 measure
    The table summarizes the A12 values from the pairwise Vargha-Delaney A measure of effect size. Green cells indicate the probability the fuzzer in the row will outperform the fuzzer in the column.
    Mann-Whitney U test
    The table summarizes the p values of pairwise Mann-Whitney U tests. Green cells indicate that the reached coverage distribution of a given fuzzer pair is significantly different.
  • Sample statistics and statistical significance (code coverage)
    Code coverage sample statistics
    count mean std min 25% median 75% max
    fuzzer time
    eclipser 82800 10.0 19458.200000 754.776906 17334.0 19563.25 19693.0 19775.75 19855.0
    afl 82800 17.0 19667.117647 195.991671 19304.0 19540.00 19683.0 19761.00 20055.0
    mopt 82800 10.0 19617.500000 134.506712 19406.0 19534.25 19619.5 19688.75 19816.0
    aflfast 82800 18.0 19218.166667 934.341243 16689.0 19329.00 19548.5 19658.50 19813.0
    honggfuzz 82800 12.0 19433.166667 99.255257 19326.0 19359.50 19420.0 19483.25 19675.0
    libafl 82800 16.0 17019.750000 51.516340 16932.0 16992.25 17016.0 17068.25 17098.0
    fairfuzz 82800 9.0 17459.888889 1768.016788 15825.0 16327.00 16569.0 19714.00 19823.0
    libfuzzer 82800 7.0 17147.857143 1442.872994 15328.0 16153.50 16510.0 18613.00 18664.0
    centipede 82800 8.0 16199.500000 286.754250 15784.0 16024.25 16183.0 16404.00 16629.0

    Vargha-Delaney A12 measure
    The table summarizes the A12 values from the pairwise Vargha-Delaney A measure of effect size. Green cells indicate the probability the fuzzer in the row will outperform the fuzzer in the column.
    Mann-Whitney U test
    The table summarizes the p values of pairwise Mann-Whitney U tests. Green cells indicate that the reached coverage distribution of a given fuzzer pair is significantly different.
  • Unique code coverage plots
    Ranking by unique code branches covered
    Each bar shows the total number of code branches found by a given fuzzer. The colored area shows the number of unique code branches (i.e., branches that were not covered by any other fuzzers).
    Pairwise unique code coverage
    Each cell represents the number of code branches covered by the fuzzer of the column but not by the fuzzer of the row

mbedtls_fuzz_dtlsclient_7c6b0e summary

Discovered bug coverage distribution
Reached code coverage distribution
Mean code coverage growth over time
Mean code coverage growth over time
Mean bug coverage growth over time
Mean bug coverage growth over time
* The error bands show the 95% confidence interval around the mean code coverage.
error
The following fuzzers do not have enough samples: libafl, centipede, libfuzzer, eclipser, mopt, fairfuzz, honggfuzz.
  • Sample statistics and statistical significance (bugs covered)
    Bug coverage sample statistics
    count mean std min 25% median 75% max
    fuzzer time
    afl 82800 18.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
    aflfast 82800 19.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
    centipede 82800 12.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
    eclipser 82800 11.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
    fairfuzz 82800 10.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
    honggfuzz 82800 10.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
    libafl 82800 13.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
    libfuzzer 82800 12.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
    mopt 82800 11.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

    Vargha-Delaney A12 measure
    The table summarizes the A12 values from the pairwise Vargha-Delaney A measure of effect size. Green cells indicate the probability the fuzzer in the row will outperform the fuzzer in the column.
    Mann-Whitney U test
    The table summarizes the p values of pairwise Mann-Whitney U tests. Green cells indicate that the reached coverage distribution of a given fuzzer pair is significantly different.
  • Sample statistics and statistical significance (code coverage)
    Code coverage sample statistics
    count mean std min 25% median 75% max
    fuzzer time
    libafl 82800 13.0 2931.692308 373.710758 2553.0 2572.00 3020.0 3189.00 3490.0
    eclipser 82800 11.0 2670.545455 221.561894 2569.0 2592.00 2603.0 2628.00 3336.0
    honggfuzz 82800 10.0 2717.800000 257.587871 2560.0 2589.25 2603.0 2618.00 3209.0
    fairfuzz 82800 10.0 2592.000000 38.233203 2496.0 2584.00 2602.5 2612.00 2631.0
    afl 82800 18.0 2578.166667 42.486330 2418.0 2574.00 2586.5 2594.25 2625.0
    mopt 82800 11.0 2663.090909 363.231456 2391.0 2577.50 2586.0 2596.00 3737.0
    libfuzzer 82800 12.0 2624.083333 183.964156 2556.0 2562.25 2570.0 2585.50 3207.0
    centipede 82800 12.0 2529.416667 17.885791 2493.0 2523.00 2536.0 2539.00 2555.0
    aflfast 82800 19.0 2472.526316 120.301181 2201.0 2504.00 2526.0 2538.00 2560.0

    Vargha-Delaney A12 measure
    The table summarizes the A12 values from the pairwise Vargha-Delaney A measure of effect size. Green cells indicate the probability the fuzzer in the row will outperform the fuzzer in the column.
    Mann-Whitney U test
    The table summarizes the p values of pairwise Mann-Whitney U tests. Green cells indicate that the reached coverage distribution of a given fuzzer pair is significantly different.
  • Unique code coverage plots
    Ranking by unique code branches covered
    Each bar shows the total number of code branches found by a given fuzzer. The colored area shows the number of unique code branches (i.e., branches that were not covered by any other fuzzers).
    Pairwise unique code coverage
    Each cell represents the number of code branches covered by the fuzzer of the column but not by the fuzzer of the row

php_php-fuzz-parser_0dbedb summary

Discovered bug coverage distribution
Reached code coverage distribution
Mean code coverage growth over time
Mean code coverage growth over time
Mean bug coverage growth over time
Mean bug coverage growth over time
* The error bands show the 95% confidence interval around the mean code coverage.
error
The following fuzzers do not have enough samples: centipede, eclipser, libfuzzer, mopt, libafl, fairfuzz, aflsmart, honggfuzz.
  • Sample statistics and statistical significance (bugs covered)
    Bug coverage sample statistics
    count mean std min 25% median 75% max
    fuzzer time
    centipede 82800 14.0 0.928571 0.267261 0.0 1.0 1.0 1.0 1.0
    afl 82800 18.0 0.166667 0.383482 0.0 0.0 0.0 0.0 1.0
    aflfast 82800 17.0 0.058824 0.242536 0.0 0.0 0.0 0.0 1.0
    aflsmart 82800 7.0 0.285714 0.487950 0.0 0.0 0.0 0.5 1.0
    eclipser 82800 13.0 0.076923 0.277350 0.0 0.0 0.0 0.0 1.0
    fairfuzz 82800 8.0 0.000000 0.000000 0.0 0.0 0.0 0.0 0.0
    honggfuzz 82800 7.0 0.142857 0.377964 0.0 0.0 0.0 0.0 1.0
    libafl 82800 12.0 0.000000 0.000000 0.0 0.0 0.0 0.0 0.0
    libfuzzer 82800 13.0 0.384615 0.506370 0.0 0.0 0.0 1.0 1.0
    mopt 82800 13.0 0.153846 0.375534 0.0 0.0 0.0 0.0 1.0

    Vargha-Delaney A12 measure
    The table summarizes the A12 values from the pairwise Vargha-Delaney A measure of effect size. Green cells indicate the probability the fuzzer in the row will outperform the fuzzer in the column.
    Mann-Whitney U test
    The table summarizes the p values of pairwise Mann-Whitney U tests. Green cells indicate that the reached coverage distribution of a given fuzzer pair is significantly different.
  • Sample statistics and statistical significance (code coverage)
    Code coverage sample statistics
    count mean std min 25% median 75% max
    fuzzer time
    centipede 82800 14.0 17761.071429 232.535676 17468.0 17598.75 17656.0 17917.50 18267.0
    honggfuzz 82800 7.0 17056.714286 93.369721 16878.0 17024.00 17099.0 17116.00 17140.0
    libafl 82800 12.0 16768.583333 45.601950 16718.0 16744.75 16753.5 16770.25 16887.0
    libfuzzer 82800 13.0 16749.923077 29.053002 16698.0 16739.00 16746.0 16774.00 16792.0
    eclipser 82800 13.0 16592.461538 38.855749 16518.0 16573.00 16603.0 16613.00 16646.0
    afl 82800 18.0 16595.166667 52.320562 16458.0 16581.50 16595.0 16619.25 16682.0
    mopt 82800 13.0 16570.846154 40.744828 16492.0 16540.00 16586.0 16601.00 16618.0
    aflsmart 82800 7.0 16580.285714 28.969606 16545.0 16563.00 16573.0 16596.50 16625.0
    aflfast 82800 17.0 16536.117647 81.067782 16334.0 16500.00 16545.0 16561.00 16747.0
    fairfuzz 82800 8.0 16283.000000 209.272208 15975.0 16142.75 16367.0 16390.75 16565.0

    Vargha-Delaney A12 measure
    The table summarizes the A12 values from the pairwise Vargha-Delaney A measure of effect size. Green cells indicate the probability the fuzzer in the row will outperform the fuzzer in the column.
    Mann-Whitney U test
    The table summarizes the p values of pairwise Mann-Whitney U tests. Green cells indicate that the reached coverage distribution of a given fuzzer pair is significantly different.
  • Unique code coverage plots
    Ranking by unique code branches covered
    Each bar shows the total number of code branches found by a given fuzzer. The colored area shows the number of unique code branches (i.e., branches that were not covered by any other fuzzers).
    Pairwise unique code coverage
    Each cell represents the number of code branches covered by the fuzzer of the column but not by the fuzzer of the row

experiment data

You can download the raw data for this report here.

Check out the documentation on how to create customized reports using this data. Also see some example Colab notebooks for doing custom analysis on the data here.

Experiment Description:

(None,)