Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Benchmark script fixes #1301

Merged
merged 4 commits into from
Apr 8, 2021
Merged

Benchmark script fixes #1301

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
35f8af9
doc fix
zphang Mar 27, 2021
cd7eaf3
doc fix
zphang Mar 27, 2021
42d7886
fixes recommended by eritain
zphang Mar 27, 2021
7302452
Merge branch 'master' into benchmark_script_fixes
jeswan Apr 8, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions guides/benchmarks/glue.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,7 @@
python benchmark_submission_formatter.py \
--benchmark GLUE \
--input_base_path $INPUT_BASE_PATH \
--output_path $OUTPUT_BASE PATH
--output_path $OUTPUT_BASE_PATH
```

where `$INPUT_BASE_PATH` contains the task folder(s) output by [runscript.py](https://github.com/jiant-dev/jiant/blob/master/jiant/proj/main/runscript.py). Alternatively, a subset of tasks can be formatted using:
Expand All @@ -18,5 +18,5 @@ python benchmark_submission_formatter.py \
--benchmark GLUE \
--tasks cola mrpc \
--input_base_path $INPUT_BASE_PATH \
--output_path $OUTPUT_BASE PATH
--output_path $OUTPUT_BASE_PATH
```
4 changes: 2 additions & 2 deletions guides/benchmarks/superglue.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,7 +6,7 @@
python benchmark_submission_formatter.py \
--benchmark SUPERGLUE \
--input_base_path $INPUT_BASE_PATH \
--output_path $OUTPUT_BASE PATH
--output_path $OUTPUT_BASE_PATH
```

where `$INPUT_BASE_PATH` contains the task folder(s) output by [runscript.py](https://github.com/nyu-mll/jiant/blob/master/jiant/proj/main/runscript.py). Alternatively, a subset of tasks can be formatted using:
Expand All @@ -16,5 +16,5 @@ python benchmark_submission_formatter.py \
--benchmark SUPERGLUE \
--tasks cola mrpc \
--input_base_path $INPUT_BASE_PATH \
--output_path $OUTPUT_BASE PATH
--output_path $OUTPUT_BASE_PATH
```
8 changes: 4 additions & 4 deletions jiant/scripts/benchmarks/benchmark_submission_formatter.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@
import os
import argparse

from jiant.scripts.postproc.benchmarks import GlueBenchmark, SuperglueBenchmark
from jiant.scripts.benchmarks.benchmarks import GlueBenchmark, SuperglueBenchmark


SUPPORTED_BENCHMARKS = {"GLUE": GlueBenchmark, "SUPERGLUE": SuperglueBenchmark}
Expand All @@ -17,7 +17,7 @@ def main():
parser.add_argument(
"--input_base_path",
required=True,
help="base input path of benchmark task predictions (contains the benchmark task folders)",
help="base path where per-task folders contain raw prediction files",
)
parser.add_argument("--output_path", required=True, help="output path for formatted files")
parser.add_argument(
Expand All @@ -31,15 +31,15 @@ def main():
benchmark = SUPPORTED_BENCHMARKS[args.benchmark]

if args.tasks:
assert args.tasks in benchmark.TASKS
assert set(args.tasks) <= benchmark.TASKS
task_names = args.tasks
else:
task_names = benchmark.TASKS

for task_name in task_names:
input_filepath = os.path.join(args.input_base_path, task_name, "test_preds.p")
output_filepath = os.path.join(
args.output_path, benchmark.BENCHMARK_SUBMISSION_FILENAMES[task_name]
os.path.abspath(args.output_path), benchmark.BENCHMARK_SUBMISSION_FILENAMES[task_name]
)
benchmark.write_predictions(task_name, input_filepath, output_filepath)

Expand Down