Merge pull request #23 from NolanTrem/main

Completed MATH pipeline
emrgnt-cmplxty · Aug 25, 2023 · 200bfd8 · 200bfd8
2 parents f86123d + ae1f889
commit 200bfd8
Show file tree

Hide file tree

Showing 5,014 changed files with 41,710 additions and 30,167 deletions.
diff --git a/commands.md b/commands.md
@@ -61,3 +61,9 @@ poetry run python zero_shot_replication/runner.py --model=... --pset=gsm8k
 ```bash
 poetry run python zero_shot_replication/runner.py --model=... --pset=math
 ```
+
+```bash
+poetry run python evals/run_MATH_eval.py --solutions_file_path=../results/openai/MATH/gpt_4_0314/openai_math__model_eq_gpt_4_0314__temperature_eq_0p7.jsonl
+
+poetry run python evals/run_MATH_eval.py --solutions_file_path=../results/openai/MATH/gpt_4_0613/openai_math__model_eq_gpt_4_0613__temperature_eq_0p7.jsonl
+```
diff --git a/psets/MATH/MATH_dataset_full_testing.csv b/psets/MATH/MATH_dataset_full_testing.csv
diff --git a/psets/MATH/README.txt b/psets/MATH/README.txt
@@ -1,4 +1,6 @@
-The 5,000 TEST examples for the MATH dataset.
+The 5,000 TEST examples for the MATH dataset modified to CSV format for easy use.
+
+CSV created with the MATH-scraper tool availible here: https://github.com/NolanTrem/MATH-Scraper.
 
 
 If you find this useful for your work, please cite

diff --git a/psets/MATH/algebra/1.json b/psets/MATH/algebra/1.json
diff --git a/psets/MATH/algebra/10.json b/psets/MATH/algebra/10.json
diff --git a/psets/MATH/algebra/1001.json b/psets/MATH/algebra/1001.json
diff --git a/psets/MATH/algebra/1002.json b/psets/MATH/algebra/1002.json
diff --git a/psets/MATH/algebra/1004.json b/psets/MATH/algebra/1004.json
diff --git a/psets/MATH/algebra/1009.json b/psets/MATH/algebra/1009.json
diff --git a/psets/MATH/algebra/101.json b/psets/MATH/algebra/101.json
diff --git a/psets/MATH/algebra/1014.json b/psets/MATH/algebra/1014.json
diff --git a/psets/MATH/algebra/1017.json b/psets/MATH/algebra/1017.json
diff --git a/psets/MATH/algebra/1019.json b/psets/MATH/algebra/1019.json
diff --git a/psets/MATH/algebra/1020.json b/psets/MATH/algebra/1020.json
diff --git a/psets/MATH/algebra/1022.json b/psets/MATH/algebra/1022.json
diff --git a/psets/MATH/algebra/1023.json b/psets/MATH/algebra/1023.json
diff --git a/psets/MATH/algebra/1025.json b/psets/MATH/algebra/1025.json
diff --git a/psets/MATH/algebra/1026.json b/psets/MATH/algebra/1026.json
diff --git a/psets/MATH/algebra/1031.json b/psets/MATH/algebra/1031.json
diff --git a/psets/MATH/algebra/1032.json b/psets/MATH/algebra/1032.json
diff --git a/psets/MATH/algebra/1034.json b/psets/MATH/algebra/1034.json
diff --git a/psets/MATH/algebra/1035.json b/psets/MATH/algebra/1035.json
diff --git a/psets/MATH/algebra/104.json b/psets/MATH/algebra/104.json
diff --git a/psets/MATH/algebra/1042.json b/psets/MATH/algebra/1042.json
diff --git a/psets/MATH/algebra/1043.json b/psets/MATH/algebra/1043.json
diff --git a/psets/MATH/algebra/1044.json b/psets/MATH/algebra/1044.json
diff --git a/psets/MATH/algebra/1049.json b/psets/MATH/algebra/1049.json
diff --git a/psets/MATH/algebra/1050.json b/psets/MATH/algebra/1050.json
diff --git a/psets/MATH/algebra/1051.json b/psets/MATH/algebra/1051.json
diff --git a/psets/MATH/algebra/1057.json b/psets/MATH/algebra/1057.json
diff --git a/psets/MATH/algebra/1060.json b/psets/MATH/algebra/1060.json
diff --git a/psets/MATH/algebra/1063.json b/psets/MATH/algebra/1063.json
diff --git a/psets/MATH/algebra/1065.json b/psets/MATH/algebra/1065.json