diff --git a/.dockerignore b/.dockerignore index 919c2da..0f3fbb2 100644 --- a/.dockerignore +++ b/.dockerignore @@ -14,4 +14,5 @@ scripts/**/*.sh # Nor copy any olmocr bench files olmOCR-bench/ olmOCR-bench*/ -html_templates*/ \ No newline at end of file +html_templates*/ +olmocr-synthmix-*/ \ No newline at end of file diff --git a/.gitignore b/.gitignore index 1b40340..77b9cb3 100644 --- a/.gitignore +++ b/.gitignore @@ -18,8 +18,6 @@ old_train/ gpt4otestset_output/* pdfs/* olmOCR-bench/* -olmOCR-bench-0825/ -olmOCR-bench-snapshot-082225/ table_data*/ /synth*/ dolma_samples/* @@ -29,6 +27,7 @@ filtered_items_prefilter/ augraphy_cache/ /*.html html_templates*/ +olmocr-synthmix*/ scoreelo.csv debug.log birrpipeline-debug.log