From 8b8c6bb8379e1acf11e80168ea9d23b186c51832 Mon Sep 17 00:00:00 2001 From: Jake Poznanski Date: Tue, 5 Aug 2025 19:42:46 +0000 Subject: [PATCH] Cleaning up some training requirements installation steps --- gantry-train-requirements.txt | 6 ------ olmocr/train/README.md | 2 +- pyproject.toml | 5 +---- scripts/train/newtrainer-beaker.sh | 2 +- 4 files changed, 3 insertions(+), 12 deletions(-) delete mode 100644 gantry-train-requirements.txt diff --git a/gantry-train-requirements.txt b/gantry-train-requirements.txt deleted file mode 100644 index e65d726..0000000 --- a/gantry-train-requirements.txt +++ /dev/null @@ -1,6 +0,0 @@ -torchvision -accelerate -peft -wandb -omegaconf -einops \ No newline at end of file diff --git a/olmocr/train/README.md b/olmocr/train/README.md index 981fe04..2ca6789 100644 --- a/olmocr/train/README.md +++ b/olmocr/train/README.md @@ -9,7 +9,7 @@ The first step is to setup your python/conda environment, and set things up the Then, add in some extra training requirements: ```bash -pip install -r gantry-train-requirements.txt +pip install .[train] pip install transformers==4.52.4 pip install flash-attn==2.8.0.post2 --no-build-isolation ``` diff --git a/pyproject.toml b/pyproject.toml index df8dc76..7174359 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -99,13 +99,10 @@ train = [ "torch", "torchvision", "accelerate", - "datasets", "peft", "wandb", "omegaconf", - "s3fs", - "necessary", - "einops", + "einops" "augraphy", ] diff --git a/scripts/train/newtrainer-beaker.sh b/scripts/train/newtrainer-beaker.sh index 6273804..bd54a08 100755 --- a/scripts/train/newtrainer-beaker.sh +++ b/scripts/train/newtrainer-beaker.sh @@ -92,7 +92,7 @@ b = Beaker.from_env(default_workspace="ai2/olmocr") # Build the training command commands = [ - "pip install -r gantry-train-requirements.txt", + "pip install .[train]", "pip install transformers==4.52.4", "pip install flash-attn==2.8.0.post2 --no-build-isolation", "pip install s5cmd",