Update README.md

README.md

@@ -1,17 +1,19 @@
----
-license: cc-by-4.0
----
-
-### Submitting Results to the Leaderboard
-First, install teh TabArena dependencies:
-```bash
-# Requires latest mainline AutoGluon (or AutoGluon 1.3+)
-git clone https://github.com/autogluon/autogluon
-./autogluon/full_install.sh
-
-git clone https://github.com/autogluon/tabrepo.git
-pip install -e tabrepo/[benchmark]
-
-git clone https://huggingface.co/datasets/TabArena/benchmark_results
-```
+---
+license: cc-by-4.0
+---
+
+### Submitting Results to the Leaderboard
+First, install the TabArena dependencies:
+```bash
+pip install uv
+uv venv --seed --python 3.11 ~/.venvs/tabarena
+source ~/.venvs/tabarena
+
+git clone --branch tabarena https://github.com/autogluon/tabrepo.git
+
+# use GIT_LFS_SKIP_SMUDGE=1 in front of the command if installing TabDPT fails due to a broken LFS/pip setup. 
+GIT_LFS_SKIP_SMUDGE=1 uv pip install -e tabrepo/[benchmark]
+
+git clone https://huggingface.co/datasets/TabArena/benchmark_results
+```
 Next, concatenate your model performance results to 'df_results.csv' and rerun 'get_leaderboard_csv.py' to obtain an updated leaderboard.