From fe650dd3260d0c24d6c5366b90d3507773872b2c Mon Sep 17 00:00:00 2001
From: Wing Lian <wing.lian@gmail.com>
Date: Wed, 29 May 2024 10:12:11 -0400
Subject: [PATCH] make sure the CI fails when pytest script fails (#1669)

* make sure the pytest script fails

* make sure the defaults come through for tests

* make sure tensorboard is loaded for test assertion
---
 cicd/cicd.sh                            | 1 +
 src/axolotl/utils/samplers/multipack.py | 4 ++--
 tests/e2e/patched/test_resume.py        | 1 +
 3 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/cicd/cicd.sh b/cicd/cicd.sh
index fa2049b6b..bc36458ab 100755
--- a/cicd/cicd.sh
+++ b/cicd/cicd.sh
@@ -1,4 +1,5 @@
 #!/bin/bash
+set -e
 
 pytest --ignore=tests/e2e/ /workspace/axolotl/tests/
 pytest /workspace/axolotl/tests/e2e/patched/
diff --git a/src/axolotl/utils/samplers/multipack.py b/src/axolotl/utils/samplers/multipack.py
index 07fd05682..1d025ca2d 100644
--- a/src/axolotl/utils/samplers/multipack.py
+++ b/src/axolotl/utils/samplers/multipack.py
@@ -80,8 +80,8 @@ class MultipackBatchSampler(BatchSampler):
         self.lengths = np.array(lengths, dtype=np.int32)
         self.batch_max_len = batch_max_len
         self.batch_size = batch_size
-        self.group_size = group_size
-        self.bin_size = bin_size
+        self.group_size = group_size if group_size is not None else 100_000
+        self.bin_size = bin_size if bin_size is not None else 200
         self.drop_last = drop_last
 
         self._efficiency = None
diff --git a/tests/e2e/patched/test_resume.py b/tests/e2e/patched/test_resume.py
index dfe9e8625..c0e791f38 100644
--- a/tests/e2e/patched/test_resume.py
+++ b/tests/e2e/patched/test_resume.py
@@ -62,6 +62,7 @@ class TestResumeLlama(unittest.TestCase):
                 "save_steps": 10,
                 "save_total_limit": 5,
                 "max_steps": 40,
+                "use_tensorboard": True,
             }
         )
         if is_torch_bf16_gpu_available():