Do not import transformer_engine on import (#3056)

* Do not import `transformer_engine` on import

* fix message

* add test

* Update test_imports.py

* resolve comment 1/2

* resolve comment 1.5/2

* lint

* more lint

* Update tests/test_imports.py

Co-authored-by: Zach Mueller <muellerzr@gmail.com>

* fmt

---------

Co-authored-by: Zach Mueller <muellerzr@gmail.com>

src/accelerate/utils/transformer_engine.py

@@ -20,8 +20,7 @@
 from .operations import GatheredParameters
 
 
-if is_fp8_available():
-    import transformer_engine.pytorch as te
+# Do not import `transformer_engine` at package level to avoid potential issues
 
 
 def convert_model(model, to_transformer_engine=True, _convert_linear=True, _convert_ln=True):
@@ -30,6 +29,8 @@ def convert_model(model, to_transformer_engine=True, _convert_linear=True, _conv
     """
     if not is_fp8_available():
         raise ImportError("Using `convert_model` requires transformer_engine to be installed.")
+    import transformer_engine.pytorch as te
+
     for name, module in model.named_children():
         if isinstance(module, nn.Linear) and to_transformer_engine and _convert_linear:
             has_bias = module.bias is not None
@@ -87,6 +88,8 @@ def has_transformer_engine_layers(model):
     """
     if not is_fp8_available():
         raise ImportError("Using `has_transformer_engine_layers` requires transformer_engine to be installed.")
+    import transformer_engine.pytorch as te
+
     for m in model.modules():
         if isinstance(m, (te.LayerNorm, te.Linear, te.TransformerLayer)):
             return True
@@ -98,6 +101,8 @@ def contextual_fp8_autocast(model_forward, fp8_recipe, use_during_eval=False):
     Wrapper for a model's forward method to apply FP8 autocast. Is context aware, meaning that by default it will
     disable FP8 autocast during eval mode, which is generally better for more accurate metrics.
     """
+    if not is_fp8_available():
+        raise ImportError("Using `contextual_fp8_autocast` requires transformer_engine to be installed.")
     from transformer_engine.pytorch import fp8_autocast
 
     def forward(self, *args, **kwargs):
@@ -115,7 +120,8 @@ def apply_fp8_autowrap(model, fp8_recipe_handler):
     """
     Applies FP8 context manager to the model's forward method
     """
-    # Import here to keep base imports fast
+    if not is_fp8_available():
+        raise ImportError("Using `apply_fp8_autowrap` requires transformer_engine to be installed.")
     import transformer_engine.common.recipe as te_recipe
 
     kwargs = fp8_recipe_handler.to_kwargs() if fp8_recipe_handler is not None else {}

tests/test_imports.py

@@ -12,7 +12,9 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 import subprocess
+import sys
 
+from accelerate.test_utils import require_transformer_engine
 from accelerate.test_utils.testing import TempDirTestCase, require_import_timer
 from accelerate.utils import is_import_timer_available
 
@@ -31,7 +33,7 @@ def convert_list_to_string(data):
 
 
 def run_import_time(command: str):
-    output = subprocess.run(["python3", "-X", "importtime", "-c", command], capture_output=True, text=True)
+    output = subprocess.run([sys.executable, "-X", "importtime", "-c", command], capture_output=True, text=True)
     return output.stderr
 
 
@@ -81,3 +83,18 @@ def test_cli_import(self):
         paths_above_threshold = get_paths_above_threshold(sorted_data, 0.05, max_depth=7)
         err_msg += f"\n{convert_list_to_string(paths_above_threshold)}"
         self.assertLess(pct_more, 20, err_msg)
+
+
+@require_transformer_engine
+class LazyImportTester(TempDirTestCase):
+    """
+    Test suite which checks if specific packages are lazy-loaded.
+
+    Eager-import will trigger circular import in some case,
+    e.g. in huggingface/accelerate#3056.
+    """
+
+    def test_te_import(self):
+        output = run_import_time("import accelerate, accelerate.utils.transformer_engine")
+
+        self.assertFalse(" transformer_engine" in output, "`transformer_engine` should not be imported on import")