3172 Add utility to save a model into TorchScript model

monai/networks/__init__.py

@@ -10,6 +10,7 @@
 # limitations under the License.
 
 from .utils import (
+    convert_to_torchscript,
     copy_model_state,
     eval_mode,
     icnr_init,

monai/networks/utils.py

@@ -15,12 +15,14 @@
 import warnings
 from collections import OrderedDict
 from contextlib import contextmanager
-from typing import Any, Callable, Mapping, Optional, Sequence, Union
+from typing import Any, Callable, Dict, Mapping, Optional, Sequence, Union
 
 import torch
 import torch.nn as nn
 
 from monai.utils.deprecate_utils import deprecated_arg
+from monai.utils.misc import ensure_tuple, set_determinism
+from monai.utils.module import PT_BEFORE_1_7
 
 __all__ = [
     "one_hot",
@@ -34,6 +36,7 @@
     "eval_mode",
     "train_mode",
     "copy_model_state",
+    "convert_to_torchscript",
 ]
 
 
@@ -424,3 +427,68 @@ def copy_model_state(
     if inplace and isinstance(dst, torch.nn.Module):
         dst.load_state_dict(dst_dict)
     return dst_dict, updated_keys, unchanged_keys
+
+
+def convert_to_torchscript(
+    model: nn.Module,
+    filename_or_obj: Optional[Any] = None,
+    extra_files: Optional[Dict] = None,
+    verify: bool = False,
+    inputs: Optional[Sequence[Any]] = None,
+    device: Optional[torch.device] = None,
+    rtol: float = 1e-4,
+    atol: float = 0.0,
+    **kwargs,
+):
+    """
+    Utility to convert a model into TorchScript model and save to file,
+    with optional input / output data verification.
+
+    Args:
+        model: source PyTorch model to save.
+        filename_or_obj: if not None, specify a file-like object (has to implement write and flush)
+            or a string containing a file path name to save the TorchScript model.
+        extra_files: map from filename to contents which will be stored as part of the save model file.
+            works for PyTorch 1.7 or later.
+            for more details: https://pytorch.org/docs/stable/generated/torch.jit.save.html.
+        verify: whether to verify the input and output of TorchScript model.
+            if `filename_or_obj` is not None, load the saved TorchScript model and verify.
+        inputs: input test data to verify model, should be a sequence of data, every item maps to a argument
+            of `model()` function.
+        device: target device to verify the model, if None, use CUDA if available.
+        rtol: the relative tolerance when comparing the outputs of PyTorch model and TorchScript model.
+        atol: the absolute tolerance when comparing the outputs of PyTorch model and TorchScript model.
+
+    """
+    model.eval()
+    with torch.no_grad():
+        script_module = torch.jit.script(model)
+        if filename_or_obj is not None:
+            if PT_BEFORE_1_7:
+                torch.jit.save(m=script_module, f=filename_or_obj)
+            else:
+                torch.jit.save(m=script_module, f=filename_or_obj, _extra_files=extra_files)
+
+    if verify:
+        if device is None:
+            device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+        if inputs is None:
+            raise ValueError("missing input data for verification.")
+
+        inputs = [i.to(device) if isinstance(i, torch.Tensor) else i for i in inputs]
+        ts_model = torch.jit.load(filename_or_obj) if filename_or_obj is not None else script_module
+        ts_model.eval().to(device)
+        model = model.to(device)
+
+        with torch.no_grad():
+            set_determinism(seed=0)
+            torch_out = ensure_tuple(model(*inputs))
+            set_determinism(seed=0)
+            torchscript_out = ensure_tuple(ts_model(*inputs))
+            set_determinism(seed=None)
+        # compare TorchScript and PyTorch results
+        for r1, r2 in zip(torch_out, torchscript_out):
+            if isinstance(r1, torch.Tensor) or isinstance(r2, torch.Tensor):
+                torch.testing.assert_allclose(r1, r2, rtol=rtol, atol=atol)
+
+    return script_module

tests/test_convert_to_torchscript.py

@@ -0,0 +1,42 @@
+# Copyright 2020 - 2021 MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import os
+import tempfile
+import unittest
+
+import torch
+
+from monai.networks import convert_to_torchscript
+from monai.networks.nets import UNet
+
+
+class TestConvertToTorchScript(unittest.TestCase):
+    def test_value(self):
+        model = UNet(
+            spatial_dims=2, in_channels=1, out_channels=3, channels=(16, 32, 64), strides=(2, 2), num_res_units=0
+        )
+        with tempfile.TemporaryDirectory() as tempdir:
+            torchscript_model = convert_to_torchscript(
+                model=model,
+                filename_or_obj=os.path.join(tempdir, "model.ts"),
+                extra_files={"foo.txt": b"bar"},
+                verify=True,
+                inputs=[torch.randn((16, 1, 32, 32), requires_grad=False)],
+                device="cuda" if torch.cuda.is_available() else "cpu",
+                rtol=1e-3,
+                atol=1e-4,
+            )
+            self.assertTrue(isinstance(torchscript_model, torch.nn.Module))
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_utils_pytorch_numpy_unification.py

@@ -15,7 +15,8 @@
 import torch
 
 from monai.transforms.utils_pytorch_numpy_unification import percentile
-from tests.utils import TEST_NDARRAYS, assert_allclose, set_determinism
+from monai.utils import set_determinism
+from tests.utils import TEST_NDARRAYS, assert_allclose
 
 
 class TestPytorchNumpyUnification(unittest.TestCase):

tests/utils.py

@@ -22,7 +22,6 @@
 import unittest
 import warnings
 from functools import partial
-from io import BytesIO
 from subprocess import PIPE, Popen
 from typing import Callable, Optional, Tuple
 from urllib.error import HTTPError, URLError
@@ -35,7 +34,8 @@
 from monai.config.deviceconfig import USE_COMPILED
 from monai.config.type_definitions import NdarrayOrTensor
 from monai.data import create_test_image_2d, create_test_image_3d
-from monai.utils import ensure_tuple, optional_import, set_determinism
+from monai.networks import convert_to_torchscript
+from monai.utils import optional_import
 from monai.utils.misc import is_module_ver_at_least
 from monai.utils.module import version_leq
 from monai.utils.type_conversion import convert_data_type
@@ -578,54 +578,26 @@ def setUp(self):
         self.segn = torch.tensor(self.segn)
 
 
-def test_script_save(net, *inputs, eval_nets=True, device=None, rtol=1e-4):
+def test_script_save(net, *inputs, device=None, rtol=1e-4, atol=0.0):
     """
     Test the ability to save `net` as a Torchscript object, reload it, and apply inference. The value `inputs` is
-    forward-passed through the original and loaded copy of the network and their results returned. Both `net` and its
-    reloaded copy are set to evaluation mode if `eval_nets` is True. The forward pass for both is done without
-    gradient accumulation.
+    forward-passed through the original and loaded copy of the network and their results returned.
+    The forward pass for both is done without gradient accumulation.
 
     The test will be performed with CUDA if available, else CPU.
     """
-    if True:
-        device = "cpu"
-    else:
-        # TODO: It would be nice to be able to use GPU if
-        # available, but this currently causes CI failures.
-        if not device:
-            device = "cuda" if torch.cuda.is_available() else "cpu"
-
-    # Convert to device
-    inputs = [i.to(device) for i in inputs]
-
-    scripted = torch.jit.script(net.cpu())
-    buffer = scripted.save_to_buffer()
-    reloaded_net = torch.jit.load(BytesIO(buffer)).to(device)
-    net.to(device)
-
-    if eval_nets:
-        net.eval()
-        reloaded_net.eval()
-
-    with torch.no_grad():
-        set_determinism(seed=0)
-        result1 = net(*inputs)
-        result2 = reloaded_net(*inputs)
-        set_determinism(seed=None)
-
-    # convert results to tuples if needed to allow iterating over pairs of outputs
-    result1 = ensure_tuple(result1)
-    result2 = ensure_tuple(result2)
-
-    for i, (r1, r2) in enumerate(zip(result1, result2)):
-        if None not in (r1, r2):  # might be None
-            np.testing.assert_allclose(
-                r1.detach().cpu().numpy(),
-                r2.detach().cpu().numpy(),
-                rtol=rtol,
-                atol=0,
-                err_msg=f"failed on comparison number: {i}",
-            )
+    # TODO: would be nice to use GPU if available, but it currently causes CI failures.
+    device = "cpu"
+    with tempfile.TemporaryDirectory() as tempdir:
+        convert_to_torchscript(
+            model=net,
+            filename_or_obj=os.path.join(tempdir, "model.ts"),
+            verify=True,
+            inputs=inputs,
+            device=device,
+            rtol=rtol,
+            atol=atol,
+        )
 
 
 def query_memory(n=2):