Fix problems with pytorch models passed to convert_model (#17255)

* Do eval() only for torch Module

* Add test

* Support decoder in convert_model

* Enable tests

src/bindings/python/src/openvino/frontend/pytorch/decoder.py

@@ -96,9 +96,7 @@ class TorchScriptPythonDecoder (Decoder):
         # We store every decoder created by this decoder so that all them are not deleted until the first decoder is deleted
         self.m_decoders = []
         self._input_signature = None
-        converted_model = False
         if graph_element is None:
-            converted_model = True
             pt_module = self._get_scripted_model(pt_module, example_input, freeze)
             self.graph_element = pt_module.inlined_graph
         else:
@@ -106,7 +104,7 @@ class TorchScriptPythonDecoder (Decoder):
         self.pt_module = pt_module
         self.raw_inputs = list(self.graph_element.inputs())
         self.raw_outputs = list(self.graph_element.outputs())
-        if self._input_signature is not None and self.raw_inputs[0].debugName() == "self":
+        if self._input_signature is not None and "self" in self.raw_inputs[0].debugName():
             self._input_signature.insert(0, "self")
 
         if isinstance(self.graph_element, torch.Graph):
@@ -137,7 +135,8 @@ class TorchScriptPythonDecoder (Decoder):
                     inputs = [inputs]
             return inputs, input_signature
 
-        pt_module.eval()
+        if isinstance(pt_module, torch.nn.Module):
+            pt_module.eval()
         input_signature = None
         if isinstance(pt_module, torch.nn.Module) and not isinstance(pt_module, (torch.jit._trace.TopLevelTracedModule, torch.jit._script.RecursiveScriptModule)):
             input_signature = list(inspect.signature(pt_module.forward).parameters.keys())

tests/layer_tests/mo_python_api_tests/test_mo_convert_pytorch.py

@@ -780,9 +780,8 @@ class TestMoConvertPyTorch(CommonMOConvertTest):
         create_pytorch_nn_module_sample_list_of_tensors,
         create_pytorch_jit_script_module,
         create_pytorch_jit_script_function,
-        # Disabled due to Ticket-109430
-        #create_pytorch_nn_module_layout_list,
-        #create_pytorch_nn_module_layout_list_case2,
+        create_pytorch_nn_module_layout_list,
+        create_pytorch_nn_module_layout_list_case2,
         create_pytorch_nn_module_mean_list,
         create_pytorch_nn_module_mean_list_default_no_compression,
         create_pytorch_nn_module_mean_list_compressin_enabled,

tests/layer_tests/py_frontend_tests/test_torch_decoder.py

@@ -598,3 +598,15 @@ def test_pytorch_decoder_can_convert_optional_tensor_none():
     outputs = list(nc_decoder.graph_element.outputs())
     assert len(outputs) == 1
     assert isinstance(outputs[0].type(), torch.NoneType)
+
+
+def f(x, y):
+    return x + y
+
+
+@pytest.mark.precommit
+def test_pytorch_decoder_can_convert_scripted_function():
+    from openvino.tools.mo import convert_model
+    scripted = torch.jit.script(f)
+    model = convert_model(scripted)
+    assert model is not None

tools/mo/openvino/tools/mo/convert_impl.py

@@ -571,8 +571,15 @@ def check_model_object(argv):
             return "tf"
     if 'torch' in sys.modules:
         import torch
-        if isinstance(model, torch.nn.Module) or isinstance(model, torch.jit.ScriptFunction):
+        if isinstance(model, (torch.nn.Module, torch.jit.ScriptFunction)):
             return "pytorch"
+        try:
+            from openvino.frontend.pytorch.decoder import TorchScriptPythonDecoder
+            
+            if isinstance(model, TorchScriptPythonDecoder):
+                return "pytorch"
+        except Exception as e:
+            pass
 
     import io
     if isinstance(model, io.BytesIO):