Fix deprecated shared memory arg (huggingface#499)

jiqing-feng · Jan 5, 2024 · 19fef90 · 19fef90
1 parent 17bf770
commit 19fef90
Show file tree

Hide file tree

Showing 3 changed files with 9 additions and 9 deletions.
diff --git a/optimum/intel/openvino/modeling_decoder.py b/optimum/intel/openvino/modeling_decoder.py
@@ -388,7 +388,7 @@ def forward(
             inputs["position_ids"] = position_ids
 
         # Run inference
-        self.request.start_async(inputs, shared_memory=True)
+        self.request.start_async(inputs, share_inputs=True)
         self.request.wait()
         logits = torch.from_numpy(self.request.get_tensor("logits").data).to(self.device)
 

diff --git a/optimum/intel/openvino/modeling_diffusion.py b/optimum/intel/openvino/modeling_diffusion.py
@@ -570,7 +570,7 @@ def __call__(self, input_ids: np.ndarray):
         inputs = {
             "input_ids": input_ids,
         }
-        outputs = self.request(inputs, shared_memory=True)
+        outputs = self.request(inputs, share_inputs=True)
         return list(outputs.values())
 
 
@@ -604,7 +604,7 @@ def __call__(
         if timestep_cond is not None:
             inputs["timestep_cond"] = timestep_cond
 
-        outputs = self.request(inputs, shared_memory=True)
+        outputs = self.request(inputs, share_inputs=True)
         return list(outputs.values())
 
 
@@ -620,7 +620,7 @@ def __call__(self, latent_sample: np.ndarray):
         inputs = {
             "latent_sample": latent_sample,
         }
-        outputs = self.request(inputs, shared_memory=True)
+        outputs = self.request(inputs, share_inputs=True)
         return list(outputs.values())
 
     def _compile(self):
@@ -641,7 +641,7 @@ def __call__(self, sample: np.ndarray):
         inputs = {
             "sample": sample,
         }
-        outputs = self.request(inputs, shared_memory=True)
+        outputs = self.request(inputs, share_inputs=True)
         return list(outputs.values())
 
     def _compile(self):

diff --git a/optimum/intel/openvino/quantization.py b/optimum/intel/openvino/quantization.py
@@ -304,18 +304,18 @@ def __call__(self, *args, **kwargs):
                 data_cache.append(*args)
                 return self.request(*args, *kwargs)
 
-            def infer(self, inputs: Any = None, shared_memory: bool = False):
+            def infer(self, inputs: Any = None, share_inputs: bool = False):
                 data_cache.append(inputs)
-                return self.request.infer(inputs, shared_memory)
+                return self.request.infer(inputs, share_inputs)
 
             def start_async(
                 self,
                 inputs: Any = None,
                 userdata: Any = None,
-                shared_memory: bool = False,
+                share_inputs: bool = False,
             ):
                 data_cache.append(inputs)
-                self.request.infer(inputs, shared_memory)
+                self.request.infer(inputs, share_inputs)
 
             def wait(self):
                 pass