Allow -1 to indicate unknown dimensions for shape override (#1300)

Signed-off-by: Tom Wildenhain <[email protected]>

Author: TomWildenhain-Microsoft

README.md

@@ -168,7 +168,7 @@ The target onnx file path.
 
 #### --inputs, --outputs
 
-TensorFlow model's input/output names, which can be found with [summarize graph tool](#summarize_graph). Those names typically end with ```:0```, for example ```--inputs input0:0,input1:0```. Inputs and outputs are ***not*** needed for models in saved-model format. Some models specify placeholders with unknown ranks and dims which can not be mapped to onnx. In those cases one can add the shape after the input name inside `[]`, for example `--inputs X:0[1,28,28,3]`
+TensorFlow model's input/output names, which can be found with [summarize graph tool](#summarize_graph). Those names typically end with ```:0```, for example ```--inputs input0:0,input1:0```. Inputs and outputs are ***not*** needed for models in saved-model format. Some models specify placeholders with unknown ranks and dims which can not be mapped to onnx. In those cases one can add the shape after the input name inside `[]`, for example `--inputs X:0[1,28,28,3]`. Use -1 to indicate unknown dimensions.
 
 #### --inputs-as-nchw
 

tf2onnx/shape_inference.py

@@ -100,7 +100,8 @@ def infer_shape_for_op(op):
             op.outputs[0].set_shape(new_shape)
             logger.debug("set placeholder op [%s] with new shape %s", op.outputs[0].name, new_shape)
             return True
-        logger.warning("Shape of placeholder %s is unknown, treated it as a scalar", op.name)
+        logger.warning("Shape of placeholder '%s' is unknown, treated it as a scalar. Please use the --input flag "
+                       "and append the shape to the input name if this input is not a scalar.", op.name)
         op.outputs[0].set_shape([])
         return True
 

tf2onnx/utils.py

@@ -107,7 +107,9 @@ def split_nodename_and_shape(name):
     for i in range(1, len(splits), 3):
         inputs.append(splits[i])
         if splits[i + 1] is not None:
-            shapes[splits[i]] = [int(n) for n in splits[i + 1][1:-1].split(",")]
+            shape = [int(n) for n in splits[i + 1][1:-1].split(",")]
+            shape = [n if n >= 0 else None for n in shape]
+            shapes[splits[i]] = shape
     if not shapes:
         shapes = None
     return inputs, shapes