| # RUN: tf-mlir-translate -graphdef-to-mlir -tf-enable-shape-inference-on-import=false %s -tf-input-arrays=p,x -tf-input-data-types="DT_INT32,DT_RESOURCE(DT_INT32)" -tf-output-arrays=p,x -o - | FileCheck %s -check-prefix=CHECK-NO-SHAPE |
| |
| # RUN: tf-mlir-translate -graphdef-to-mlir -tf-enable-shape-inference-on-import=false %s -tf-input-arrays=p,x -tf-input-shapes=512,1024: -tf-input-data-types="DT_INT32,DT_RESOURCE(512,1024:DT_INT32)" -tf-output-arrays=p,x -o - | FileCheck %s -check-prefix=CHECK-SHAPE |
| |
| # Test the handling of the input data types. In particular, if the data type |
| # for an input graph node is specified via command line options, use it. |
| # otherwise, use the data type of the node in the graph. |
| |
| node { |
| name: "p" |
| op: "Placeholder" |
| attr { |
| key: "dtype" |
| value { |
| type: DT_INT32 |
| } |
| } |
| attr { |
| key: "shape" |
| value { |
| shape { |
| unknown_rank: true |
| } |
| } |
| } |
| } |
| node { |
| name: "x" |
| op: "Placeholder" |
| attr { |
| key: "dtype" |
| value { |
| type: DT_RESOURCE |
| } |
| } |
| attr { |
| key: "shape" |
| value { |
| shape { |
| unknown_rank: true |
| } |
| } |
| } |
| } |
| versions { |
| producer: 216 |
| } |
| |
| # CHECK-NO-SHAPE: func @main(%arg0: tensor<i32>, %arg1: tensor<!tf_type.resource<tensor<i32>>>) -> (tensor<i32>, tensor<!tf_type.resource<tensor<i32>>>) |
| |
| # CHECK-SHAPE: func @main(%arg0: tensor<512x1024xi32>, %arg1: tensor<!tf_type.resource<tensor<512x1024xi32>>>) -> (tensor<512x1024xi32>, tensor<!tf_type.resource<tensor<512x1024xi32>>>) |
