Fix program constructed by programdesc incomplete bug

[chenwhql]

PR types

Bug fixes

PR changes

Others

Describe

Fix program constructed by programdesc incomplete bug

only update var name is not enough, need use BlockDesc._rename_var update var info.

original:

the var linear_0.w_0.load_0 and linear_0.b_0.load_0

{ // block 0
    var linear_1.tmp_0 : fluid.VarType.LOD_TENSOR.shape(16, 10).astype(VarType.FP32)
    var linear_1.tmp_1 : fluid.VarType.LOD_TENSOR.shape(16, 10).astype(VarType.FP32)
    var linear_0.w_0.load_0 : fluid.VarType.LOD_TENSOR.shape().astype(VarType.BOOL)
    var translated_layer/scale_0.tmp_0 : fluid.VarType.LOD_TENSOR.shape(16, 10).astype(VarType.FP32)
    var linear_0.b_0.load_0 : fluid.VarType.LOD_TENSOR.shape().astype(VarType.BOOL)
    var generated_var_0 : fluid.VarType.LOD_TENSOR.shape(16, 784).astype(VarType.FP32)

    {Out=['linear_1.tmp_0']} = matmul(inputs={X=['generated_var_0'], Y=['linear_0.w_0.load_0']}, Scale_out = 1.0, Scale_x = 1.0, Scale_y = 1.0, alpha = 1.0, force_fp32_output = False, fused_reshape_Out = [], fused_reshape_X = [], fused_reshape_Y = [], fused_transpose_Out = [], fused_transpose_X = [], fused_transpose_Y = [], mkldnn_data_type = float32, op_device = , op_namescope = /, op_role = 0, op_role_var = [], transpose_X = False, transpose_Y = False, use_mkldnn = False, use_quantizer = False)
    {Out=['linear_1.tmp_1']} = elementwise_add(inputs={X=['linear_1.tmp_0'], Y=['linear_0.b_0.load_0']}, Scale_out = 1.0, Scale_x = 1.0, Scale_y = 1.0, axis = 1, mkldnn_data_type = float32, op_device = , op_namescope = /, op_role = 0, op_role_var = [], use_mkldnn = False, use_quantizer = False, x_data_format = , y_data_format = )
    {Out=['translated_layer/scale_0.tmp_0']} = scale(inputs={ScaleTensor=[], X=['linear_1.tmp_1']}, bias = 0.0, bias_after_scale = True, op_device = , op_namescope = /, op_role = 0, op_role_var = [], scale = 1.0)
}

• fixed:

{ // block 0
    persist var linear_0.b_0.load_0 : fluid.VarType.LOD_TENSOR.shape(10,).astype(VarType.FP32)
    var linear_1.tmp_0 : fluid.VarType.LOD_TENSOR.shape(16, 10).astype(VarType.FP32)
    var linear_1.tmp_1 : fluid.VarType.LOD_TENSOR.shape(16, 10).astype(VarType.FP32)
    persist var linear_0.w_0.load_0 : fluid.VarType.LOD_TENSOR.shape(784, 10).astype(VarType.FP32)
    var translated_layer/scale_0.tmp_0 : fluid.VarType.LOD_TENSOR.shape(16, 10).astype(VarType.FP32)
    var generated_var_0 : fluid.VarType.LOD_TENSOR.shape(16, 784).astype(VarType.FP32)

    {Out=['linear_1.tmp_0']} = matmul(inputs={X=['generated_var_0'], Y=['linear_0.w_0.load_0']}, Scale_out = 1.0, Scale_x = 1.0, Scale_y = 1.0, alpha = 1.0, force_fp32_output = False, fused_reshape_Out = [], fused_reshape_X = [], fused_reshape_Y = [], fused_transpose_Out = [], fused_transpose_X = [], fused_transpose_Y = [], mkldnn_data_type = float32, op_device = , op_namescope = /, op_role = 0, op_role_var = [], transpose_X = False, transpose_Y = False, use_mkldnn = False, use_quantizer = False)
    {Out=['linear_1.tmp_1']} = elementwise_add(inputs={X=['linear_1.tmp_0'], Y=['linear_0.b_0.load_0']}, Scale_out = 1.0, Scale_x = 1.0, Scale_y = 1.0, axis = 1, mkldnn_data_type = float32, op_device = , op_namescope = /, op_role = 0, op_role_var = [], use_mkldnn = False, use_quantizer = False, x_data_format = , y_data_format = )
    {Out=['translated_layer/scale_0.tmp_0']} = scale(inputs={ScaleTensor=[], X=['linear_1.tmp_1']}, bias = 0.0, bias_after_scale = True, op_device = , op_namescope = /, op_role = 0, op_role_var = [], scale = 1.0)
}

[paddle-bot-old]

Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

[Channingss]

LGTM

[liym27]

LGTM