std::vector<TF_Output> inputs_;
  std::vector<TF_Tensor*> input_values_;
  std::vector<TF_Output> outputs_;
  std::vector<TF_Tensor*> output_values_;
  std::vector<TF_Operation*> targets_;
};

using ImmediateTensorHandlePtr =
    std::unique_ptr<ImmediateExecutionTensorHandle,
                    internal::ImmediateExecutionTensorHandleDeleter>;

}  // namespace tensorflow

    }
  }
};
}  // namespace internal

using AbstractOperationPtr =
    std::unique_ptr<AbstractOperation, internal::AbstractOperationDeleter>;

}  // namespace tensorflow

    ::testing::Combine(::testing::Values("graphdef", "mlir"),
                       /*tfrt*/ ::testing::Values(false),
                       /*executing_eagerly*/ ::testing::Values(true, false)));
#endif
}  // namespace
}  // namespace internal
}  // namespace gradients

      absl::Span<AbstractTensorHandle* const> output_gradients,
      absl::Span<AbstractTensorHandle*> result);
};

}  // namespace gradients
}  // namespace tensorflow

<div>
<table width="100%" border="0" cellpadding="0" cellspacing="0" style="padding:0;margin:0;line-height:1px;font-size:1px;">
<!--<![endif]-->
<tbody>
<tr>
<td width="50" style="width:50px;padding:0;margin:0;line-height:1px;font-size:1px;" class="margins"></td>
<td align="center" style="padding:0;margin:0;line-height:1px;font-size:1px;">

                                                          size_t proto_len,
                                                          TF_Status* status);

#ifdef __cplusplus
} /* end extern "C" */
#endif

    ctx.device_type = DLDeviceType::kDLCPU;
  } else if (device_type == "GPU") {
#if TENSORFLOW_USE_ROCM
    ctx.device_type = DLDeviceType::kDLROCM;
#else
    ctx.device_type = DLDeviceType::kDLCUDA;
#endif
  } else {
    status->status = tensorflow::errors::InvalidArgument(
        "Unsupported Device Type for dlpack");
  }

  return ctx;
}

// Converts DLDevice to TF device name.

{% endfor -%}
{%- for sponsor in sponsors.silver -%}
<a href="{{ sponsor.url }}" target="_blank" title="{{ sponsor.title }}"><img src="{{ sponsor.img }}"></a>
{% endfor %}
{% endif %}
"""


def generate_readme_content() -> str:
    en_index = en_docs_path / "docs" / "index.md"
    content = en_index.read_text("utf-8")
    match_pre = re.search(r"</style>\n\n", content)

  }
  return ParallelTensor::FromTensorHandles(*this, std::move(components),
                                           status);
}

}  // namespace parallel_device
}  // namespace tensorflow