batch_timeout_micros = 1000000,
      container = "container",
      shared_name = "shared_name",
      batching_queue = "batching_queue",
      enable_large_batch_splitting = false,
      Tin = [i32],
      Tcaptured = [i32],
      Tout = [i32]} : 1

  %result_2 = tfrt_fallback_async.batch_function device("/device:CPU:0") @matmul_cpu (%a2, %b) {
      num_batch_threads = 1,
      max_batch_size = 4,

    value {
      shape {
      }
    }
  }
  attr {
    key: "shared_name"
    value {
      s: ""
    }
  }
}
node {
  name: "Call"
  op: "PartitionedCall"
  attr {
    key: "Tin"
    value {
      list {
      }
    }
  }
  attr {
    key: "Tout"
    value {
      list {
        type: DT_RESOURCE
      }
    }
  }
  attr {
    key: "f"
    value {

    // CHECK-NEXT: op: "PartitionedCall"
    // CHECK:   func {
    // CHECK:     name: "foo"
    // CHECK:   }
    // CHECK: }
    %0 = tf_executor.island wraps "tf.PartitionedCall"() {Tin = [], Tout = [], config = "", config_proto = "", device = "", executor_type = "", f = @foo, name = "Call_foo"} : () -> ()
    tf_executor.fetch
  }
  func.return
}

// CHECK:      library {
// CHECK:        function {

    }
  }
  attr {
    key: "index"
    value {
      i: 0
    }
  }
}
node {
  name: "spc1"
  op: "StatefulPartitionedCall"
  input: "input0"
  input: "args_0"
  attr {
    key: "Tin"
    value {
      list {
        type: DT_FLOAT
        type: DT_RESOURCE
      }
    }
  }
  attr {
    key: "Tout"
    value {
      list {
        type: DT_FLOAT
        type: DT_RESOURCE

# Những bước đầu tiên

Tệp tin FastAPI đơn giản nhất có thể trông như này:

```Python
{!../../../docs_src/first_steps/tutorial001.py!}
```

Sao chép sang một tệp tin `main.py`.

Chạy live server:

<div class="termy">

```console
$ uvicorn main:app --reload

<span style="color: green;">INFO</span>:     Uvicorn running on http://127.0.0.1:8000 (Press CTRL+C to quit)

      %outputs_0, %control_1 = tf_executor.island wraps "tf.Const"() {device = "", value = dense<0> : tensor<i32>} : () -> tensor<i32>

    tensor<3x3x9x10x!tf_type.qint8>
  %0 = "tf.UniformQuantizedConvolution"(
    %quant_input, %quant_filter, %input_scale, %input_zp,
    %filter_scale, %filter_zp, %accum_scale, %accum_zp
  ) {
    Tin = "tfdtype$DT_QINT8", Tout = "tfdtype$DT_QINT32",
    attr_map = "", batch_group_count = 1 : i64,
    dimension_numbers = "\10\03\1A\02\01\02 \02(\032\02\00\01@\03J\02\01\02",

module attributes {tf.versions = {bad_consumers = [], min_consumer = 12 : i32, producer = 29 : i32}} {
  func.func @main() {
    tf_executor.graph {
      %0 = tf_executor.island wraps "tf.PartitionedCall"() {Tin = [], Tout = [], config = "", config_proto = "", device = "", executor_type = "", f = @foo, name = "Call_foo"} : () -> ()
      tf_executor.fetch
    }
    func.return
  }
  func.func @foo() {
    tf_executor.graph {

    %conv_out = "tf.UniformQuantizedConvolution"(%input, %filter,
                                %input_scale, %input_zp, %filter_scale, %filter_zp, %out_scale, %out_zp) {
        Tin = "tfdtype$DT_QINT8",
        Tout = "tfdtype$DT_QINT32",
        window_strides = [1, 1],
        padding = "SAME",
        explicit_padding = [],
        lhs_dilation = [],
        rhs_dilation = [],

  }
}
node {
  name: "StatefulIf"
  op: "If"
  input: "tf.Less"
  input: "a"
  input: "b"
  attr {
    key: "Tcond"
    value {
      type: DT_BOOL
    }
  }
  attr {
    key: "Tin"
    value {
      list {
        type: DT_FLOAT
        type: DT_FLOAT
      }
    }
  }
  attr {
    key: "Tout"
    value {
      list {
        type: DT_FLOAT
      }
    }
  }