# tf_agents.drivers.dynamic_step_driver.DynamicStepDriver

<br />

|---|
| [![](https://www.tensorflow.org/images/GitHub-Mark-32px.png) View source on GitHub](https://github.com/tensorflow/agents/blob/v0.19.0/tf_agents/drivers/dynamic_step_driver.py#L47-L224) |

A driver that takes N steps in an environment using a tf.while_loop.

Inherits From: [`Driver`](https://www.tensorflow.org/agents/api_docs/python/tf_agents/drivers/driver/Driver)

    tf_agents.drivers.dynamic_step_driver.DynamicStepDriver(
        env, policy, observers=None, transition_observers=None, num_steps=1
    )

### Used in the notebooks

| Used in the tutorials |
|---|
| - [Checkpointer and PolicySaver](https://www.tensorflow.org/agents/tutorials/10_checkpointer_policysaver_tutorial) - [Replay Buffers](https://www.tensorflow.org/agents/tutorials/5_replay_buffers_tutorial) - [Tutorial on Multi Armed Bandits in TF-Agents](https://www.tensorflow.org/agents/tutorials/bandits_tutorial) - [A Tutorial on Multi-Armed Bandits with Per-Arm Features](https://www.tensorflow.org/agents/tutorials/per_arm_bandits_tutorial) - [Tutorial on Ranking in TF-Agents](https://www.tensorflow.org/agents/tutorials/ranking_tutorial) |

The while loop will run num_steps in the environment, only counting steps that
result in an environment transition, i.e. (time_step, action, next_time_step).
If a step results in environment resetting, i.e. time_step.is_last() and
next_time_step.is_first() (traj.is_boundary()), this is not counted toward the
num_steps.

As environments run batched time_steps, the counters for all batch elements
are summed, and execution stops when the total exceeds num_steps. When
batch_size \> 1, there is no guarantee that exactly num_steps are taken -- it
may be more but never less.

This termination condition can be overridden in subclasses by implementing the
self._loop_condition_fn() method.

<br />

<br />

<br />

| ## Args ||
|---|---|
| `env` | A tf_environment.Base environment. |
| `policy` | A tf_policy.TFPolicy policy. |
| `observers` | A list of observers that are updated after every step in the environment. Each observer is a callable(time_step.Trajectory). |
| `transition_observers` | A list of observers that are updated after every step in the environment. Each observer is a callable((TimeStep, PolicyStep, NextTimeStep)). |
| `num_steps` | The number of steps to take in the environment. For batched or parallel environments, this is the total number of steps taken summed across all environments. |

<br />

<br />

<br />

<br />

| ## Raises ||
|---|---|
| `ValueError` | If env is not a tf_environment.Base or policy is not an instance of tf_policy.TFPolicy. |

<br />

<br />

<br />

<br />

| ## Attributes ||
|---|---|
| `env` | <br /> <br /> |
| `info_observers` | <br /> <br /> |
| `observers` | <br /> <br /> |
| `policy` | <br /> <br /> |
| `transition_observers` | <br /> <br /> |

<br />

## Methods

### `run`

[View source](https://github.com/tensorflow/agents/blob/v0.19.0/tf_agents/drivers/dynamic_step_driver.py#L176-L197)

    run(
        time_step=None, policy_state=None, maximum_iterations=None
    )

Takes steps in the environment using the policy while updating observers.

<br />

<br />

<br />

| Args ||
|---|---|
| `time_step` | optional initial time_step. If None, it will use the current_time_step of the environment. Elements should be shape \[batch_size, ...\]. |
| `policy_state` | optional initial state for the policy. |
| `maximum_iterations` | Optional maximum number of iterations of the while loop to run. If provided, the cond output is AND-ed with an additional condition ensuring the number of iterations executed is no greater than maximum_iterations. |

<br />

<br />

<br />

<br />

| Returns ||
|---|---|
| `time_step` | TimeStep named tuple with final observation, reward, etc. |
| `policy_state` | Tensor with final step policy state. |

<br />