(self, state)
| 273 | |
| 274 | |
| 275 | def act(self, state): |
| 276 | if np.random.rand() <= self.epsilon: |
| 277 | return np.random.choice(self.action_size) |
| 278 | act_values = self.model.predict(state, verbose=0) |
| 279 | return np.argmax(act_values[0]) # returns action |
| 280 | |
| 281 | @tf.function |
| 282 | def replay(self, batch_size=32): |
no test coverage detected