* Get actions based on a state-tensor input. * * @param {tf.Tensor} inputs A tf.Tensor instance of shape `[batchSize, 4]`. * @param {Float32Array} inputs The actions for the inputs, with length * `batchSize`.
(inputs)
| 222 | * `batchSize`. |
| 223 | */ |
| 224 | getActions(inputs) { |
| 225 | return this.getLogitsAndActions(inputs)[1].dataSync(); |
| 226 | } |
| 227 | |
| 228 | /** |
| 229 | * Push a new dictionary of gradients into records. |
no test coverage detected