regal

REGAL

This dataset contains dataflow computational graphs generated procedurally, intended for training and evaluating algorithms that optimize execution (e.g. placement and scheduling), in TensorFlow's CostGraphDef protocol buffer format and encoded as text.

Original paper REGAL: Transfer Learning For Fast Optimization of Computation Graphs (Paliwal, Gimeno, Nair, Li, Lubin, Kohli, Vinyals)

Folder structure

There are 10000 training graphs, 1000 validation graphs and 1000 test graphs. The file names follow the format of "graph_" plus a hash of the graph topology plus ".pbtxt".

Filtering

For each set (train, valid, test) there are not two graphs with the same topology. We used the Biased Random Key Genetic Algorithm (BRKGA) to filter out graphs that did not have "room for improvement"

"Room for improvement" was defined as the union of two conditions:

• if BRKGA with a low fitness evaluation limit (number of calls to fitness function) did not fit the hardware constraints and BRKGA with a high number did.
• if BRKGA with a high fitness evaluation limit was 20% faster in running time that BRKGA with a low number.

Example Graph

node {
  name: "_SOURCE"
}
node {
  name: "node_0"
  id: 1
  control_input: 0
}
node {
  name: "node_1"
  id: 2
  output_info {
    size: 70
    alias_input_port: -1
  }
  control_input: 0
  compute_cost: 58
}
node {
  name: "node_2"
  id: 3
  output_info {
    size: 52
    alias_input_port: -1
  }
  control_input: 0
  compute_cost: 47
}
node {
  name: "node_3"
  id: 4
  input_info {
    preceding_node: 2
  }
  output_info {
    size: 55
    alias_input_port: -1
  }
  control_input: 0
  compute_cost: 58
}

Dataset Location

The dataset is available in the following link

Dataset Metadata

The following table is necessary for this dataset to be indexed by search engines such as Google Dataset Search.

property	value
name	REGAL CostGraphDef Synthetic Dataset
url	https://github.com/deepmind/deepmind-research/tree/master/regal
sameAs	https://github.com/deepmind/deepmind-research/tree/master/regal
description	This dataset contains dataflow computational graphs generated procedurally, intended for training and evaluating algorithms that optimize execution (e.g. placement and scheduling), in [TensorFlow's CostGraphDef](https://github.com/tensorflow/tensorflow/blob/59ee7f9138482d85cd93c004aca961bea35820c7/tensorflow/core/framework/cost_graph.proto#L12) [protocol buffer](https://en.wikipedia.org/wiki/Protocol_Buffers) format and encoded as [text](https://developers.google.com/protocol-buffers/docs/reference/cpp/google.protobuf.text_format).
provider	property value name DeepMind sameAs https://en.wikipedia.org/wiki/DeepMind
citation	https://identifiers.org/arxiv:1905.02494

Disclaimer

This is not an officially supported Google product.