deephyper.core.analytics

deephyper.core.analytics

Top-K Configuration

A command line to extract the top-k best configuration from a DeepHyper execution.

It can be used with:

$ deephyper-analytics --help
usage: deephyper-analytics topk [-h] [-k K] [-o OUTPUT] path

positional arguments:
path                  Path to the input CSV file.

optional arguments:
-h, --help            show this help message and exit
-k K                  Number of best configurations to output in decreasing order of best objective.
-o OUTPUT, --output OUTPUT
                        Path to the output file.

An example usage is:

$ deephyper-analytics topk combo_8gpu_8_agebo/infos/results.csv -k 2
'0':
arch_seq: '[229, 0, 22, 1, 1, 53, 29, 1, 119, 1, 0, 116, 123, 1, 273, 0, 1, 388]'
batch_size: 59
elapsed_sec: 10259.2741303444
learning_rate: 0.0001614947
loss: log_cosh
objective: 0.9236862659
optimizer: adam
patience_EarlyStopping: 22
patience_ReduceLROnPlateau: 10
'1':
arch_seq: '[229, 0, 22, 0, 1, 235, 29, 1, 313, 1, 0, 116, 123, 1, 37, 0, 1, 388]'
batch_size: 51
elapsed_sec: 8818.2674164772
learning_rate: 0.0001265946
loss: mae
objective: 0.9231553674
optimizer: nadam
patience_EarlyStopping: 23
patience_ReduceLROnPlateau: 14

An --output argument is also available to save the output in a YAML, JSON or CSV format.

Quick Plot

A tool to have quick and simple visualization from your data.

It can be use such as:

$ deephyper-analytics quickplot nas_big_data/combo/exp_sc21/combo_1gpu_8_age/infos/results.csv
$ deephyper-analytics quickplot save/history/*.json --xy time val_r2
$ deephyper-analytics quickplot save/history/*.json --xy epochs val_r2