# Using the config.toml File¶

Admins can edit a config.toml file when starting the Driverless AI Docker image. The config.toml file includes all possible configuration options that would otherwise be specified in the nvidia-docker run command. This file is located in a folder on the container. You can make updates to environment variables directly in this file. Driverless AI will use the updated config.toml file when starting from native installs. Docker users can specify that updated config.toml file when starting Driverless AI Docker image.

## Configuration Override Chain¶

The configuration engine reads and overrides variables in the following order:

1. h2oai/config/config.toml - This is an internal file that is not visible.

2. config.toml - Place this file in a folder or mount it in a Docker container and specify the path in the “DRIVERLESS_AI_CONFIG_FILE” environment variable.

3. Environment variable - Configuration variables can also be provided as environment variables. They must have the prefix DRIVERLESS_AI_ followed by the variable name in all caps. For example, “authentication_method” can be provided as “DRIVERLESS_AI_AUTHENTICATION_METHOD”.

## Docker Image Users¶

1. Copy the config.toml file from inside the Docker image to your local filesystem.

# Make a config directory
mkdir config

# Copy the config.toml file to the new config directory.
nvidia-docker run \
--pid=host \
--rm \
--init \
-u id -u:id -g \
-v pwd/config:/config \
--entrypoint bash \
h2oai/dai-centos7-x86_64:TAG
-c "cp /etc/dai/config.toml /config"

1. Edit the desired variables in the config.toml file. Save your changes when you are done.

2. Start Driverless AI with the DRIVERLESS_AI_CONFIG_FILE environment variable. Make sure this points to the location of the edited config.toml file so that the software finds the configuration file.

nvidia-docker run \
--pid=host \
--init \
--rm \
--shm-size=256m \
-u id -u:id -g \
-p 12345:12345 \
-e DRIVERLESS_AI_CONFIG_FILE="/config/config.toml" \
-v pwd/config:/config \
-v pwd/data:/data \
-v pwd/log:/log \
-v pwd/license:/license \
-v pwd/tmp:/tmp \
h2oai/dai-centos7-x86_64:TAG


## Native Install Users¶

Native installs include DEBs, RPMs, and TAR SH installs.

1. Export the Driverless AI config.toml file or add it to ~/.bashrc. For example:

export DRIVERLESS_AI_CONFIG_FILE=“/config/config.toml”

1. Edit the desired variables in the config.toml file. Save your changes when you are done.

2. Start Driverless AI. Note that the command used to start Driverless AI varies depending on your install type.

For reference, below is a copy of the standard config.toml file included with this version of Driverless AI. The sections that follow describe some examples showing how to set different environment variables, data connectors, authentication, and notifications.

## Sample Config.toml File¶

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270  ############################################################################## # DRIVERLESS AI CONFIGURATION FILE # # Comments: # This file is authored in TOML (see https://github.com/toml-lang/toml) # # Config Override Chain # Configuration variables for Driverless AI can be provided in several ways, # the config engine reads and overrides variables in the following order # # 1. h2oai/config/config.toml # [internal not visible to users] # # 2. config.toml # [place file in a folder/mount file in docker container and provide path # in "DRIVERLESS_AI_CONFIG_FILE" environment variable] # # 3. Environment variable # [configuration variables can also be provided as environment variables # they must have the prefix "DRIVERLESS_AI_" followed by # variable name in caps e.g "authentication_method" can be provided as # "DRIVERLESS_AI_AUTHENTICATION_METHOD"] ############################################################################## # Whether to allow user to change non-server toml parameters per experiment in expert page. #allow_config_overrides_in_expert_page = true # Every *.toml file is read from this directory and process the same way as main config file. #user_config_directory = "" # IP address and port of autoviz process. #vis_server_ip = "127.0.0.1" # IP and port of autoviz process. #vis_server_port = 12346 # IP address and port of procsy process. #procsy_ip = "127.0.0.1" # IP address and port of procsy process. #procsy_port = 12347 # IP address and port of H2O instance. #h2o_ip = "127.0.0.1" # IP address and port of H2O instance for use by MLI. #h2o_port = 12348 # Enable h2o recipes server. #enable_h2o_recipes = true # URL of H2O instance for use by transformers, models, or scorers. #h2o_recipes_url = "None" # IP of H2O instance for use by transformers, models, or scorers. #h2o_recipes_ip = "None" # Port of H2O instance for use by transformers, models, or scorers. No other instances must be on that port or on next port. #h2o_recipes_port = 50341 # Name of H2O instance for use by transformers, models, or scorers. #h2o_recipes_name = "None" # Number of threads for H2O instance for use by transformers, models, or scorers. #h2o_recipes_nthreads = 4 # Log Level of H2O instance for use by transformers, models, or scorers. #h2o_recipes_log_level = "None" # Maximum memory size of H2O instance for use by transformers, models, or scorers. #h2o_recipes_max_mem_size = "None" # Minimum memory size of H2O instance for use by transformers, models, or scorers. #h2o_recipes_min_mem_size = "None" # General user overrides of kwargs dict to pass to h2o.init() for recipe server #h2o_recipes_kwargs = "{}" # IP address and port for Driverless AI HTTP server. #ip = "127.0.0.1" # IP address and port for Driverless AI HTTP server. #port = 12345 # A list of two integers indicating the port range to search over, and dynamically find an open port to bind to (e.g., [11111,20000]). #port_range = "[]" # File upload limit (default 100GB) #max_file_upload_size = 104857600000 # Verbosity of logging # 0: quiet (CRITICAL, ERROR, WARNING) # 1: default (CRITICAL, ERROR, WARNING, INFO, DATA) # 2: verbose (CRITICAL, ERROR, WARNING, INFO, DATA, DEBUG) # Affects server and all experiments #log_level = 1 # Whether to collect relevant server logs (h2oai_server.log, dai.log from systemctl or docker, and h2o log) # Useful for when sending logs to H2O.ai #collect_server_logs_in_experiment_logs = false # Redis settings #redis_ip = "127.0.0.1" # Redis settings #redis_port = 6379 # Redis settings #master_redis_password = "" # https settings # You can make a self-signed certificate for testing with the following commands: # sudo openssl req -x509 -newkey rsa:4096 -keyout private_key.pem -out cert.pem -days 3650 -nodes -subj '/O=Driverless AI' # sudo chown dai:dai cert.pem private_key.pem # sudo chmod 600 cert.pem private_key.pem # sudo mv cert.pem private_key.pem /etc/dai #enable_https = false # https settings # You can make a self-signed certificate for testing with the following commands: # sudo openssl req -x509 -newkey rsa:4096 -keyout private_key.pem -out cert.pem -days 3650 -nodes -subj '/O=Driverless AI' # sudo chown dai:dai cert.pem private_key.pem # sudo chmod 600 cert.pem private_key.pem # sudo mv cert.pem private_key.pem /etc/dai #ssl_key_file = "/etc/dai/private_key.pem" # https settings # You can make a self-signed certificate for testing with the following commands: # sudo openssl req -x509 -newkey rsa:4096 -keyout private_key.pem -out cert.pem -days 3650 -nodes -subj '/O=Driverless AI' # sudo chown dai:dai cert.pem private_key.pem # sudo chmod 600 cert.pem private_key.pem # sudo mv cert.pem private_key.pem /etc/dai #ssl_crt_file = "/etc/dai/cert.pem" # SSL TLS #ssl_no_sslv2 = true # SSL TLS #ssl_no_sslv3 = true # SSL TLS #ssl_no_tlsv1 = true # SSL TLS #ssl_no_tlsv1_1 = true # SSL TLS #ssl_no_tlsv1_2 = false # SSL TLS #ssl_no_tlsv1_3 = false # https settings # Sets the client verification mode. # CERT_NONE: Client does not need to provide the certificate and if it does any # verification errors are ignored. # CERT_OPTIONAL: Client does not need to provide the certificate and if it does # certificate is verified agains set up CA chains. # CERT_REQUIRED: Client needs to provide a certificate and certificate is # verified. # You'll need to set 'ssl_client_key_file' and 'ssl_client_crt_file' # When this mode is selected for Driverless to be able to verify # it's own callback requests. # #ssl_client_verify_mode = "CERT_NONE" # https settings # Path to the Certification Authority certificate file. This certificate will be # used when to verify client certificate when client authentication is turned on. # If this is not set, clients are verified using default system certificates. # #ssl_ca_file = "" # https settings # path to the private key that Driverless will use to authenticate itself when # CERT_REQUIRED mode is set. # #ssl_client_key_file = "" # https settings # path to the client certificate that Driverless will use to authenticate itself # when CERT_REQUIRED mode is set. # #ssl_client_crt_file = "" # Data directory. All application data and files related datasets and # experiments are stored in this directory. #data_directory = "./tmp" # Whether to run quick performance benchmark at start of application #enable_quick_benchmark = true # Whether to run extended performance benchmark at start of application #enable_extended_benchmark = false # Scaling factor for number of rows for extended performance benchmark. For rigorous performance benchmarking, # values of 1 or larger are recommended. #extended_benchmark_scale_num_rows = 0.1 # Whether to run quick startup checks at start of application #enable_startup_checks = true # Whether to opt in to usage statistics and bug reporting #usage_stats_opt_in = false # authentication_method # unvalidated : Accepts user id and password. Does not validate password. # none: Does not ask for user id or password. Authenticated as admin. # openid: Users OpenID Connect provider for authentication. See additional OpenID settings below. # pam: Accepts user id and password. Validates user with operating system. # ldap: Accepts user id and password. Validates against an ldap server. Look # for additional settings under LDAP settings. # local: Accepts a user id and password. Validated against an htpasswd file provided in local_htpasswd_file. # ibm_spectrum_conductor: Authenticate with IBM conductor auth api. # tls_certificate: Authenticate with Driverless by providing a TLS certificate. # #authentication_method = "unvalidated" # default amount of time in hours before we force user to login again (if not provided by authentication_method) #authentication_default_timeout_hours = 72 # OpenID Connect Settings: # Refer to OpenID Connect Basic Client Implementation Guide for details on how OpenID authentication flow works # https://openid.net/specs/openid-connect-basic-1_0.html # base server uri to the OpenID Provider server (ex: https://oidp.ourdomain.com #auth_openid_provider_base_uri = "" # uri to pull OpenID config data from (you can extract most of required OpenID config from this url) # usually located at: /auth/realms/master/.well-known/openid-configuration #auth_openid_configuration_uri = "" # uri to start authentication flow #auth_openid_auth_uri = "" # uri to make request for token after callback from OpenID server was received #auth_openid_token_uri = "" # uri to get user information once access_token has been acquired (ex: list of groups user belongs to will be provided here) #auth_openid_userinfo_uri = "" # uri to logout user #auth_openid_logout_uri = "" # callback uri that OpenID provide will use to send 'authentication_code' # This is OpenID callback endpoint in Driverless AI. Most OpenID providers need this to be HTTPs. # (ex. https://driverless.ourdomin.com/openid/callback) #auth_openid_redirect_uri = "" # OAuth2 grant type (usually authorization_code for OpenID, can be access_token also) #auth_openid_grant_type = "" # OAuth2 response type (usually code) #auth_openid_response_type = "" # Client ID registered with OpenID provider #auth_openid_client_id = "" # Client secret provided by OpenID provider when registering Client ID #auth_openid_client_secret = "" # Scope of info (usually openid). Can be list of more than one, space delimited, possible # values listed at https://openid.net/specs/openid-connect-basic-1_0.html#Scopes #auth_openid_scope = "" # What key in user_info json should we check to authorize user #auth_openid_userinfo_auth_key = "" # What value should the key have in user_info json in order to authorize user #auth_openid_userinfo_auth_value = "" # Key that specifies username in user_info json (we will use the value of this key as username in Driverless AI) #auth_openid_userinfo_username_key = "" # Quote method from urllib.parse used to encode payload dict in Authentication Request #auth_openid_urlencode_quote_via = "quote" # Key in Token Response JSON that holds the value for access token expiry #auth_openid_access_token_expiry_key = "expires_in" # Key in Token Response JSON that holds the value for access token expiry #auth_openid_refresh_token_expiry_key = "refresh_expires_in" # Expiration time in seconds for access token #auth_openid_token_expiration_secs = 3600 # Enables advanced matching for OpenID Connect authentication. # When enabled ObjectPath () expression is used to # evaluate the user identity. # #auth_openid_use_objectpath_match = false # ObjectPath () expression that will be used # to evaluate whether user is allowed to login into Driverless. # Any expression that evaluates to True means user is allowed to log in. # Examples: # Simple claim equality: \$.our_claim is "our_value" # List of claims contains required value: "expected_role" in @.roles # #auth_openid_use_objectpath_expression = "" # ldap server domain or ip #ldap_server = "" # ldap server port #ldap_port = "" # Complete DN of the LDAP bind user #ldap_bind_dn = "" # Password for the LDAP bind #ldap_bind_password = "" # Provide Cert file location #ldap_tls_file = "" # use true to use ssl or false #ldap_use_ssl = false # the location in the DIT where the search will start #ldap_search_base = "" # A string that describes what you are searching for. You can use Pythonsubstitution to have this constructed dynamically.(only {{DAI_USERNAME}} is supported) #ldap_search_filter = "" # ldap attributes to return from search #ldap_search_attributes = "" # specify key to find user name #ldap_user_name_attribute = "" # When using this recipe, needs to be set to "1" #ldap_recipe = "0" # Deprecated do not use #ldap_user_prefix = "" # Deprecated, Use ldap_bind_dn #ldap_search_user_id = "" # Deprecated, ldap_bind_password #ldap_search_password = "" # Deprecated, use ldap_search_base instead #ldap_ou_dn = "" # Deprecated, use ldap_base_dn #ldap_dc = "" # Deprecated, use ldap_search_base #ldap_base_dn = "" # Deprecated, use ldap_search_filter #ldap_base_filter = "" # Path to the CRL file that will be used to verify client certificate. #auth_tls_crl_file = "" # What field of the subject would used as source for username or other values used for further validation. #auth_tls_subject_field = "CN" # Regular expression that will be used to parse subject field to obtain the username or other values used for further validation. #auth_tls_field_parse_regexp = "(?P.*)" # Sets up the way how user identity would be obtained # REGEXP_ONLY: Will use 'auth_tls_subject_field' and 'auth_tls_field_parse_regexp' # to extract the username from the client certificate. # LDAP_LOOKUP: Will use LDAP server to lookup for the username. # 'ldap_server', 'ldap_use_ssl', 'ldap_tls_file', 'ldap_bind_dn', # 'ldap_bind_password' options are used to establish # the connection with the LDAP server. # 'auth_tls_subject_field' and 'auth_tls_field_parse_regexp' # options are used to parse the certificate. # 'ldap_search_base', 'ldap_search_filter', and # 'ldap_username_attribute' options are used to do the lookup. # 'ldap_search_filter' can be built dynamically using the named # capturing groups from the 'auth_tls_field_parse_regexp' for # substitution. # Example: # auth_tls_field_parse_regexp = "\w+ (?P\d+)" # ldap_search_filter = "(&(objectClass=person)(id={{id}}))" # #auth_tls_user_lookup = "REGEXP_ONLY" # Sets optional additional lookup filter that is performed after the # user is found. This can be used for example to check whether the is member of # particular group. # Filter can be built dynamically from the attributes returned by the lookup. # Authorization fails when search does not return any entry. If one ore more # entries are returned authorization succeeds. # Example: # auth_tls_field_parse_regexp = "\w+ (?P\d+)" # ldap_search_filter = "(&(objectClass=person)(id={{id}}))" # auth_tls_ldap_authorization_lookup_filter = "(&(objectClass=group)(member=uid={{uid}},dc=example,dc=com))" # If this option is empty no additional lookup is done and just a successful user # lookup is enough to authorize the user. # #auth_tls_ldap_authorization_lookup_filter = "" # Base DN where to start the Authorization lookup. Used when 'auth_tls_ldap_authorization_lookup_filter' is set. #auth_tls_ldap_authorization_search_base = "" # Local password file # Generating a htpasswd file: see syntax below # htpasswd -B '' '' # note: -B forces use of brcypt, a secure encryption method #local_htpasswd_file = "" # Supported file formats (file name endings must match for files to show up in file browser) #supported_file_types = "csv, tsv, txt, dat, tgz, gz, bz2, zip, xz, xls, xlsx, jay, feather, bin, arff, parquet, pkl, orc" # Supported file formats of data recipe files (file name endings must match for files to show up in file browser) #recipe_supported_file_types = "py, pyc" # By default, only supported file types (based on the file extensions listed above) will be listed for import into DAI # Some data pipelines generate parquet files without any extensions. Enabling the below option will cause files # without an extension to be listed in the file import dialog. # DAI will import files without extensions as parquet files; if cannot be imported, an error is generated # #list_files_without_extensions = false # File System Support # upload : standard upload feature # file : local file system/server file system # hdfs : Hadoop file system, remember to configure the HDFS config folder path and keytab below # dtap : Blue Data Tap file system, remember to configure the DTap section below # s3 : Amazon S3, optionally configure secret and access key below # gcs : Google Cloud Storage, remember to configure gcs_path_to_service_account_json below # gbq : Google Big Query, remember to configure gcs_path_to_service_account_json below # minio : Minio Cloud Storage, remember to configure secret and access key below # snow : Snowflake Data Warehouse, remember to configure Snowflake credentials below (account name, username, password) # kdb : KDB+ Time Series Database, remember to configure KDB credentials below (hostname and port, optionally: username, password, classpath, and jvm_args) # azrbs : Azure Blob Storage, remember to configure Azure credentials below (account name, account key) # jdbc: JDBC Connector, remember to configure JDBC below. (jdbc_app_configs) # recipe_file: Custom recipe file upload # recipe_url: Custom recipe upload via url # #enabled_file_systems = "upload, file, hdfs, s3, recipe_file, recipe_url" #max_files_listed = 100 # do_not_log_list : add configurations that you do not wish to be recorded in logs here #do_not_log_list = "local_htpasswd_file, aws_access_key_id, aws_secret_access_key, snowflake_password, snowflake_url, snowflake_user, snowflake_account, minio_endpoint_url, minio_access_key_id, minio_secret_access_key, kdb_user, kdb_password, ldap_bind_password, gcs_path_to_service_account_json, azure_blob_account_name, azure_blob_account_key, deployment_aws_access_key_id, deployment_aws_secret_access_key, master_minio_access_key_id, master_minio_secret_access_key, master_redis_password, auth_openid_client_id, auth_openid_client_secret, auth_openid_userinfo_auth_key, auth_openid_userinfo_auth_value, auth_openid_userinfo_username_key" # Minio is used for file distribution on multinode architecture. # These settings are used to specify the local Minio connection to master nodes. #master_minio_address = ":" # Minio is used for file distribution on multinode architecture. # These settings are used to specify the local Minio connection to master nodes. #master_minio_access_key_id = "" # Minio is used for file distribution on multinode architecture. # These settings are used to specify the local Minio connection to master nodes. #master_minio_secret_access_key = "" # Allow using browser localstorage, to improve UX. #allow_localstorage = true # Allow original dataset columns to be present in downloaded predictions CSV #allow_orig_cols_in_predictions = true # If the experiment is not done after this many minutes, stop feature engineering and model tuning as soon as possible and proceed with building the final modeling pipeline and deployment artifacts, independent of model score convergence or pre-determined number of iterations. Only active is not in reproducible mode. Depending on the data and experiment settings, overall experiment runtime can differ significantly from this setting. #max_runtime_minutes = 1440 # If the experiment is not done after this many minutes, push the abort button. Preserves experiment artifacts made so far for summary and log zip files, but further artifacts are made. #max_runtime_minutes_until_abort = 10080 # Recipe type # Recipes override any GUI settings # 'auto' : all models and features automatically determined by experiment settings, toml settings, and feature_engineering_effort # 'compliant' : like 'auto' except: # * interpretability=10 (to avoid complexity, overrides GUI or python client chose for interpretability) # * enable_glm='on' (rest 'off', to avoid complexity and be compatible with algorithms supported by MLI) # * fixed_ensemble_level=0: Don't use any ensemble (to avoid complexity) # * feature_brain_level=0: No feature brain used (to ensure every restart is identical) # * max_feature_interaction_depth=1: interaction depth is set to 1 (no multi-feature interactions to avoid complexity) # * target_transformer='identity': for regression (to avoid complexity) # * check_distribution_shift_drop='off': Don't use distribution shift between train, valid, and test to drop features (bit risky without fine-tuning # 'kaggle' : like 'auto' except: # * external validation set is concatenated with train set, with target marked as missing # * test set is concatenated with train set, with target marked as missing # * transformers that do not use the target are allowed to fit_transform across entire train + validation + test # * several config toml expert options open-up limits (e.g. more numerics are treated as categoricals) # Note: If plentiful memory, can: # * choose kaggle mode and then change fixed_feature_interaction_depth to large negative number, # otherwise default number of features given to transformer is limited to 50 by default # * choose mutation_mode = "full", so even more types are transformations are done at once per transformer # #recipe = "auto" # How much effort to spend on feature engineering (0...10) # Heuristic combination of various developer-level toml parameters # 0 : keep only numeric features, only model tuning during evolution # 1 : keep only numeric features and frequency-encoded categoricals, only model tuning during evolution # 2 : Like #1 but instead just no Text features. Some feature tuning before evolution. # 3 : Like #5 but only tuning during evolution. Mixed tuning of features and model parameters. # 4 : Like #5, but slightly more focused on model tuning # 5 : Default. Balanced feature-model tuning # 6-7 : Like #5, but slightly more focused on feature engineering # 8 : Like #6-7, but even more focused on feature engineering with high feature generation rate, no feature dropping even if high interpretability # 9-10: Like #8, but no model tuning during feature evolution #feature_engineering_effort = 5 # Whether to enable train/valid and train/test distribution shift detection ('auto'/'on'/'off') #check_distribution_shift = "auto" # Whether to drop high-shift features ('auto'/'on'/'off'). Auto disables for time series. #check_distribution_shift_drop = "auto" # If distribution shift detection is enabled, drop features (except ID, text, date/datetime, time, weight) for # which shift AUC, GINI, or Spearman correlation is above this value # (e.g. AUC of a binary classifier that predicts whether given feature value # belongs to train or test data) #drop_features_distribution_shift_threshold_auc = 0.999 # Whether to check leakage for each feature (True/False). # If fold column, this checks leakage without fold column used. #check_leakage = "auto" # If leakage detection is enabled, # drop features for which AUC (R2 for regression), GINI, # or Spearman correlation is above this value. # If fold column present, features are not dropped, # because leakage test applies without fold column used. # #drop_features_leakage_threshold_auc = 0.999 # Max number of rows x number of columns to trigger (stratified) sampling for leakage checks #leakage_max_data_size = 10000000 # Whether to create the Python scoring pipeline at the end of each experiment. #make_python_scoring_pipeline = "auto" # Whether to create the MOJO scoring pipeline at the end of each experiment. If set to "auto", will attempt to # create it if possible (without dropping capabilities). If set to "on", might need to drop some models, # transformers or custom recipes. #make_mojo_scoring_pipeline = "auto" # Whether to measure the MOJO scoring latency at the time of MOJO creation. #benchmark_mojo_latency = "auto" # Max size of pipeline.mojo file (in MB) for automatic mode of MOJO scoring latency measurement #benchmark_mojo_latency_auto_size_limit = 100 # If MOJO creation times out at end of experiment, can still make MOJO from the GUI or from the R/Py clients (timeout doesn't apply there). #mojo_building_timeout = 1800.0 # If MOJO creation is too slow, increase this value. Higher values can finish faster, but use more memory. # If MOJO creation fails due to an out-of-memory error, reduce this value to 1. # Set to -1 for all physical cores. # #mojo_building_parallelism = -1 # Whether to create the pipeline visualization at the end of each experiment. #make_pipeline_visualization = "auto" # Whether to create the experiment Autoreport after end of experiment. # #make_autoreport = true # Max number of CPU cores to use per experiment. Set to <= 0 to use all cores. # One can also set environment variable 'OMP_NUM_THREADS' to number of cores to use for OpenMP # (e.g., in bash: 'export OMP_NUM_THREADS=32' and 'export OPENBLAS_NUM_THREADS=32'). #max_cores = 0 # Max number of CPU cores to use across all of DAI experiments and tasks. # -1 is all available, with stall_subprocess_submission_dai_fork_threshold_count=0 means restricted to core count. # #max_cores_dai = -1 # Stall submission of tasks if total DAI fork count exceeds count (-1 to disable, 0 for automatic of max_cores_dai) #stall_subprocess_submission_dai_fork_threshold_count = 0 # Stall submission of tasks if system memory available is less than this threshold in percent (set to 0 to disable). # Above this threshold, the number of workers in any pool of workers is linearly reduced down to 1 once hitting this threshold. # #stall_subprocess_submission_mem_threshold_pct = 2 # Whether to set automatic number of cores by physical (True) or logical (False) count. # Using all logical cores can lead to poor performance due to cache thrashing. #max_cores_by_physical = true # Absolute limit to core count #max_cores_limit = 100 # Control maximum number of cores to use for a model's fit call (0 = all physical cores >= 1 that count) #max_fit_cores = 10 # Control maximum number of cores to use for a model's predict call (0 = all physical cores >= 1 that count) #max_predict_cores = 0 # Control maximum number of cores to use for a model's transform and predict call when doing operations inside DAI-MLI GUI and R/Py client (0 = all physical cores >= 1 that count) #max_predict_cores_in_dai = 4 # Control number of workers used in CPU mode for tuning (0 = socket count -1 = all physical cores >= 1 that count). More workers will be more parallel but models learn less from each other. #batch_cpu_tuning_max_workers = 0 # Control number of workers used in CPU mode for training (0 = socket count -1 = all physical cores >= 1 that count) #cpu_max_workers = 0 # Number of GPUs to use per experiment for training task. Set to -1 for all GPUs. # An experiment will generate many different models. # Currently num_gpus_per_experiment!=-1 disables GPU locking, so is only recommended for # single experiments and single users. # Ignored if GPUs disabled or no GPUs on system. # More info at: https://github.com/NVIDIA/nvidia-docker/wiki/nvidia-docker#gpu-isolation #num_gpus_per_experiment = -1 # Number of CPU cores per GPU. Limits number of GPUs in order to have sufficient cores per GPU. # Set to -1 to disable. #min_num_cores_per_gpu = 2 # Number of GPUs to use per model training task. Set to -1 for all GPUs. # For example, when this is set to -1 and there are 4 GPUs available, all of them can be used for the training of a single model. # Currently num_gpus_per_model!=1 disables GPU locking, so is only recommended for single # experiments and single users. # Ignored if GPUs disabled or no GPUs on system. # More info at: https://github.com/NVIDIA/nvidia-docker/wiki/nvidia-docker#gpu-isolation #num_gpus_per_model = 1 # Number of GPUs to use for predict for models and transform for transformers when running outside of fit/fit_transform. # If predict/transform are called in same process as fit/fit_transform, number of GPUs will match, # while new processes will use this count for number of GPUs for applicable models/transformers. # If tensorflow_nlp_have_gpus_in_production=true, then that overrides this setting for relevant # TensorFflow NLP transformers. # #num_gpus_for_prediction = 0 # Minimum number of threads for datatable (and OpenMP) during data munging (per process). # datatable is the main data munging tool used within Driverless ai (source : # https://github.com/h2oai/datatable) #min_dt_threads_munging = 1 # Like min_datatable (and OpenMP)_threads_munging but for final pipeline munging #min_dt_threads_final_munging = 1 # Maximum number of threads for datatable during data munging (per process) (0 = all, -1 = auto). #max_dt_threads_munging = -1 # Maximum number of threads for datatable during data reading and writing (per process) (0 = all, -1 = auto). #max_dt_threads_readwrite = -1 # Maximum number of threads for datatable stats and openblas (per process) (0 = all, -1 = auto). #max_dt_threads_stats_openblas = -1 # Maximum number of threads for datatable during TS properties preview panel computations). #max_dt_threads_do_timeseries_split_suggestion = 1 # Which gpu_id to start with # If using CUDA_VISIBLE_DEVICES=... to control GPUs (preferred method), gpu_id=0 is the # first in that restricted list of devices. # E.g. if CUDA_VISIBLE_DEVICES='4,5' then gpu_id_start=0 will refer to the # device #4. # E.g. from expert mode, to run 2 experiments, each on a distinct GPU out of 2 GPUs: # Experiment#1: num_gpus_per_model=1, num_gpus_per_experiment=1, gpu_id_start=0 # Experiment#2: num_gpus_per_model=1, num_gpus_per_experiment=1, gpu_id_start=1 # E.g. from expert mode, to run 2 experiments, each on a distinct GPU out of 8 GPUs: # Experiment#1: num_gpus_per_model=1, num_gpus_per_experiment=4, gpu_id_start=0 # Experiment#2: num_gpus_per_model=1, num_gpus_per_experiment=4, gpu_id_start=4 # E.g. Like just above, but now run on all 4 GPUs/model # Experiment#1: num_gpus_per_model=4, num_gpus_per_experiment=4, gpu_id_start=0 # Experiment#2: num_gpus_per_model=4, num_gpus_per_experiment=4, gpu_id_start=4 # If num_gpus_per_model!=1, global GPU locking is disabled # (because underlying algorithms don't support arbitrary gpu ids, only sequential ids), # so must setup above correctly to avoid overlap across all experiments by all users # More info at: https://github.com/NVIDIA/nvidia-docker/wiki/nvidia-docker#gpu-isolation # Note that gpu selection does not wrap, so gpu_id_start + num_gpus_per_model must be less than number of visibile gpus #gpu_id_start = 0 # Maximum number of workers for Driverless AI server pool (only 1 needed currently) #max_workers = 1 # Period (in seconds) of ping by Driverless AI server to each experiment # (in order to get logger info like disk space and memory usage). # 0 means don't print anything. #ping_period = 60 # Period between checking DAI status. #ping_sleep_period = 1 # Minimum amount of disk space in GB needed to run experiments. # Experiments will fail if this limit is crossed. # This limit exists because Driverless AI needs to generate data for model training # feature engineering, documentation and other such processes. #disk_limit_gb = 5 # Minimum amount of disk space in GB needed to before stall forking of new processes during an experiment. #stall_disk_limit_gb = 1 # Minimum amount of system memory in GB needed to start experiments. # Similarly with disk space, a certain amount of system memory is needed to run some basic # operations. #memory_limit_gb = 5 # Minimum number of rows needed to run experiments (values lower than 100 might not work). # A minimum threshold is set to ensure there is enough data to create a statistically # reliable model and avoid other small-data related failures. #min_num_rows = 100 # Minimum required number of rows (in the training data) for each class label for classification problems. #min_rows_per_class = 5 # Minimum required number of rows for each split when generating validation samples. #min_rows_per_split = 5 # Precision of how data is stored # 'float32' best for speed, 'float64' best for accuracy or very large input values # 'float32' allows numbers up to about +-3E38 with relative error of about 1E-7 # 'float64' allows numbers up to about +-1E308 with relative error of about 1E-16 # Some calculations, like the GLM standardization, can only handle up to sqrt() of these maximums for data values, # So GLM with 32-bit precision can only handle up to about a value of 1E19 before standardization generates inf values. # If you see "Best individual has invalid score" you may require higher precision. #data_precision = "float32" # Precision of most data transformers (same options and notes as data_precision). # Useful for higher precision in transformers with numerous operations that can accumulate error. # Also useful if want faster performance for transformers but otherwise want data stored in high precision. #transformer_precision = "float32" # Whether to change ulimit soft limits up to hard limits (for DAI server app, which is not a generic user app). # Prevents resource limit problems in some cases. # Restricted to no more than limit_nofile and limit_nproc for those resources. #ulimit_up_to_hard_limit = true #disable_core_files = false # number of file limit # Below should be consistent with start-dai.sh #limit_nofile = 65535 # number of threads limit # Below should be consistent with start-dai.sh #limit_nproc = 16384 # Level of reproducibility desired (for same data and same inputs). # Only active if 'reproducible' mode is enabled (GUI button enabled or a seed is set from the client API). # Supported levels are: # reproducibility_level = 1 for same experiment results as long as same O/S, same CPU(s) and same GPU(s) # reproducibility_level = 2 for same experiment results as long as same O/S, same CPU architecture and same GPU architecture # reproducibility_level = 3 for same experiment results as long as same O/S, same CPU architecture, not using GPUs # reproducibility_level = 4 for same experiment results as long as same O/S, (best effort) #reproducibility_level = 1 # Seed for random number generator to make experiments reproducible, to a certain reproducibility level (see above). # Only active if 'reproducible' mode is enabled (GUI button enabled or a seed is set from the client API). #seed = 1234 # The list of values that should be interpreted as missing values during data import. # This applies to both numeric and string columns. Note that the dataset must be reloaded after applying changes to this config via the expert settings. # Also note that 'nan' is always interpreted as a missing value for numeric columns. #missing_values = "['', '?', 'None', 'nan', 'NA', 'N/A', 'unknown', 'inf', '-inf', '1.7976931348623157e+308', '-1.7976931348623157e+308']" # For tensorflow, what numerical value to give to missing values, where numeric values are standardized. # So 0 is center of distribution, and if Normal distribution then +-5 is 5 standard deviations away from the center. # In many cases, an out of bounds value is a good way to represent missings, but in some cases the mean (0) may be better. #tf_nan_impute_value = -5 # Internal threshold for number of rows x number of columns to trigger certain statistical # techniques (small data recipe like including one hot encoding for all model types, and smaller learning rate) # to increase model accuracy #statistical_threshold_data_size_small = 100000 # Internal threshold for number of rows x number of columns to trigger certain statistical # techniques (fewer genes created, removal of high max_depth for tree models, etc.) that can speed up modeling. # Also controls maximum rows used in training final model, # by sampling statistical_threshold_data_size_large / columns number of rows #statistical_threshold_data_size_large = 500000000 # Internal threshold for number of rows x number of columns to trigger sampling for auxiliary data uses, # like imbalanced data set detection and bootstrap scoring sample size and iterations #aux_threshold_data_size_large = 10000000 # Internal threshold for number of rows x number of columns to trigger certain changes in performance # (fewer threads if beyond large value) to help avoid OOM or unnecessary slowdowns # (fewer threads if lower than small value) to avoid excess forking of tasks #performance_threshold_data_size_small = 100000 # Internal threshold for number of rows x number of columns to trigger certain changes in performance # (fewer threads if beyond large value) to help avoid OOM or unnecessary slowdowns # (fewer threads if lower than small value) to avoid excess forking of tasks #performance_threshold_data_size_large = 100000000 # Maximum number of columns to start an experiment. This threshold exists to constraint the # complexity and the length of the Driverless AI's processes. #max_cols = 10000 # Largest number of rows to use for column stats, otherwise sample randomly #max_rows_col_stats = 1000000 # Whether to obtain permutation feature importance on original features for reporting in logs and file. # #orig_features_fs_report = false # Maximum number of rows when doing permutation feature importance, reduced by (stratified) random sampling. # #max_rows_fs = 1000000 # How many workers to use for feature selection by permutation for predict phase # (0 = auto, > 0: min of DAI value and this value, < 0: exactly negative of this value) # #max_workers_fs = 0 # Maximum number of columns selected out of original set of original columns, using feature selection # The selection is based upon how well target encoding (or frequency encoding if not available) on categoricals and numerics treated as categoricals # This is useful to reduce the final model complexity. First the best # [max_orig_cols_selected] are found through feature selection methods and then # these features are used in feature evolution (to derive other features) and in modelling. #max_orig_cols_selected = 10000 # Maximum number of numeric columns selected, above which will do feature selection # same as above (max_orig_cols_selected) but for numeric columns. #max_orig_numeric_cols_selected = 10000 # Maximum number of non-numeric columns selected, above which will do feature selection on all features and avoid treating numerical as categorical # same as above (max_orig_numeric_cols_selected) but for categorical columns. #max_orig_nonnumeric_cols_selected = 300 # The factor times max_orig_cols_selected, by which column selection is based upon no target encoding and no treating numerical as categorical # in order to limit performance cost of feature engineering #max_orig_cols_selected_simple_factor = 2 # Like max_orig_cols_selected, but columns above which add special individual with original columns reduced. # #fs_orig_cols_selected = 500 # Like max_orig_numeric_cols_selected, but applicable to special individual with original columns reduced. # A separate individual in the genetic algorithm is created by doing feature selection by permutation importance on original features. # #fs_orig_numeric_cols_selected = 500 # Like max_orig_nonnumeric_cols_selected, but applicable to special individual with original columns reduced. # A separate individual in the genetic algorithm is created by doing feature selection by permutation importance on original features. # #fs_orig_nonnumeric_cols_selected = 200 # Like max_orig_cols_selected_simple_factor, but applicable to special individual with original columns reduced. #fs_orig_cols_selected_simple_factor = 2 # Maximum allowed fraction of unique values for integer and categorical columns (otherwise will treat column as ID and drop) #max_relative_cardinality = 0.95 # Maximum allowed number of unique values for integer and categorical columns (otherwise will treat column as ID and drop) #max_absolute_cardinality = 1000000 # Whether to treat some numerical features as categorical. # For instance, sometimes an integer column may not represent a numerical feature but # represent different numerical codes instead. #num_as_cat = true # Max number of unique values for integer/real columns to be treated as categoricals (test applies to first statistical_threshold_data_size_small rows only) #max_int_as_cat_uniques = 50 # Number of folds for models used during the feature engineering process. # Increasing this will put a lower fraction of data into validation and more into training # (e.g., num_folds=3 means 67%/33% training/validation splits). # Actual value will vary for small or big data cases. #num_folds = 3 # For multiclass problems only. Whether to allow different sets of target classes across (cross-)validation # fold splits. Especially important when passing a fold column that isn't balanced w.r.t class distribution. # #allow_different_classes_across_fold_splits = true # Accuracy setting equal and above which enables full cross-validation (multiple folds) during feature evolution # as opposed to only a single holdout split (e.g. 2/3 train and 1/3 validation holdout) #full_cv_accuracy_switch = 8 # Accuracy setting equal and above which enables stacked ensemble as final model. # Stacking commences at the end of the feature evolution process.. # It quite often leads to better model performance, but it does increase the complexity # and execution time of the final model. #ensemble_accuracy_switch = 5 # Number of fold splits to use for ensemble_level >= 2. # The ensemble modelling may require predictions to be made on out-of-fold samples # hence the data needs to be split on different folds to generate these predictions. # Less folds (like 2 or 3) normally create more stable models, but may be less accurate # More folds can get to higher accuracy at the expense of more time, but the performance # may be less stable when the training data is not enough (i.e. higher chance of overfitting). # Actual value will vary for small or big data cases. #num_ensemble_folds = 5 # Number of repeats for each fold for all validation # (modified slightly for small or big data cases) #fold_reps = 1 #max_num_classes_hard_limit = 10000 # Maximum number of classes to allow for a classification problem. # High number of classes may make certain processes of Driverless AI time-consuming. # Memory requirements also increase with higher number of classes #max_num_classes = 200 # Maximum number of classes to compute ROC and CM for, # beyond which roc_reduce_type choice for reduction is applied. # Too many classes can take much longer than model building time. #max_num_classes_compute_roc = 200 # Maximum number of classes to show in GUI for confusion matrix, showing first max_num_classes_client_and_gui labels. # The number 20 is default, but beyond 6 classes the diagnostics launched from GUI are visually truncated. # This will only modify client-GUI launched diagnostics if changed in config.toml and server is restarted, # while this value can be changed in expert settings to control experiment plots. #max_num_classes_client_and_gui = 10 # If too many classes when computing roc, # reduce by "rows" by randomly sampling rows, # or reduce by truncating classes to no more than max_num_classes_compute_roc. # If have sufficient rows for class count, can reduce by rows. #roc_reduce_type = "rows" #min_roc_sample_size = 1 # Number of actuals vs. predicted data points to use in order to generate in the relevant # plot/graph which is shown at the right part of the screen within an experiment. #num_actuals_vs_predicted = 100 # Whether to use H2O.ai brain: the local caching and smart re-use of prior experiments, # in order to generate more useful features and models for new experiments. # It can also be used to control checkpointing for experiments that have been paused or interrupted. # DAI will use H2O.ai brain cache if cache file has # a) any matching column names and types for a similar experiment type # b) exactly matches classes # c) exactly matches class labels # d) matches basic time series choices # e) interpretability of cache is equal or lower # f) main model (booster) is allowed by new experiment. # Level of brain to use (for chosen level, where higher levels will also do all lower level operations automatically) # -1 = Don't use any brain cache and don't write any cache # 0 = Don't use any brain cache but still write cache # Use case: Want to save model for later use, but want current model to be built without any brain models # 1 = smart checkpoint from latest best individual model # Use case: Want to use latest matching model, but match can be loose, so needs caution # 2 = smart checkpoint from H2O.ai brain cache of individual best models # Use case: DAI scans through H2O.ai brain cache for best models to restart from # 3 = smart checkpoint like level #1, but for entire population. Tune only if brain population insufficient size # (will re-score entire population in single iteration, so appears to take longer to complete first iteration) # 4 = smart checkpoint like level #2, but for entire population. Tune only if brain population insufficient size # (will re-score entire population in single iteration, so appears to take longer to complete first iteration) # 5 = like #4, but will scan over entire brain cache of populations to get best scored individuals # (can be slower due to brain cache scanning if big cache) # 1000 + feature_brain_level (above positive values) = use resumed_experiment_id and actual feature_brain_level, # to use other specific experiment as base for individuals or population, # instead of sampling from any old experiments # GUI has 3 options and corresponding settings: # 1) New Experiment: Uses feature brain level default of 2 # 2) New Model With Same Parameters: Re-uses the same feature brain level as parent experiment # 3) Restart From Last Checkpoint: Resets feature brain level to 1003 and sets experiment ID to resume from # (continued genetic algorithm iterations) # 4) Retrain Final Pipeline: Like Restart but also time=0 so skips any tuning and heads straight to final model # (assumes had at least one tuning iteration in parent experiment) # Other use cases: # a) Restart on different data: Use same column names and fewer or more rows (applicable to 1 - 5) # b) Re-fit only final pipeline: Like (a), but choose time=1 and feature_brain_level=3 - 5 # c) Restart with more columns: Add columns, so model builds upon old model built from old column names (1 - 5) # d) Restart with focus on model tuning: Restart, then select feature_engineering_effort = 3 in expert settings # e) can retrain final model but ignore any original features except those in final pipeline (normal retrain but set brain_add_features_for_new_columns=false) # Notes: # 1) In all cases, we first check the resumed experiment id if given, and then the brain cache # 2) For Restart cases, may want to set min_dai_iterations to non-zero to force delayed early stopping, else may not be enough iterations to find better model. # 3) A "New model with Same Params" of a Restart will use feature_brain_level=1003 for default Restart mode (revert to 2, or even 0 if want to start a fresh experiment otherwise) #feature_brain_level = 2 # Relative number of columns that must match between current reference individual and brain individual. # 0.0: perfect match # 1.0: All columns are different, worst match # e.g. 0.1 implies no more than 10% of columns mismatch between reference set of columns and brain individual. # #brain_maximum_diff_score = 0.1 # Maximum number of brain individuals pulled from H2O.ai brain cache for feature_brain_level=1, 2 #max_num_brain_indivs = 3 # Save feature brain iterations every iter_num % feature_brain_iterations_save_every_iteration == 0, to be able to restart/refit with which_iteration_brain >= 0 # 0 means disable #feature_brain_save_every_iteration = 0 # When doing restart or re-fit type feature_brain_level with resumed_experiment_id, choose which iteration to start from, instead of only last best # -1 means just use last best # Usage: # 1) Run one experiment with feature_brain_iterations_save_every_iteration=1 or some other number # 2) Identify which iteration brain dump one wants to restart/refit from # 3) Restart/Refit from original experiment, setting which_iteration_brain to that number in expert settings # Note: If restart from a tuning iteration, this will pull in entire scored tuning population and use that for feature evolution #which_iteration_brain = -1 # When doing re-fit, if change columns or features, population of individuals used to refit from may change order of which was best, # leading to better result chosen (False case). But sometimes want to see exact same model/features with only one feature added, # and then would need to set this to True case. # E.g. if refit with just 1 extra column and have interpretability=1, then final model will be same features, # with one more engineered feature applied to that new original feature. # #refit_same_best_individual = false # Directory, relative to data_directory, to store H2O.ai brain meta model files #brain_rel_dir = "H2O.ai_brain" # Maximum size in bytes the brain will store # We reserve this memory to save data in order to ensure we can retrieve an experiment if # for any reason it gets interrupted. # -1: unlimited # >=0 number of GB to limit brain to #brain_max_size_GB = 20 # Whether to take any new columns and add additional features to pipeline, even if doing retrain final model. # In some cases, one might have a new dataset but only want to keep same pipeline regardless of new columns, # in which case one sets this to False. For example, new data might lead to new dropped features, # due to shift or leak detection. To avoid change of feature set, one can disable all dropping of columns, # but set this to False to avoid adding any columns as new features, # so pipeline is perfectly preserved when changing data. #brain_add_features_for_new_columns = true # Whether to enable early stopping # Early stopping refers to stopping the feature evolution/engineering process # when there is no performance uplift after a certain number of iterations. # After early stopping has been triggered, Driverless AI will initiate the ensemble # process if selected. #early_stopping = true # Whether to enable early stopping per individual # Each individual in the generic algorithm will stop early if no improvement, # and it will no longer be mutated. # Instead, the best individual will be additionally mutated. #early_stopping_per_individual = true # Minimum number of Driverless AI iterations to stop the feature evolution/engineering # process even if score is not improving. Driverless AI needs to run for at least that many # iterations before deciding to stop. It can be seen a safeguard against suboptimal (early) # convergence. #min_dai_iterations = 0 # Maximum features per model (and each model within the final model if ensemble) kept just after scoring them # Keeps top variable importance features, prunes rest away, after each scoring. # Final ensemble will exclude any pruned-away features and only train on kept features, # but may contain a few new features due to fitting on different data view (e.g. new clusters) # Final scoring pipeline will exclude any pruned-away features, # but may contain a few new features due to fitting on different data view (e.g. new clusters) # -1 means no restrictions except internally-determined memory and interpretability restrictions #nfeatures_max = -1 # Maximum genes (transformer instances) per model (and each model within the final model if ensemble) kept. # Controls number of genes before features are scored, so just randomly samples genes if pruning occurs. # If restriction occurs after scoring features, then aggregated gene importances are used for pruning genes. # Instances includes all possible transformers, including original transformer for numeric features. # -1 means no restrictions except internally-determined memory and interpretability restrictions #ngenes_max = -1 # Whether to limit feature counts by interpretability setting via features_allowed_by_interpretability #limit_features_by_interpretability = true # Max. number of epochs for TensorFlow models for making NLP features #tensorflow_max_epochs_nlp = 2 # Accuracy setting equal and above which will add all enabled TensorFlow NLP models below at start of experiment for text dominated problems # when TensorFlow NLP transformers are set to auto. If set to on, this parameter is ignored. # Otherwise, at lower accuracy, TensorFlow NLP transformations will only be created as a mutation. #enable_tensorflow_nlp_accuracy_switch = 5 # Whether to use Word-based CNN TensorFlow models for NLP if TensorFlow enabled #enable_tensorflow_textcnn = "auto" # Whether to use Word-based Bi-GRU TensorFlow models for NLP if TensorFlow enabled #enable_tensorflow_textbigru = "auto" # Whether to use Character-level CNN TensorFlow models for NLP if TensorFlow enabled #enable_tensorflow_charcnn = "auto" # Path to pretrained embeddings for TensorFlow NLP models # For example, download and unzip https://nlp.stanford.edu/data/glove.6B.zip # tensorflow_nlp_pretrained_embeddings_file_path = /path/on/server/to/glove.6B.300d.txt #tensorflow_nlp_pretrained_embeddings_file_path = "" # Allow training of all weights of the neural network graph, including the pretrained embedding layer weights. If disabled, then the embedding layer is frozen, but all other weights are still fine-tuned. #tensorflow_nlp_pretrained_embeddings_trainable = false # Whether Python/MOJO scoring runtime will have GPUs (otherwise BiGRU will fail in production if this is enabled). # Enabling this can speed up training for BiGRU, but will require GPUs and CuDNN in production. #tensorflow_nlp_have_gpus_in_production = false # Fraction of text columns out of all features to be considered a text-dominated problem #text_fraction_for_text_dominated_problem = 0.3 # Fraction of text transformers to all transformers above which to trigger that text dominated problem #text_transformer_fraction_for_text_dominated_problem = 0.3 # Threshold for average string-is-text score as determined by internal heuristics # It decides when a string column will be treated as text (for an NLP problem) or just as # a standard categorical variable. # Higher values will favor string columns as categoricals, lower values will favor string columns as text #string_col_as_text_threshold = 0.3 # Mininum fraction of unique values for string columns to be considered as possible text (otherwise categorical) #string_col_as_text_min_relative_cardinality = 0.1 # Mininum number of uniques for string columns to be considered as possible text (otherwise categorical) #string_col_as_text_min_absolute_cardinality = 100 # Interpretability setting equal and above which will use monotonicity constraints in GBM # You may read the following source to understand what these constraints connote and why # they may be important, especially when the end goal is a very interpretable machine # learning model: https://blog.datadive.net/monotonicity-constraints-in-machine-learning/ #monotonicity_constraints_interpretability_switch = 7 # Threshold, of Pearson product-moment correlation coefficient between numerical or encoded transformed # feature and target, above (below negative for) which will enforce positive (negative) monotonicity # for XGBoost and LightGBM. # Enabled when interpretability >= monotonicity_constraints_interpretability_switch config toml value #monotonicity_constraints_correlation_threshold = 0.1 # Exploring feature interactions can be important in gaining better predictive performance. # The interaction can take multiple forms (i.e. feature1 + feature2 or feature1 * feature2 + ... featureN) # Although certain machine learning algorithms (like tree-based methods) can do well in # capturing these interactions as part of their training process, still generating them may # help them (or other algorithms) yield better performance. # The depth of the interaction level (as in "up to" how many features may be combined at # once to create one single feature) can be specified to control the complexity of the # feature engineering process. For transformers that use both numeric and categorical features, this constrains # the number of each type, not the total number. Higher values might be able to make more predictive models # at the expense of time (-1 means automatic). #max_feature_interaction_depth = -1 # Instead of sampling from min to max (up to max_feature_interaction_depth unless all specified) # columns allowed for each transformer (0), choose fixed non-zero number of columns to use. # Can make same as number of columns to use all columns for each transformers if allowed by each transformer. # -n can be chosen to do 50/50 sample and fixed of n features. # #fixed_feature_interaction_depth = 0 # Like fixed_feature_interaction_depth but for categoricals if doing numcat (i.e. numeric and categoricals separated) # transformers. Only applies if also fixed_feature_interaction_depth > 0, and then will use sampling (0) unless # override since doesn't make sense to group by all columns usually. #fixed_feature_interaction_depth_numcat_cat = 0 # Accuracy setting equal and above which enables tuning of model parameters # Only applicable if parameter_tuning_num_models=-1 (auto) #tune_parameters_accuracy_switch = 3 # Accuracy setting equal and above which enables tuning of target transform for regression. # This is useful for time series when instead of predicting the actual target value, it # might be better to predict a transformed target variable like sqrt(target) or log(target) # as a means to control for outliers. #tune_target_transform_accuracy_switch = 3 # Select a target transformation for regression problems. Must be one of: ['auto', # 'identity', 'unit_box', 'log', 'square', 'sqrt', 'double_sqrt', 'inverse', 'anscombe', 'logit', 'sigmoid']. # If set to 'auto', will automatically pick the best target transformer (if accuracy is set to # tune_target_transform_accuracy_switch or larger). # #target_transformer = "auto" # Tournament style (method to decide which models are best at each iteration) # 'auto' : Choose based upon accuracy, etc. # 'fullstack' : Choose among optimal model and feature types # 'model' : individuals with same model type compete # 'feature' : individuals with similar feature types compete # 'uniform' : all individuals in population compete to win as best # 'model' and 'feature' styles preserve at least one winner for each type (and so 2 total indivs of each type after mutation) # For each case, a round robin approach is used to choose best scores among type of models to choose from #tournament_style = "auto" # Interpretability above which will use 'uniform' tournament style #tournament_uniform_style_interpretability_switch = 6 # Accuracy below which will use uniform style if tournament_style = 'auto' (regardless of other accuracy tournament style switch values) #tournament_uniform_style_accuracy_switch = 6 # Accuracy equal and above which uses model style if tournament_style = 'auto' #tournament_model_style_accuracy_switch = 6 # Accuracy equal and above which uses feature style if tournament_style = 'auto' #tournament_feature_style_accuracy_switch = 7 # Accuracy equal and above which uses fullstack style if tournament_style = 'auto' #tournament_fullstack_style_accuracy_switch = 8 # Whether to use penalized score for GA tournament or actual score #tournament_use_feature_penalized_score = true # Driverless AI uses a genetic algorithm (GA) to find the best features, best models and # best hyper parameters for these models. The GA facilitates getting good results while not # requiring torun/try every possible model/feature/parameter. This version of GA has # reinforcement learning elements - it uses a form of exploration-exploitation to reach # optimum solutions. This means it will capitalise on models/features/parameters that seem # to be working well and continue to exploit them even more, while allowing some room for # trying new (and semi-random) models/features/parameters to avoid settling on a local # minimum. # These models/features/parameters tried are what-we-call individuals of a population. More # individuals connote more models/features/parameters to be tried and compete to find the best # ones. #num_individuals = 2 # set fixed number of individuals (if > 0) - useful to compare different hardware configurations #fixed_num_individuals = 0 # set fixed number of fold reps (if > 0) - useful for quick runs regardless of data #fixed_fold_reps = 0 # number of unique targets or folds counts after which switch to faster/simpler non-natural sorting and print outs #sanitize_natural_sort_limit = 1000 # Whether target encoding could be enabled # Target encoding refers to several different feature transformations (primarily focused on # categorical data) that aim to represent the feature using information of the actual # target variable. A simple example can be to use the mean of the target to replace each # unique category of a categorical feature. This type of features can be very predictive, # but are prone to overfitting and require more memory as they need to store mappings of # the unique categories and the target values. #enable_target_encoding = "auto" #enable_lexilabel_encoding = "off" #enable_isolation_forest = "off" # Whether one hot encoding could be enabled. If auto, then only applied for small data and GLM. #enable_one_hot_encoding = "auto" #isolation_forest_nestimators = 200 # Driverless AI categorises all data (feature engineering) transformers # More information for these transformers can be viewed here: # http://docs.h2o.ai/driverless-ai/latest-stable/docs/userguide/transformations.html # This section allows including/excluding these transformations and may be useful when # simpler (more interpretable) models are sought at the expense of accuracy. # the interpretability setting) # for multi-class: '['NumCatTETransformer', 'TextLinModelTransformer', # 'FrequentTransformer', 'CVTargetEncodeTransformer', 'ClusterDistTransformer', # 'WeightOfEvidenceTransformer', 'TruncSVDNumTransformer', 'CVCatNumEncodeTransformer', # 'DatesTransformer', 'TextTransformer', 'OriginalTransformer', # 'NumToCatWoETransformer', 'NumToCatTETransformer', 'ClusterTETransformer', # 'InteractionsTransformer']' # for regression/binary: '['TextTransformer', 'ClusterDistTransformer', # 'OriginalTransformer', 'TextLinModelTransformer', 'NumToCatTETransformer', # 'DatesTransformer', 'WeightOfEvidenceTransformer', 'InteractionsTransformer', # 'FrequentTransformer', 'CVTargetEncodeTransformer', 'NumCatTETransformer', # 'NumToCatWoETransformer', 'TruncSVDNumTransformer', 'ClusterTETransformer', # 'CVCatNumEncodeTransformer']' # This list appears in the experiment logs (search for 'Transformers used') # #included_transformers = "[]" # Auxiliary to included_transformers # e.g. to disable all Target Encoding: excluded_transformers = # '['NumCatTETransformer', 'CVTargetEncodeF', 'NumToCatTETransformer', # 'ClusterTETransformer']' # #excluded_transformers = "[]" # Include list of genes (i.e. genes (built on top of transformers) to use, # independent of the interpretability setting) # Some transformers are used by multiple genes, so this allows different control over feature engineering # for multi-class: '['InteractionsGene', 'WeightOfEvidenceGene', # 'NumToCatTargetEncodeSingleGene', 'OriginalGene', 'TextGene', 'FrequentGene', # 'NumToCatWeightOfEvidenceGene', 'NumToCatWeightOfEvidenceMonotonicGene', ' # CvTargetEncodeSingleGene', 'DateGene', 'NumToCatTargetEncodeMultiGene', ' # DateTimeGene', 'TextLinRegressorGene', 'ClusterIDTargetEncodeSingleGene', # 'CvCatNumEncodeGene', 'TruncSvdNumGene', 'ClusterIDTargetEncodeMultiGene', # 'NumCatTargetEncodeMultiGene', 'CvTargetEncodeMultiGene', 'TextLinClassifierGene', # 'NumCatTargetEncodeSingleGene', 'ClusterDistGene']' # for regression/binary: '['CvTargetEncodeSingleGene', 'NumToCatTargetEncodeSingleGene', # 'CvCatNumEncodeGene', 'ClusterIDTargetEncodeSingleGene', 'TextLinRegressorGene', # 'CvTargetEncodeMultiGene', 'ClusterDistGene', 'OriginalGene', 'DateGene', # 'ClusterIDTargetEncodeMultiGene', 'NumToCatTargetEncodeMultiGene', # 'NumCatTargetEncodeMultiGene', 'TextLinClassifierGene', 'WeightOfEvidenceGene', # 'FrequentGene', 'TruncSvdNumGene', 'InteractionsGene', 'TextGene', # 'DateTimeGene', 'NumToCatWeightOfEvidenceGene', # 'NumToCatWeightOfEvidenceMonotonicGene', ''NumCatTargetEncodeSingleGene']' # This list appears in the experiment logs (search for 'Genes used') # e.g. to only enable interaction gene, use: included_genes = # '['InteractionsGene']' #included_genes = "[]" # Exclude list of genes (i.e. genes (built on top of transformers) to not use, # independent of the interpretability setting) # Some transformers are used by multiple genes, so this allows different control over feature engineering 
 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216 2217 2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254 2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 2331 2332 2333 2334 2335 2336 2337 2338 2339 2340 2341 2342 2343 2344 2345 2346 2347 2348 2349 2350 2351 2352 2353 2354 2355 2356 2357 2358 2359 2360 2361 2362 2363 2364 2365 2366 2367 2368 2369 2370 2371 2372 2373 2374 2375 2376 2377 2378 2379 2380 2381 2382 2383 2384 2385 2386 2387 2388 2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2449 2450 2451 2452 2453 2454 2455 2456 2457 2458 2459 2460 2461 2462 2463 2464 2465 2466 2467 2468 2469 2470 2471 2472 2473 2474 2475 2476 2477 2478 2479 2480 2481 2482 2483 2484 2485 2486 2487 2488 2489 2490 2491 2492 2493 2494 2495 2496 2497 2498 2499 2500 2501 2502 2503 2504 2505 2506 2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541 2542 2543 2544 2545 2546 2547 2548 2549 2550 2551 2552 2553 2554 2555 2556 2557 2558 2559 2560 2561 2562 2563 2564 2565 2566 2567 2568 2569 2570 2571 2572 2573 2574 2575 2576 2577 2578 2579 2580 2581 2582 2583 2584 2585 2586 2587 2588 2589 2590 2591 2592 2593 2594 2595 2596 2597 2598 2599 2600 2601 2602 2603 2604 2605 2606 2607 2608 2609 2610 2611 2612 2613 2614 2615 2616 2617 2618 2619 2620 2621 2622 2623 2624 2625 2626 2627 2628 2629 2630 2631 2632 2633 2634 2635 2636 2637 2638 2639 2640 2641 2642 2643 2644 2645 2646 2647 2648 2649 2650 2651 2652 2653 2654 2655 2656 2657 2658 2659 2660 2661 2662 2663 2664 2665 2666 2667 2668 2669 2670 2671 2672 2673 2674 2675 2676 2677 2678 2679 2680 2681 2682 2683 2684 2685 2686 2687 2688 2689 2690 2691 2692 2693 2694 2695 2696 2697 2698 2699 2700 2701 # for multi-class: '['InteractionsGene', 'WeightOfEvidenceGene', # 'NumToCatTargetEncodeSingleGene', 'OriginalGene', 'TextGene', 'FrequentGene', # 'NumToCatWeightOfEvidenceGene', 'NumToCatWeightOfEvidenceMonotonicGene', ' # CvTargetEncodeSingleGene', 'DateGene', 'NumToCatTargetEncodeMultiGene', ' # DateTimeGene', 'TextLinRegressorGene', 'ClusterIDTargetEncodeSingleGene', # 'CvCatNumEncodeGene', 'TruncSvdNumGene', 'ClusterIDTargetEncodeMultiGene', # 'NumCatTargetEncodeMultiGene', 'CvTargetEncodeMultiGene', 'TextLinClassifierGene', # 'NumCatTargetEncodeSingleGene', 'ClusterDistGene']' # for regression/binary: '['CvTargetEncodeSingleGene', 'NumToCatTargetEncodeSingleGene', # 'CvCatNumEncodeGene', 'ClusterIDTargetEncodeSingleGene', 'TextLinRegressorGene', # 'CvTargetEncodeMultiGene', 'ClusterDistGene', 'OriginalGene', 'DateGene', # 'ClusterIDTargetEncodeMultiGene', 'NumToCatTargetEncodeMultiGene', # 'NumCatTargetEncodeMultiGene', 'TextLinClassifierGene', 'WeightOfEvidenceGene', # 'FrequentGene', 'TruncSvdNumGene', 'InteractionsGene', 'TextGene', # 'DateTimeGene', 'NumToCatWeightOfEvidenceGene', # 'NumToCatWeightOfEvidenceMonotonicGene', ''NumCatTargetEncodeSingleGene']' # This list appears in the experiment logs (search for 'Genes used') # e.g. to disable interaction gene, use: excluded_genes = # '['InteractionsGene']' #excluded_genes = "[]" #included_models = "[]" # Auxiliary to included_models #excluded_models = "[]" #included_scorers = "[]" # Auxiliary to included_scorers #excluded_scorers = "[]" # Whether to enable XGBoost GBM models ('auto'/'on'/'off') #enable_xgboost_gbm = "auto" # Whether to enable XGBoost Dart models ('auto'/'on'/'off') #enable_xgboost_dart = "auto" # Internal threshold for number of rows x number of columns to trigger no xgboost models due to high memory use # Overridden if enable_xgboost_gbm = "on" or enable_xgboost_dart = "on", in which case always allow each model type to be used #xgboost_threshold_data_size_large = 100000000 # Internal threshold for number of rows x number of columns to trigger no xgboost models due to limits on GPU memory capability # Overridden if enable_xgboost_gbm = "on" or enable_xgboost_dart = "on", in which case always allow each model type to be used #xgboost_gpu_threshold_data_size_large = 30000000 # Whether to enable GLM models ('auto'/'on'/'off') #enable_glm = "auto" # Whether to enable Decision Tree models ('auto'/'on'/'off') #enable_decision_tree = "auto" # Whether to enable LightGBM models ('auto'/'on'/'off') #enable_lightgbm = "auto" # Whether to enable TensorFlow models ('auto'/'on'/'off') #enable_tensorflow = "auto" # Whether to enable FTRL support (beta version, no mojo) (follow the regularized leader) model ('auto'/'on'/'off') #enable_ftrl = "auto" # Whether to enable RuleFit support (beta version, no mojo) ('auto'/'on'/'off') #enable_rulefit = "auto" # Which boosting types to enable for LightGBM (gbdt = boosted trees, rf_early_stopping = random forest with early stopping rf = random forest (no early stopping), dart = drop-out boosted trees with no early stopping #enable_lightgbm_boosting_types = "['gbdt']" # Whether to enable LightGBM categorical feature support (only CPU mode currently) #enable_lightgbm_cat_support = false # Whether to enable constant models ('auto'/'on'/'off') #enable_constant_model = "auto" # Whether to show constant models in iteration panel #show_constant_model = false #drop_constant_model_final_ensemble = true # Parameters for LightGBM to override DAI parameters # parameters should be given as XGBoost equivalent unless unique LightGBM parameter # e.g. 'eval_metric' instead of 'metric' should be used # e.g. params_lightgbm = "{'objective': 'binary:logistic', 'n_estimators': 100, 'max_leaves': 64, 'random_state': 1234}" # e.g. params_lightgbm = {'n_estimators': 600, 'learning_rate': 0.1, 'reg_alpha': 0.0, 'reg_lambda': 0.5, 'gamma': 0, 'max_depth': 0, 'max_bin': 128, 'max_leaves': 256, 'scale_pos_weight': 1.0, 'max_delta_step': 3.469919910597877, 'min_child_weight': 1, 'subsample': 0.9, 'colsample_bytree': 0.3, 'tree_method': 'gpu_hist', 'grow_policy': 'lossguide', 'min_data_in_bin': 3, 'min_child_samples': 5, 'early_stopping_rounds': 20, 'num_classes': 2, 'objective': 'binary:logistic', 'eval_metric': 'logloss', 'random_state': 987654, 'early_stopping_threshold': 0.01, 'monotonicity_constraints': False, 'silent': True, 'debug_verbose': 0, 'subsample_freq': 1}" # avoid including "system"-level parameters like 'n_gpus': 1, 'gpu_id': 0, , 'n_jobs': 1, 'booster': 'lightgbm' # also likely should avoid parameters like: 'objective': 'binary:logistic', unless one really knows what one is doing (e.g. alternative objectives) # See: https://xgboost.readthedocs.io/en/latest/parameter.html # And see: https://github.com/Microsoft/LightGBM/blob/master/docs/Parameters.rst # Can also pass objective parameters if choose (or in case automatically chosen) certain objectives # https://lightgbm.readthedocs.io/en/latest/Parameters.html#metric-parameters #params_lightgbm = "{}" # Parameters for XGBoost to override DAI parameters # similar parameters as lightgbm since lightgbm parameters are transcribed from xgboost equivalent versions # e.g. params_xgboost = '{'n_estimators': 100, 'max_leaves': 64, 'max_depth': 0, 'random_state': 1234}' # See: https://xgboost.readthedocs.io/en/latest/parameter.html #params_xgboost = "{}" # Like params_xgboost but for XGBoost's dart method #params_dart = "{}" # Parameters for TensorFlow to override DAI parameters # e.g. params_tensorflow = '{'lr': 0.01, 'add_wide': False, 'add_attention': True, 'epochs': 30, 'layers': [100, 100], 'activation': 'selu', 'batch_size': 64, 'chunk_size': 1000, 'dropout': 0.3, 'strategy': 'one_shot', 'l1': 0.0, 'l2': 0.0, 'ort_loss': 0.5, 'ort_loss_tau': 0.01, 'normalize_type': 'streaming'}' # See: https://keras.io/ , e.g. for activations: https://keras.io/activations/ # Example layers: [500, 500, 500], [100, 100, 100], [100, 100], [50, 50] # Strategies: '1cycle' or 'one_shot', See: https://github.com/fastai/fastai # normalize_type: 'streaming' or 'global' (using sklearn StandardScaler) #params_tensorflow = "{}" # Parameters for XGBoost's gblinear to override DAI parameters # e.g. params_gblinear = '{'n_estimators': 100}' # See: https://xgboost.readthedocs.io/en/latest/parameter.html #params_gblinear = "{}" # Parameters for Decision Tree to override DAI parameters # parameters should be given as XGBoost equivalent unless unique LightGBM parameter # e.g. 'eval_metric' instead of 'metric' should be used # e.g. params_decision_tree = "{'objective': 'binary:logistic', 'n_estimators': 100, 'max_leaves': 64, 'random_state': 1234}" # e.g. params_decision_tree = {'n_estimators': 1, 'learning_rate': 1, 'reg_alpha': 0.0, 'reg_lambda': 0.5, 'gamma': 0, 'max_depth': 0, 'max_bin': 128, 'max_leaves': 256, 'scale_pos_weight': 1.0, 'max_delta_step': 3.469919910597877, 'min_child_weight': 1, 'subsample': 0.9, 'colsample_bytree': 0.3, 'tree_method': 'gpu_hist', 'grow_policy': 'lossguide', 'min_data_in_bin': 3, 'min_child_samples': 5, 'early_stopping_rounds': 20, 'num_classes': 2, 'objective': 'binary:logistic', 'eval_metric': 'logloss', 'random_state': 987654, 'early_stopping_threshold': 0.01, 'monotonicity_constraints': False, 'silent': True, 'debug_verbose': 0, 'subsample_freq': 1}" # avoid including "system"-level parameters like 'n_gpus': 1, 'gpu_id': 0, , 'n_jobs': 1, 'booster': 'lightgbm' # also likely should avoid parameters like: 'objective': 'binary:logistic', unless one really knows what one is doing (e.g. alternative objectives) # See: https://xgboost.readthedocs.io/en/latest/parameter.html # And see: https://github.com/Microsoft/LightGBM/blob/master/docs/Parameters.rst # Can also pass objective parameters if choose (or in case automatically chosen) certain objectives # https://lightgbm.readthedocs.io/en/latest/Parameters.html#metric-parameters #params_decision_tree = "{}" # Parameters for Rulefit to override DAI parameters # e.g. params_rulefit = '{'max_leaves': 64}' # See: https://xgboost.readthedocs.io/en/latest/parameter.html #params_rulefit = "{}" # Parameters for FTRL to override DAI parameters #params_ftrl = "{}" # Dictionary of key:lists of values to use for LightGBM tuning, overrides DAI's choice per key # e.g. params_tune_lightgbm = '{'min_child_samples': [1,2,5,100,1000], 'min_data_in_bin': [1,2,3,10,100,1000]}' #params_tune_lightgbm = "{}" # Like params_tune_lightgbm but for XGBoost # e.g. params_tune_xgboost = '{'max_leaves': [8, 16, 32, 64]}' #params_tune_xgboost = "{}" # Like params_tune_lightgbm but for XGBoost's Dart # e.g. params_tune_dart = '{'max_leaves': [8, 16, 32, 64]}' #params_tune_dart = "{}" # Like params_tune_lightgbm but for TensorFlow # e.g. params_tune_tensorflow = '{'layers': [[10,10,10], [10, 10, 10, 10]]}' #params_tune_tensorflow = "{}" # Like params_tune_lightgbm but for gblinear # e.g. params_tune_gblinear = '{'reg_lambda': [.01, .001, .0001, .0002]}' #params_tune_gblinear = "{}" # Like params_tune_lightgbm but for rulefit # e.g. params_tune_rulefit = '{'max_depth': [4, 5, 6]}' #params_tune_rulefit = "{}" # Like params_tune_lightgbm but for ftrl #params_tune_ftrl = "{}" # Whether to force max_leaves and max_depth to be 0 if grow_policy is depthwise and lossguide, respectively. #params_tune_grow_policy_simple_trees = true # Maximum number of GBM trees or GLM iterations # Early-stopping usually chooses less #max_nestimators = 3000 # LightGBM dart mode and normal rf mode do not use early stopping and will sample from these values for n_estimators. #n_estimators_list_no_early_stopping = [50, 100, 200, 300] # Lower limit on learning rate for final ensemble GBM models. # In some cases, the maximum number of treess/iterations is insufficient for the final learning rate, # which can lead to no early stopping triggered and poor final model performance. # Then, one can try increasing the learning rate by raising this minimum, # or one can try increasing the maximum number of trees/iterations. # #min_learning_rate_final = 0.01 # Upper limit on learning rate for final ensemble GBM models #max_learning_rate_final = 0.05 # factor by which max_nestimators is reduced for tuning and feature evolution #max_nestimators_feature_evolution_factor = 0.2 # Lower limit on learning rate for feature engineering GBM models #min_learning_rate = 0.05 # Upper limit on learning rate for GBM models # If want to override min_learning_rate and min_learning_rate_final, set this to smaller value #max_learning_rate = 0.5 # Max. number of epochs for TensorFlow and FTRL models #max_epochs = 10 # Maximum tree depth (and corresponding max max_leaves as 2**max_max_depth) #max_max_depth = 12 # Default max_bin for tree methods #default_max_bin = 256 # Default max_bin for lightgbm (recommended for GPU lightgbm) #default_lightgbm_max_bin = 64 # Maximum max_bin for tree features #max_max_bin = 256 # Minimum max_bin for any tree #min_max_bin = 32 # Amount of memory which can handle max_bin = 256 can handle 125 columns and max_bin = 32 for 1000 columns # As available memory on system goes higher than this scale, can handle proportionally more columns at higher max_bin # Currently set to 10GB #scale_mem_for_max_bin = 10737418240 # Factor by which rf gets more depth than gbdt #factor_rf = 1.25 # Whether TensorFlow will use all CPU cores, or if it will split among all transformers #tensorflow_use_all_cores = true # Whether TensorFlow will use all CPU cores if reproducible is set, or if it will split among all transformers #tensorflow_use_all_cores_even_if_reproducible_true = false # How many cores to use for each TensorFlow model, regardless if GPU or CPU based (0 = auto mode) #tensorflow_cores = 0 # Max number of rules to be used for RuleFit models (-1 for all) #rulefit_max_num_rules = -1 # Max tree depth for RuleFit models #rulefit_max_tree_depth = 6 # Max number of trees for RuleFit models #rulefit_max_num_trees = 100 # Internal threshold for number of rows x number of columns to trigger no rulefit models due to being too slow currently #rulefit_threshold_data_size_large = 100000000 # Enable One-Hot-Encoding (which does binning to limit to number of bins to no more than 100 anyway) for categorical columns with fewer than this many unique values # Set to 0 to disable #one_hot_encoding_cardinality_threshold = 50 # Fixed ensemble_level # -1 = auto, based upon ensemble_accuracy_switch, accuracy, size of data, etc. # 0 = No ensemble, only final single model on validated iteration/tree count # 1 = 1 model, multiple ensemble folds (cross-validation) # 2 = 2 models, multiple ensemble folds (cross-validation) # 3 = 3 models, multiple ensemble folds (cross-validation) # 4 = 4 models, multiple ensemble folds (cross-validation) #fixed_ensemble_level = -1 # If enabled, use cross-validation to determine optimal parameters for single final model, # and to be able to create training holdout predictions. #cross_validate_single_final_model = true # Number of models to tune during pre-evolution phase # Can make this lower to avoid excessive tuning, or make higher to do enhanced tuning. # -1 : auto # #parameter_tuning_num_models = -1 #validate_meta_learner = true #validate_meta_learner_extra = false # Specify the fixed number of cross-validation folds (if >= 2) for feature evolution. (The actual number of splits allowed can be less and is determined at experiment run-time). #fixed_num_folds_evolution = -1 # Specify the fixed number of cross-validation folds (if >= 2) for the final model. (The actual number of splits allowed can be less and is determined at experiment run-time). #fixed_num_folds = -1 # set "on" to force only first fold for models - useful for quick runs regardless of data #fixed_only_first_fold_model = "auto" #num_fold_ids_show = 10 #fold_scores_instability_warning_threshold = 0.25 # Upper limit on the number of rows x number of columns for feature evolution (applies to both training and validation/holdout splits) # feature evolution is the process that determines which features will be derived. # Depending on accuracy settings, a fraction of this value will be used #feature_evolution_data_size = 100000000 # Upper limit on the number of rows x number of columns for training final pipeline. # #final_pipeline_data_size = 500000000 # Smaller values can speed up final pipeline model training, as validation data is only used for early stopping. # Note that final model predictions and scores will always be provided on the full dataset provided. # #max_validation_to_training_size_ratio_for_final_ensemble = 2.0 # Ratio of minority to majority class of the target column beyond which stratified sampling is done for binary classification. Otherwise perform random sampling. Set to 0 to always do random sampling. Set to 1 to always do stratified sampling. #force_stratified_splits_for_imbalanced_threshold_binary = 0.01 # Sampling method for imbalanced binary classification problems. Choices are: # "auto": sample both classes as needed, depending on data # "over_under_sampling": over-sample the minority class and under-sample the majority class, depending on data # "under_sampling": under-sample the majority class to reach class balance # "off": do not perform any sampling # #imbalance_sampling_method = "off" # For imbalanced binary classification: ratio of majority to minority class equal and above which to enable # special imbalanced models with sampling techniques (specified by imbalance_sampling_method) to attempt to improve model performance. #imbalance_ratio_sampling_threshold = 5 # For heavily imbalanced binary classification: ratio of majority to minority class equal and above which to enable only # special imbalanced models on full original data, without upfront sampling. #heavy_imbalance_ratio_sampling_threshold = 25 # -1: automatic #imbalance_sampling_number_of_bags = -1 # -1: automatic #imbalance_sampling_max_number_of_bags = 10 # Only for shift/leakage/tuning/feature evolution models. Not used for final models. Final models can # be limited by imbalance_sampling_max_number_of_bags. #imbalance_sampling_max_number_of_bags_feature_evolution = 3 # Max. size of data sampled during imbalanced sampling (in terms of dataset size), # controls number of bags (approximately). Only for imbalance_sampling_number_of_bags == -1. #imbalance_sampling_max_multiple_data_size = 1.0 # Rank averaging can be helpful when ensembling diverse models when ranking metrics like AUC/Gini # metrics are optimized. No MOJO support yet. #imbalance_sampling_rank_averaging = "auto" # A value of 0.5 means that models/algorithms will be presented a balanced target class distribution # after applying under/over-sampling techniques on the training data. Sometimes it makes sense to # choose a smaller value like 0.1 or 0.01 when starting from an extremely imbalanced original target # distribution. -1.0: automatic #imbalance_sampling_target_minority_fraction = -1.0 # For binary classification: ratio of majority to minority class equal and above which to notify # of imbalance in GUI to say slightly imbalanced. # More than imbalance_ratio_sampling_threshold will say problem is imbalanced. # #imbalance_ratio_notification_threshold = 2.0 # list of possible bins for FTRL (largest is default best value) #nbins_ftrl_list = "[1000000, 10000000, 100000000]" # Samples the number of automatic FTRL interactions terms to no more than this value (for each of 2nd, 3rd, 4th order terms) #ftrl_max_interaction_terms_per_degree = 10000 # list of possible bins for target encoding (first is default value) #te_bin_list = "[25, 10, 100, 250]" # list of possible bins for weight of evidence encoding (first is default value) # If only want one value: woe_bin_list = [2] #woe_bin_list = "[25, 10, 100, 250]" # list of possible bins for ohe hot encoding (first is default value) #ohe_bin_list = "[10, 25, 50, 75, 100]" # Whether to drop columns with constant values #drop_constant_columns = true # Whether to drop columns that appear to be an ID #drop_id_columns = true # Whether to avoid dropping any columns (original or derived) #no_drop_features = false # Direct control over columns to drop in bulk so can copy-paste large lists instead of selecting each one separately in GUI #cols_to_drop = "" # Control over columns to group by, default is empty list that means DAI automatically searches all columns, # selected randomly or by which have top variable importance. #cols_to_group_by = "" # Whether to sample from given features to group by (True) or to always group by all features (False). #sample_cols_to_group_by = false # Aggregation functions to use for groupby operations. #agg_funcs_for_group_by = "['mean', 'sd', 'min', 'max', 'count']" # Out of fold aggregations ensure less overfitting, but see less data in each fold. #folds_for_group_by = 5 # Strategy to apply when doing mutations on transformers. # Sample mode is default, with tendency to sample transformer parameters. # Batched mode tends to do multiple types of the same transformation together. # Full mode does even more types of the same transformation together. # #mutation_mode = "sample" # Whether to enable checking text for shift, currently only via label encoding. #shift_check_text = false # Whether to use LightGBM random forest mode without early stopping for shift detection. #use_rf_for_shift_if_have_lgbm = true # Normalized training variable importance above which to check the feature for shift # Useful to avoid checking likely unimportant features #shift_key_features_varimp = 0.01 # Whether to only check certain features based upon the value of shift_key_features_varimp #shift_check_reduced_features = true # Number of trees to use to train model to check shift in distribution # No larger than max_nestimators #shift_trees = 100 # The value of max_bin to use for trees to use to train model to check shift in distribution #shift_max_bin = 256 # The min. value of max_depth to use for trees to use to train model to check shift in distribution #shift_min_max_depth = 4 # The max. value of max_depth to use for trees to use to train model to check shift in distribution #shift_max_max_depth = 8 # If distribution shift detection is enabled, show features for which shift AUC is above this value # (AUC of a binary classifier that predicts whether given feature value belongs to train or test data) #detect_features_distribution_shift_threshold_auc = 0.55 # Minimum number of features to keep, keeping least shifted feature at least if 1 #drop_features_distribution_shift_min_features = 1 # Whether to enable checking text for leakage, currently only via label encoding. #leakage_check_text = true # Normalized training variable importance (per 1 minus AUC/R2 to control for leaky varimp dominance) above which to check the feature for leakage # Useful to avoid checking likely unimportant features #leakage_key_features_varimp = 0.001 # Like leakage_key_features_varimp, but applies if early stopping disabled when can trust multiple leaks to get uniform varimp. #leakage_key_features_varimp_if_no_early_stopping = 0.05 # Whether to only check certain features based upon the value of leakage_key_features_varimp. If any feature has AUC near 1, will consume all variable importance, even if another feature is also leaky. So False is safest option, but True generally good if many columns. #leakage_check_reduced_features = true # Whether to use LightGBM random forest mode without early stopping for leakage detection. #use_rf_for_leakage_if_have_lgbm = true # Number of trees to use to train model to check for leakage # No larger than max_nestimators #leakage_trees = 100 # The value of max_bin to use for trees to use to train model to check for leakage #leakage_max_bin = 256 # The value of max_depth to use for trees to use to train model to check for leakage #leakage_min_max_depth = 4 # The value of max_depth to use for trees to use to train model to check for leakage #leakage_max_max_depth = 8 # When leakage detection is enabled, if AUC (R2 for regression) on original data (label-encoded) # is above or equal to this value, then trigger per-feature leakage detection #detect_features_leakage_threshold_auc = 0.95 # When leakage detection is enabled, show features for which AUC (R2 for regression, # for whether that predictor/feature alone predicts the target) is above or equal to this value. # Feature is dropped if AUC/R2 is above or equal to drop_features_leakage_threshold_auc #detect_features_per_feature_leakage_threshold_auc = 0.8 # Minimum number of features to keep, keeping least leakage feature at least if 1 #drop_features_leakage_min_features = 1 # Ratio of train to validation holdout when testing for leakage #leakage_train_test_split = 0.25 # Whether to enable detailed traces (in GUI Trace) #detailed_traces = false # Whether to enable debug log level (in log files) #debug_log = false # Whether to add logging of system information such as CPU, GPU, disk space at the start of each experiment log. Same information is already logged in system logs. #log_system_info_per_experiment = true # How close to the optimal value (usually 1 or 0) does the validation score need to be to be considered perfect (to stop the experiment)? #abs_tol_for_perfect_score = 0.0001 # Timeout in seconds to wait for data ingestion. #data_ingest_timeout = 86400.0 # Enable time series recipe #time_series_recipe = true # Provide date or datetime timestamps (in same format as the time column) for custom training and validation splits like this: "tr_start1, tr_end1, va_start1, va_end1, ..., tr_startN, tr_endN, va_startN, va_endN" #time_series_validation_fold_split_datetime_boundaries = "" # Timeout in seconds for time-series properties detection in UI. #timeseries_split_suggestion_timeout = 30.0 # Whether to use lag transformers when using causal-split for validation (as occurs when not using time-based lag recipe). # If no time groups columns, lag transformers will still use time-column as sole time group column. # #use_lags_if_not_time_series_recipe = false # earliest datetime for automatic conversion of integers in %Y%m%d format to a time column during parsing #min_ymd_timestamp = 19000101 # lastet datetime for automatic conversion of integers in %Y%m%d format to a time column during parsing #max_ymd_timestamp = 21000101 # maximum number of data samples (randomly selected rows) for date/datetime format detection #max_rows_datetime_format_detection = 100000 # Automatically generate is-holiday features from date columns #holiday_features = true # County code to use to look up holiday calendar (Python package 'holiday') #holiday_country = "US" # Max. sample size for automatic determination of time series train/valid split properties, only if time column is selected #max_time_series_properties_sample_size = 250000 # Maximum number of lag sizes to use for lags-based time-series experiments. are sampled from if sample_lag_sizes==True, else all are taken (-1 == automatic) #max_lag_sizes = 30 # Minimum required autocorrelation threshold for a lag to be considered for feature engineering #min_lag_autocorrelation = 0.1 # How many samples of lag sizes to use for a single time group (single time series signal) #max_signal_lag_sizes = 100 # Whether to sample lag sizes #sample_lag_sizes = false # How many samples of lag sizes to use, chosen randomly out of original set of lag sizes #max_sampled_lag_sizes = 10 # Override lags to be used # e.g. [7, 14, 21] # this exact list # e.g. 21 # produce from 1 to 21 # e.g. 21:3 produce from 1 to 21 in step of 3 # e.g. 5-21 produce from 5 to 21 # e.g. 5-21:3 produce from 5 to 21 in step of 3 #override_lag_sizes = [] # Smallest considered lag size #min_lag_size = -1 # Whether to enable feature engineering based on selected time column, e.g. Date~weekday. #allow_time_column_as_feature = true # Whether to enable integer time column to be used as a numeric feature. # If using time series recipe, using time column (numeric time stamps) as input features can lead to model that # memorizes the actual time stamps instead of features that generalize to the future. # #allow_time_column_as_numeric_feature = false # Allowed date or date-time transformations. # Date transformers include: year, quarter, month, week, weekday, day, dayofyear, num. # Date transformers also include: hour, minute, second. # Features in DAI will show up as get_ + transformation name. # E.g. num is a direct numeric value representing the floating point value of time, # which can lead to over-fitting if used on IID problems. So this is turned off by default. #datetime_funcs = "['year', 'quarter', 'month', 'week', 'weekday', 'day', 'dayofyear', 'hour', 'minute', 'second']" # Whether to consider time groups columns (tgc) as standalone features. # Note that 'time_column' is treated separately via 'Allow to engineer features from time column'. # Use allowed_coltypes_for_tgc_as_features for control per feature type. # #allow_tgc_as_features = false # Which time groups columns (tgc) feature types to consider as standalone features, # if the corresponding flag "Consider time groups columns as standalone features" is set to true. # E.g. all column types would be ["numeric", "categorical", "ohe_categorical", "datetime", "date", "text"] # Note that 'time_column' is treated separately via 'Allow to engineer features from time column'. # Note that if lag-based time series recipe is disabled, then all tgc are allowed features. # #allowed_coltypes_for_tgc_as_features = "['numeric', 'categorical', 'ohe_categorical', 'datetime', 'date', 'text']" # Whether various transformers (clustering, truncated SVD) are enabled, # that otherwise would be disabled for time series due to # potential to overfit by leaking across time within the fit of each fold. #enable_time_unaware_transformers = "auto" # Whether to group by all time groups columns for creating lag features, instead of sampling from them #tgc_only_use_all_groups = true # Enable creation of holdout predictions on training data # using moving windows (useful for MLI, but can be slow) #time_series_holdout_preds = true # Set fixed number of time-based splits for internal model validation (actual number of splits allowed can be less and is determined at experiment run-time). #time_series_validation_splits = -1 # Maximum overlap between two time-based splits. Higher values increase the amount of possible splits. #time_series_splits_max_overlap = 0.5 # Max number of splits used for creating final time-series model's holdout/backtesting predictions. With the default value '-1' the same amount of splits as during model validation will be used. Use 'time_series_validation_splits' to control amount of time-based splits used for model validation. #time_series_max_holdout_splits = -1 # Whether to speed up time-series holdout predictions for back-testing on training data (used for MLI and metrics calculation). Can be slightly less accurate. #mli_ts_fast_approx = false # Whether to speed up Shapley values for time-series holdout predictions for back-testing on training data (used for MLI). Can be slightly less accurate. #mli_ts_fast_approx_contribs = true # Enable creation of Shapley values for holdout predictions on training data # using moving windows (useful for MLI, but can be slow), at the time of the experiment. If disabled, MLI will # generate Shapley values on demand. #mli_ts_holdout_contribs = true # Values of 5 or more can improve generalization by more aggressive dropping of least important features. Set to 1 to disable. #time_series_min_interpretability = 5 # Dropout mode for lag features in order to achieve an equal n.a.-ratio between train and validation/test. The independent mode performs a simple feature-wise dropout, whereas the dependent one takes lag-size dependencies per sample/row into account. #lags_dropout = "dependent" # Normalized probability of choosing to lag non-targets relative to targets (-1.0 = auto) #prob_lag_non_targets = -1.0 # Method to create rolling test set predictions, if the forecast horizon is shorter than the time span of the test set. One can choose between test time augmentation (TTA) and a successive refitting of the final pipeline. #rolling_test_method = "tta" # Probability for new Lags/EWMA gene to use default lags (determined by frequency/gap/horizon, independent of data) (-1.0 = auto) #prob_default_lags = -1.0 # Unnormalized probability of choosing other lag time-series transformers based on interactions (-1.0 = auto) #prob_lagsinteraction = -1.0 # Unnormalized probability of choosing other lag time-series transformers based on aggregations (-1.0 = auto) #prob_lagsaggregates = -1.0 # Maximum amount of columns send from UI to backend in order to auto-detect TGC #tgc_via_ui_max_ncols = 10 # Maximum frequency of duplicated timestamps for TGC detection #tgc_dup_tolerance = 0.01 # Build (if missing but available) the MOJO pipeline to be used for predictions in MLI. For certain models this can improve performance. #mli_use_mojo_pipeline = false # When number of rows are above this limit sample for MLI for scoring UI data #mli_sample_above_for_scoring = 1000000 # When number of rows are above this limit sample for MLI for training surrogate models #mli_sample_above_for_training = 100000 # When sample for MLI how many rows to sample #mli_sample_size = 100000 # how many bins to do quantile binning #mli_num_quantiles = 10 # mli random forest number of trees #mli_drf_num_trees = 100 # Whether to speed up predictions used inside MLI with a fast approximation #mli_fast_approx = true # mli number of trees for fast_approx during predict for Shapley #fast_approx_num_trees = 50 # whether to do only 1 fold and 1 model of all folds and models if ensemble #fast_approx_do_one_fold_one_model = true # mli random forest max depth #mli_drf_max_depth = 20 # not only sample training, but also sample scoring #mli_sample_training = true # regularization strength for k-LIME GLM's #klime_lambda = "[1e-06, 1e-08]" # regularization strength for k-LIME GLM's #klime_alpha = 0.0 # mli converts numeric columns to enum when cardinality is <= this value #mli_max_numeric_enum_cardinality = 25 # Maximum number of features allowed for k-LIME k-means clustering #mli_max_number_cluster_vars = 6 # Use all columns for k-LIME k-means clustering (this will override mli_max_number_cluster_vars if set to True #use_all_columns_klime_kmeans = false # Strict version check for MLI #mli_strict_version_check = true # MLI cloud name #mli_cloud_name = "" # Compute original model ICE using per feature's bin predictions (true) or use "one frame" strategy (false) #mli_ice_per_bin_strategy = false # By default DIA will run for categorical columns with cardinality <= mli_dia_default_max_cardinality #mli_dia_default_max_cardinality = 10 # Enable MLI Sensitivity Analysis #enable_mli_sa = true # When number of rows are above this limit, then sample for DIA #mli_dia_sample_size = 100000 # When number of rows are above this limit, then sample for DAI PD/ICE #mli_pd_sample_size = 100000 # When number of rows are above this limit, then sample for MLI Shapley calculation #mli_shapley_sample_size = 100000 # Enable async/await-based non-blocking MLI API #enable_mli_async_api = true # Enable main chart aggregator in Sensitivity Analysis #enable_mli_sa_main_chart_aggregator = true # When to sample for Sensitivity Analysis (number of rows after sampling) #mli_sa_sampling_limit = 500000 # Run main chart aggregator in Sensitivity Analysis when the number of dataset instances is bigger than given limit #mli_sa_main_chart_aggregator_limit = 1000 # Use predict_safe() (true) or predict_base() (false) in MLI (PD, ICE, SA, ... #mli_predict_safe = false # Allow predict method with fast approximation in MLI (PD, ICE, SA, ... #enable_mli_predict_fast_approx = false # Number of max retries should the surrogate model fail to build. #mli_max_surrogate_retries = 5 # Number of rows per batch when scoring using MOJO. #mli_mojo_batch_size = 50 # Tokenizer used to extract tokens from text columns for MLI. #mli_nlp_tokenizer = "tfidf" # Number of tokens used for MLI NLP explanations. -1 means all. #mli_nlp_top_n = 20 # Maximum number of records on which we'll perform MLI NLP #mli_nlp_sample_limit = 10000 # Number of parallel workers when scoring using MOJO in MLI NLP. #mli_nlp_workers = 4 # Minimum number of documents in which token has to appear. Integer mean absolute count, float means percentage. #mli_nlp_min_df = 3 # Maximum number of documents in which token has to appear. Integer mean absolute count, float means percentage. #mli_nlp_max_df = 0.9 # The minimum value in the ngram range. The tokenizer will generate all possible tokens in the (mli_nlp_min_ngram, mli_nlp_max_ngram) range. #mli_nlp_min_ngram = 1 # The maximum value in the ngram range. The tokenizer will generate all possible tokens in the (mli_nlp_min_ngram, mli_nlp_max_ngram) range. #mli_nlp_max_ngram = 1 # Mode used to choose N tokens for MLI NLP. # "top" chooses N top tokens. # "bottom" chooses N bottom tokens. # "top-bottom" chooses math.floor(N/2) top and math.ceil(N/2) bottom tokens. # "linspace" chooses N evenly spaced out tokens. #mli_nlp_min_token_mode = "top" # The number of top tokens to be used as features when building token based feature importance. #mli_nlp_tokenizer_max_features = -1 # The number of top tokens to be used as features when computing text LOCO. #mli_nlp_loco_max_features = -1 # The number of top tokens to be used as features when building surrogate models. #mli_nlp_surrogate_tokens = 100 # Whether to dump every scored individual's variable importance (both derived and original) to csv/tabulated/json file # produces files like: individual_scored_id%d.iter%d*features* #dump_varimp_every_scored_indiv = false # Whether to dump every scored individual's model parameters to csv/tabulated/json file # produces files like: individual_scored.params.[txt, csv, json] #dump_modelparams_every_scored_indiv = true # Number of features to show in model dump every scored individual #dump_modelparams_every_scored_indiv_feature_count = 3 # Whether to append (false) or have separate files, files like: individual_scored_id%d.iter%d*params*, (true) for modelparams every scored indiv #dump_modelparams_separate_files = false # Whether to dump every scored fold's timing and feature info to a *timings*.txt file #dump_trans_timings = false # whether to delete preview timings if wrote transformer timings #delete_preview_trans_timings = true # Location of the AutoDoc template #autodoc_template = "report_template.docx" # Specify the output of the report. # Options are docx or md. #autodoc_output_type = "docx" # Specify the name of the report. #autodoc_report_name = "report" # Specify the maximum number of classes in the confusion # matrix. #autodoc_max_cm_size = 10 # Set the number of models for which a glm coefficients # table is shown in the Autoreport. coef_table_num_models must # be -1 or an integer >= 1 (-1 shows all models). #autodoc_coef_table_num_models = 1 # Set the number of folds per model for which a glm # coefficients table is shown in the Autoreport. coef_table_num_folds # must be -1 or an integer >= 1 (-1 shows all folds per model). #autodoc_coef_table_num_folds = -1 # Set the number of coefficients to show within a glm # coefficients table in the Autoreport. coef_table_num_coef, controls # the number of rows shown in a glm table and must be -1 or # an integer >= 1 (-1 shows all coefficients). #autodoc_coef_table_num_coef = 50 # Set the number of classes to show within a glm # coefficients table in the Autoreport. coef_table_num_classes controls # the number of class-columns shown in a glm table and must be -1 or # an integer >= 4 (-1 shows all classes). #autodoc_coef_table_num_classes = 9 # Specify whether to show the full glm coefficient # table(s) in the appendix. coef_table_appendix_results_table must be # a boolean: True to show tables in appendix, False to not show them. #autodoc_coef_table_appendix_results_table = false # Specify the minimum relative importance in order # for a feature to be displayed. autodoc_min_relative_importance # must be a float >= 0 and < 1. #autodoc_min_relative_importance = 0.003 # Specify the number of top features to display in # the document. setting to -1 disables this restriction #autodoc_num_features = 50 # Specify the number of rows to include in PDP and ICE plot # if individual rows are not specified. #autodoc_num_rows = 0 # Maximum number of seconds Partial Dependency computation # can take when generating report. Set to -1 for no time limit. #autodoc_pd_max_runtime = 20 # Number of standard deviations outside of the range of # a column to include in partial dependence plots. This shows how the # model will react to data it has not seen before. #autodoc_out_of_range = 3 # Number of columns to be show in data summary. Value # must be an integer. Values lower than 1, f.e. 0 or -1, indicate that # all columns should be shown. #autodoc_data_summary_col_num = -1 # Whether to include prediction statistics information if # experiment is binary classification/regression. #autodoc_prediction_stats = false # Number of quantiles to use for prediction statistics # computation. #autodoc_prediction_stats_n_quantiles = 20 # Whether to include population stability index if # experiment is binary classification/regression. #autodoc_population_stability_index = false # Number of quantiles to use for population stability index # computation. #autodoc_population_stability_index_n_quantiles = 10 # Whether to compute permutation based feature importance. #autodoc_include_permutation_feature_importance = false # Name of the scorer to be used to calculate feature # importance. Leave blank to use experiments default scorer #autodoc_feature_importance_scorer = "" # Number of permutations to make per feature when computing # feature importance. #autodoc_feature_importance_num_perm = 1 # Whether to include response rates information if # experiment is binary classification. #autodoc_response_rate = false # Number of quantiles to use for response rates information # computation. #autodoc_response_rate_n_quantiles = 10 # The number feature in a KLIME global GLM coefficients # table. Must be an integer greater than 0 or -1. To # show all features set to -1. #autodoc_global_klime_num_features = 10 # Set the number of KLIME global GLM coefficients tables. Set # to 1 to show one table with coefficients sorted by absolute # value. Set to 2 to two tables one with the top positive # coefficients and one with the top negative coefficients. Must # be set to the integer 1 or 2. #autodoc_global_klime_num_tables = 1 # Whether to show the Gini Plot. #autodoc_gini_plot = false # Whether to show all config settings. If False, only # the changed settings (config overrides) are listed, otherwise all # settings are listed. #autodoc_list_all_config_settings = false # Maximum number of columns autoviz will work with. # If dataset has more columns than this number, # autoviz will pick columns randomly, prioritizing numerical columns # #autoviz_max_num_columns = 50 # ' # Whether to compute training, validation, and test correlation matrix (table and heatmap pdf) and save to disk # alpha: currently single threaded and slow for many columns #compute_correlation = false # Whether to dump to disk a correlation heatmap #produce_correlation_heatmap = false # Value to report high correlation between original features #high_correlation_value_to_report = 0.95 # Whether to delete preview cache on server exit #preview_cache_upon_server_exit = true # Configurations for a HDFS data source # Path of hdfs coresite.xml # core_site_xml_path is deprecated, please use hdfs_config_path #core_site_xml_path = "" # (Required) HDFS config folder path. Can contain multiple config files. #hdfs_config_path = "" # Path of the principal key tab file. Required when hdfs_aut_type='principal'. #key_tab_path = "" # The option disable access to DAI data_directory from file browser #file_hide_data_directory = true # Enable usage of path filters #file_path_filtering_enabled = false # List of absolute path prefixes to restrict access to in file browser. # For example: # file_path_filter_include = "['/data','/home/michal/']" #file_path_filter_include = [] # (Required) HDFS connector # Auth type can be Principal/keytab/keytabPrincipal # Specify HDFS Auth Type, allowed options are: # noauth : No authentication needed # principal : Authenticate with HDFS with a principal user # keytab : Authenticate with a Key tab (recommended). If running # DAI as a service, then the Kerberos keytab needs to # be owned by the DAI user. # keytabimpersonation : Login with impersonation using a keytab #hdfs_auth_type = "noauth" # Kerberos app principal user. Required when hdfs_aut_type='keytab'; recommended otherwise. #hdfs_app_principal_user = "" # Deprecated - Do Not Use, login user is taken from the user name from login #hdfs_app_login_user = "" # JVM args for HDFS distributions, provide args seperate by space # -Djava.security.krb5.conf=/krb5.conf # -Dsun.security.krb5.debug=True # -Dlog4j.configuration=file:///log4j.properties #hdfs_app_jvm_args = "" # hdfs class path #hdfs_app_classpath = "" # List of supported DFS schemas. Ex. "['hdfs://', 'maprfs://', 'swift://']" # Supported schemas list is used as an initial check to ensure valid input to connector # #hdfs_app_supported_schemes = "['hdfs://', 'maprfs://', 'swift://']" # Maximum number of files viewable in connector ui. Set to larger number to view more files #hdfs_max_files_listed = 100 # Blue Data DTap connector settings are similar to HDFS connector settings. # Specify DTap Auth Type, allowed options are: # noauth : No authentication needed # principal : Authenticate with DTab with a principal user # keytab : Authenticate with a Key tab (recommended). If running # DAI as a service, then the Kerberos keytab needs to # be owned by the DAI user. # keytabimpersonation : Login with impersonation using a keytab # NOTE: "hdfs_app_classpath" and "core_site_xml_path" are both required to be set for DTap connector #dtap_auth_type = "noauth" # Dtap (HDFS) config folder path , can contain multiple config files #dtap_config_path = "" # Path of the principal key tab file #dtap_key_tab_path = "" # Kerberos app principal user (recommended) #dtap_app_principal_user = "" # Specify the user id of the current user here as user@realm #dtap_app_login_user = "" # JVM args for DTap distributions, provide args seperate by space #dtap_app_jvm_args = "" # DTap (HDFS) class path. NOTE: set 'hdfs_app_classpath' also #dtap_app_classpath = "" # S3 Connector credentials #aws_access_key_id = "" # S3 Connector credentials #aws_secret_access_key = "" # S3 Connector credentials #aws_role_arn = "" # What region to use when none is specified in the s3 url. # Ignored when aws_s3_endpoint_url is set. # #aws_default_region = "" # Sets enpoint URL that will be used to access S3. #aws_s3_endpoint_url = "" # If set to true S3 Connector will try to to obtain credentials assiciated with # the role attached to the EC2 instance. #aws_use_ec2_role_credentials = false # Starting S3 path displayed in UI S3 browser #s3_init_path = "s3://" # GCS Connector credentials # example (suggested) -- '/licenses/my_service_account_json.json' #gcs_path_to_service_account_json = "" # Minio Connector credentials #minio_endpoint_url = "" # Minio Connector credentials #minio_access_key_id = "" # Minio Connector credentials #minio_secret_access_key = "" # Recommended Provide: url, user, password # Optionally Provide: account, user, password # Example URL: https://..snowflakecomputing.com # Snowflake Connector credentials #snowflake_url = "" # Snowflake Connector credentials #snowflake_user = "" # Snowflake Connector credentials #snowflake_password = "" # Snowflake Connector credentials #snowflake_account = "" # KDB Connector credentials #kdb_user = "" # KDB Connector credentials #kdb_password = "" # KDB Connector credentials #kdb_hostname = "" # KDB Connector credentials #kdb_port = "" # KDB Connector credentials #kdb_app_classpath = "" # KDB Connector credentials #kdb_app_jvm_args = "" # Azure Blob Store Connector credentials #azure_blob_account_name = "" # Azure Blob Store Connector credentials #azure_blob_account_key = "" # Azure Blob Store Connector credentials #azure_connection_string = "" # Configuration for JDBC Connector. # JSON/Dictionary String with multiple keys. # Format as a single line without using carriage returns (the following example is formatted for readability). # Use triple quotations to ensure that the text is read as a single string. # Example: # '{ # "postgres": { # "url": "jdbc:postgresql://ip address:port/postgres", # "jarpath": "/path/to/postgres_driver.jar", # "classpath": "org.postgresql.Driver" # }, # "mysql": { # "url":"mysql connection string", # "jarpath": "/path/to/mysql_driver.jar", # "classpath": "my.sql.classpath.Driver" # } # }' # #jdbc_app_configs = "{}" # extra jvm args for jdbc connector #jdbc_app_jvm_args = "-Xmx4g" # alternative classpath for jdbc connector #jdbc_app_classpath = "" # Notification scripts # - the variable points to a location of script which is executed at given event in experiment lifecycle # - the script should have executable flag enabled # - use of absolute path is suggested # The on experiment start notification script location #listeners_experiment_start = "" # The on experiment finished notification script location #listeners_experiment_done = "" # Address of the H2O Storage endpoint. Keep empty to use the local storage only. #h2o_storage_address = "" # Whether the channel to the storage should be encrypted. #h2o_storage_tls_enabled = true # Path to the certification authority certificate that H2O Storage server identity will be checked against. #h2o_storage_tls_ca_path = "" # Path to the client certificate to authenticate with H2O Storage server #h2o_storage_tls_cert_path = "" # Path to the client key to authenticate with H2O Storage server #h2o_storage_tls_key_path = "" # UUID of a Storage project to use instead of the remote HOME folder. #h2o_storage_internal_default_project_id = "" # Default AWS credentials to be used for scorer deployments. #deployment_aws_access_key_id = "" # Default AWS credentials to be used for scorer deployments. #deployment_aws_secret_access_key = "" # AWS S3 bucket to be used for scorer deployments. #deployment_aws_bucket_name = "" # Allow the browser to store e.g. login credentials in login form (set to false for higher security) #allow_form_autocomplete = true # Enable Projects workspace (alpha version, for evaluation) #enable_projects = true # Enable custom recipes. #enable_custom_recipes = true # Enable uploading of custom recipes. #enable_custom_recipes_upload = true # Include custom recipes in default inclusion lists (warning: enables all custom recipes) #include_custom_recipes_by_default = false # Default application language - options are 'en', 'ja' #app_language = "en" # Whether to check if config.toml keys are valid and fail if not valid #check_invalid_config_toml_keys = true #enable_funnel = true #clean_funnel = true #quiet_funnel = false #debug_daimodel_level = 0 #interaction_finder_max_rows_x_cols = 200000.0 #interaction_finder_corr_threshold = 0.95 # Required GINI relative improvement for InteractionTransformer. # If GINI is not better than this relative improvement compared to original features considered # in the interaction, then the interaction is not returned. If noisy data, and no clear signal # in interactions but still want interactions, then can decrease this number. #interaction_finder_gini_rel_improvement_threshold = 0.5 # Number of transformed Interactions to make as best out of many generated trial interactions. #interaction_finder_return_limit = 5 # Whether to enable bootstrap sampling. Provides error bars to validation and test scores based on the standard error of the bootstrap mean. #enable_bootstrap = true # Minimum number of bootstrap samples to use for estimating score and its standard deviation # Actual number of bootstrap samples will vary between the min and max, # depending upon row count (more rows, fewer samples) and accuracy settings (higher accuracy, more samples) # #min_bootstrap_samples = 1 # Maximum number of bootstrap samples to use for estimating score and its standard deviation # Actual number of bootstrap samples will vary between the min and max, # depending upon row count (more rows, fewer samples) and accuracy settings (higher accuracy, more samples) # #max_bootstrap_samples = 100 # Minimum fraction of row size to take as sample size for bootstrap estimator # Actual sample size used for bootstrap estimate will vary between the min and max, # depending upon row count (more rows, smaller sample size) and accuracy settings (higher accuracy, larger sample size) # #min_bootstrap_sample_size_factor = 1.0 # Maximum fraction of row size to take as sample size for bootstrap estimator # Actual sample size used for bootstrap estimate will vary between the min and max, # depending upon row count (more rows, smaller sample size) and accuracy settings (higher accuracy, larger sample size) # #max_bootstrap_sample_size_factor = 10.0 # Seed to use for final model bootstrap sampling, -1 means use experiment-derived seed. # E.g. one can retrain final model with different seed to get different final model error bars for scores. # #bootstrap_final_seed = -1 #varimp_threshold_at_interpretability_10 = 0.05 #features_allowed_by_interpretability = "{1: 10000000, 2: 10000, 3: 1000, 4: 500, 5: 300, 6: 200, 7: 150, 8: 100, 9: 80, 10: 50, 11: 50, 12: 50, 13: 50}" #nfeatures_max_threshold = 200 #rdelta_percent_score_penalty_per_feature_by_interpretability = "{1: 0.0, 2: 0.1, 3: 1.0, 4: 2.0, 5: 5.0, 6: 10.0, 7: 20.0, 8: 30.0, 9: 50.0, 10: 100.0, 11: 100.0, 12: 100.0, 13: 100.0}" #drop_low_meta_weights = true #meta_weight_allowed_by_interpretability = "{1: 1E-7, 2: 1E-5, 3: 1E-4, 4: 1E-3, 5: 1E-2, 6: 0.03, 7: 0.05, 8: 0.08, 9: 0.10, 10: 0.15, 11: 0.15, 12: 0.15, 13: 0.15}" #feature_cost_mean_interp_for_penalty = 5 #features_cost_per_interp = 0.25 #varimp_threshold_shift_report = 0.3 #apply_featuregene_limits_after_tuning = true #remove_scored_0gain_genes_in_postprocessing_above_interpretability = 13 #remove_scored_0gain_genes_in_postprocessing_above_interpretability_final_population = 2 #remove_scored_by_threshold_genes_in_postprocessing_above_interpretability_final_population = 7 # Unnormalized probability to add genes or instances of transformers with specific attributes. # If no genes can be added, other mutations # (mutating models hyper parmaters, pruning genes, pruning features, etc.) are attempted. # #prob_add_genes = 0.5 # Unnormalized probability, conditioned on prob_add_genes, # to add genes or instances of transformers with specific attributes # that have shown to be beneficial to other individuals within the population. # #prob_addbest_genes = 0.5 # Unnormalized probability to prune genes or instances of transformers with specific attributes. # If a variety of transformers with many attributes exists, default value is reasonable. # However, if one has fixed set of transformers that should not change or no new transformer attributes # can be added, then setting this to 0.0 is reasonable to avoid undesired loss of transformations. # #prob_prune_genes = 0.5 # Unnormalized probability change model hyper parameters. # #prob_perturb_xgb = 0.25 # Unnormalized probability to prune features that have low variable importance, # as opposed to pruning entire instances of genes/transformers. # #prob_prune_by_features = 0.25 #max_absolute_feature_expansion = 1000 # Number of classes above which to always use TensorFlow (if TensorFlow is enabled), # instead of others models set on 'auto' (models set to 'on' are still used). #tensorflow_num_classes_switch = 10 # Class count above which do not use TextLin Transformer. #textlin_num_classes_switch = 5 #text_gene_dim_reduction_choices = "[50]" #text_gene_max_ngram = "[1, 2, 3]" # Max. number of top variable importances to show in logs during feature evolution #max_num_varimp_to_log = 10 # Max. number of top variable importance shifts to show in logs and GUI after final model built #max_num_varimp_shift_to_log = 10 # Dictionary to control recipes for each experiment and particular custom recipes. # E.g. if inserting into the GUI as any toml string, can use: # ""recipe_dict="{'key1': 2, 'key2': 'value2'}""" # E.g. if putting into config.toml as a dict, can use: # recipe_dict="{'key1': 2, 'key2': 'value2'}" # #recipe_dict = "{}" # location of custom recipes packages installed (relative to data_directory) # We will try to install packages dynamically, but can also do (before or after server started): # (inside docker running docker instance if running docker, or as user server is running as (e.g. dai user) if deb/tar native installation: # PYTHONPATH=//lib/python3.6/site-packages/ dai-env.sh python -m pip install --prefix=/ --upgrade --upgrade-strategy only-if-needed --log-file pip_log_file.log # where is /opt/h2oai/dai/ for native rpm/deb installation # Note can also install wheel files if is name of wheel file or archive. # #contrib_env_relative_directory = "contrib/env" # pip install retry for call to pip. Sometimes need to try twice #pip_install_overall_retries = 2 # pip install verbosity level (number of -v's given to pip, up to 3 #pip_install_verbosity = 2 # pip install timeout in seconds, Sometimes internet issues would mean want to fail faster #pip_install_timeout = 15 # pip install retry count #pip_install_retries = 5 # pip install options: string of list of other options, e.g. "['--proxy', 'http://user:password@proxyserver:port']" #pip_install_options = "" # Whether to enable basic acceptance testing. Tests if can pickle the state, etc. #enable_basic_acceptance_tests = true # Whether acceptance tests should run for custom genes / models / scorers / etc. #enable_acceptance_tests = true # Minutes to wait until a recipe's acceptance testing is aborted. A recipe is rejected if acceptance # testing is enabled and times out. # One may also set timeout for a specific recipe by setting the class's staticmethod function called # acceptance_test_timeout to return number of minutes to wait until timeout doing acceptance testing. # This timeout does not include the time to install required packages. # #acceptance_test_timeout = 20.0 # Skipping just avoids the failed transformer. # Sometimes python multiprocessing swallows exceptions, # so skipping and logging exceptions is also more reliable way to handle them. # Recipe can raise h2oaicore.systemutils.IgnoreError to ignore error and avoid logging error. # #skip_transformer_failures = true # Skipping just avoids the failed model. Failures are logged depending upon detailed_skip_failure_messages_level." # Recipe can raise h2oaicore.systemutils.IgnoreError to ignore error and avoid logging error. # #skip_model_failures = true # How much verbosity to log failure messages for failed and then skipped transformers or models. # Full failures always go to disk as *.stack files, # which upon completion of experiment goes into details folder within experiment log zip file. # #detailed_skip_failure_messages_level = 1 # Instructions for 'Add to config.toml via toml string' in GUI expert page # Self-referential toml parameter, for setting any other toml parameters as string of tomls separated by # (spaces around # are ok). # Useful when toml parameter is not in expert mode but want per-experiment control. # Setting this will override all other choices. # In expert page, each time expert options saved, the new state is set without memory of any prior settings. # The entered item is a fully compliant toml string that would be processed directly by toml.load(). # One should include 2 double quotes around the entire setting, or double quotes need to be escaped. # One enters into the expert page text as follows: # e.g. enable_glm="off" # enable_xgboost_gbm="off" # enable_lightgbm="on" # e.g. ""enable_glm="off" # enable_xgboost_gbm="off" # enable_lightgbm="off" # enable_tensorflow="on""" # e.g. fixed_num_individuals=4 # e.g. params_lightgbm="{'objective':'poisson'}" # e.g. ""params_lightgbm="{'objective':'poisson'}""" # e.g. max_cores=10 # data_precision="float32" # max_rows_feature_evolution=50000000000 # ensemble_accuracy_switch=11 # feature_engineering_effort=1 # target_transformer="identity" # tournament_feature_style_accuracy_switch=5 # params_tensorflow="{'layers': [100, 100, 100, 100, 100, 100]}" # e.g. ""max_cores=10 # data_precision="float32" # max_rows_feature_evolution=50000000000 # ensemble_accuracy_switch=11 # feature_engineering_effort=1 # target_transformer="identity" # tournament_feature_style_accuracy_switch=5 # params_tensorflow="{'layers': [100, 100, 100, 100, 100, 100]}""" # If you see: "toml.TomlDecodeError" then ensure toml is set correctly. # When set in the expert page of an experiment, these changes only affect experiments and not the server # Usually should keep this as empty string in this toml file. #config_overrides = "" # Whether user can download dataset as csv file #enable_dataset_downloading = true # Extra HTTP headers. #extra_http_headers = "{}" # After how many days the audit log records are removed. # Set equal to 0 to disable removal of old records. # #audit_log_retention_period = 5 # Replace all the downloads on the experiment page to exports and allow users to push to the artifact store configured with artifacts_store #enable_artifacts_upload = false # Artifacts store. # file_system: stores artifacts on a file system directory denoted by artifacts_file_system_directory. # #artifacts_store = "file_system" # File system location where artifacts will be copied in case artifacts_store is set to file_system #artifacts_file_system_directory = "tmp" #num_models_for_resume_graph = 1000