From 338b82519f041fbd5a42dba701e533e44dbc690f Mon Sep 17 00:00:00 2001 From: Quarto GHA Workflow Runner Date: Wed, 13 Aug 2025 10:45:04 +0000 Subject: [PATCH] Built site for gh-pages --- .github/workflows/main.yml | 18 + .nojekyll | 2 +- TODO.html | 893 -------------------------------- docs/api/cli.utils.train.html | 35 +- docs/api/integrations.base.html | 12 +- search.json | 22 +- sitemap.xml | 398 +++++++------- 7 files changed, 255 insertions(+), 1125 deletions(-) delete mode 100644 TODO.html diff --git a/.github/workflows/main.yml b/.github/workflows/main.yml index 891300246..3daf39e43 100644 --- a/.github/workflows/main.yml +++ b/.github/workflows/main.yml @@ -98,6 +98,12 @@ jobs: python_version: "3.11" pytorch: 2.7.1 axolotl_extras: + is_latest: + - cuda: 126 + cuda_version: 12.6.3 + python_version: "3.11" + pytorch: 2.7.1 + axolotl_extras: vllm is_latest: true - cuda: 128 cuda_version: 12.8.1 @@ -151,6 +157,18 @@ jobs: python_version: "3.11" pytorch: 2.6.0 axolotl_extras: + - cuda: 126 + cuda_version: 12.6.3 + python_version: "3.11" + pytorch: 2.7.1 + axolotl_extras: + is_latest: + - cuda: 126 + cuda_version: 12.6.3 + python_version: "3.11" + pytorch: 2.7.1 + axolotl_extras: vllm + is_latest: true runs-on: axolotl-gpu-runner steps: - name: Checkout diff --git a/.nojekyll b/.nojekyll index 894962a8b..9af084580 100644 --- a/.nojekyll +++ b/.nojekyll @@ -1 +1 @@ -f74eb4cd \ No newline at end of file +a893a00c \ No newline at end of file diff --git a/TODO.html b/TODO.html deleted file mode 100644 index e0239ec97..000000000 --- a/TODO.html +++ /dev/null @@ -1,893 +0,0 @@ - - - - - - - - - -todo – Axolotl - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
-
- - -
- -
- - -
- - - -
- - - - -
-

todo list

-
    -
  • [] Validation of parameters for combinations that won’t work
    • -
-
-

things that are known not to work

-
    -
  • FSDP offload and gradient_checkpointing - https://github.com/pytorch/pytorch/issues/82203
    • -
  • adamw_bnb_8bit doesn’t play well with FSDP offload
    • -
- - -
-
- -
- -
- - - - - \ No newline at end of file diff --git a/docs/api/cli.utils.train.html b/docs/api/cli.utils.train.html index 6ad56b4bc..3909f4688 100644 --- a/docs/api/cli.utils.train.html +++ b/docs/api/cli.utils.train.html @@ -529,7 +529,7 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true}); generate_config_files -Generate list of configuration files to process. +Generate list of configuration files to process. Yields a tuple of the configuration file name and a boolean indicating launch_training @@ -597,7 +597,35 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true});

generate_config_files

cli.utils.train.generate_config_files(config, sweep)
-

Generate list of configuration files to process.

+

Generate list of configuration files to process. Yields a tuple of the configuration file name and a boolean indicating +whether this is a group of configurations (i.e., a sweep).

+
+

Parameters

+ + + + + + + + + + + + + + + + + + + + + + + +
NameTypeDescriptionDefault
configstrBase configuration filerequired
sweepstr | NoneSweep configuration filerequired
+

launch_training

@@ -607,7 +635,8 @@ gtag('config', 'G-9KYCVJBNMQ', { 'anonymize_ip': true}); cloud, kwargs, launcher_args=None, -) + use_exec=False, +)

Execute training with the given configuration.

diff --git a/docs/api/integrations.base.html b/docs/api/integrations.base.html index 39c1cb84a..084b71338 100644 --- a/docs/api/integrations.base.html +++ b/docs/api/integrations.base.html @@ -689,7 +689,7 @@ training.

register -Registers the plugin with the given configuration. +Registers the plugin with the given configuration as an unparsed dict. @@ -1469,16 +1469,10 @@ callbacks that require access to the model or trainer.

register
integrations.base.BasePlugin.register(cfg)
-

Registers the plugin with the given configuration.

+

Registers the plugin with the given configuration as an unparsed dict.

Parameters
------ @@ -1490,7 +1484,7 @@ callbacks that require access to the model or trainer.

- + diff --git a/search.json b/search.json index 69b4682d4..9e9fe9ded 100644 --- a/search.json +++ b/search.json @@ -1692,14 +1692,14 @@ "href": "docs/api/cli.utils.train.html", "title": "cli.utils.train", "section": "", - "text": "cli.utils.train\nUtilities for axolotl train CLI command.\n\n\n\n\n\nName\nDescription\n\n\n\n\nbuild_command\nBuild command list from base command and options.\n\n\ngenerate_config_files\nGenerate list of configuration files to process.\n\n\nlaunch_training\nExecute training with the given configuration.\n\n\n\n\n\ncli.utils.train.build_command(base_cmd, options)\nBuild command list from base command and options.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\nbase_cmd\nlist[str]\nCommand without options.\nrequired\n\n\noptions\ndict[str, Any]\nOptions to parse and append to base command.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[str]\nList of strings giving shell command.\n\n\n\n\n\n\n\ncli.utils.train.generate_config_files(config, sweep)\nGenerate list of configuration files to process.\n\n\n\ncli.utils.train.launch_training(\n cfg_file,\n launcher,\n cloud,\n kwargs,\n launcher_args=None,\n)\nExecute training with the given configuration." + "text": "cli.utils.train\nUtilities for axolotl train CLI command.\n\n\n\n\n\nName\nDescription\n\n\n\n\nbuild_command\nBuild command list from base command and options.\n\n\ngenerate_config_files\nGenerate list of configuration files to process. Yields a tuple of the configuration file name and a boolean indicating\n\n\nlaunch_training\nExecute training with the given configuration.\n\n\n\n\n\ncli.utils.train.build_command(base_cmd, options)\nBuild command list from base command and options.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\nbase_cmd\nlist[str]\nCommand without options.\nrequired\n\n\noptions\ndict[str, Any]\nOptions to parse and append to base command.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[str]\nList of strings giving shell command.\n\n\n\n\n\n\n\ncli.utils.train.generate_config_files(config, sweep)\nGenerate list of configuration files to process. Yields a tuple of the configuration file name and a boolean indicating\nwhether this is a group of configurations (i.e., a sweep).\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\nconfig\nstr\nBase configuration file\nrequired\n\n\nsweep\nstr | None\nSweep configuration file\nrequired\n\n\n\n\n\n\n\ncli.utils.train.launch_training(\n cfg_file,\n launcher,\n cloud,\n kwargs,\n launcher_args=None,\n use_exec=False,\n)\nExecute training with the given configuration." }, { "objectID": "docs/api/cli.utils.train.html#functions", "href": "docs/api/cli.utils.train.html#functions", "title": "cli.utils.train", "section": "", - "text": "Name\nDescription\n\n\n\n\nbuild_command\nBuild command list from base command and options.\n\n\ngenerate_config_files\nGenerate list of configuration files to process.\n\n\nlaunch_training\nExecute training with the given configuration.\n\n\n\n\n\ncli.utils.train.build_command(base_cmd, options)\nBuild command list from base command and options.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\nbase_cmd\nlist[str]\nCommand without options.\nrequired\n\n\noptions\ndict[str, Any]\nOptions to parse and append to base command.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[str]\nList of strings giving shell command.\n\n\n\n\n\n\n\ncli.utils.train.generate_config_files(config, sweep)\nGenerate list of configuration files to process.\n\n\n\ncli.utils.train.launch_training(\n cfg_file,\n launcher,\n cloud,\n kwargs,\n launcher_args=None,\n)\nExecute training with the given configuration." + "text": "Name\nDescription\n\n\n\n\nbuild_command\nBuild command list from base command and options.\n\n\ngenerate_config_files\nGenerate list of configuration files to process. Yields a tuple of the configuration file name and a boolean indicating\n\n\nlaunch_training\nExecute training with the given configuration.\n\n\n\n\n\ncli.utils.train.build_command(base_cmd, options)\nBuild command list from base command and options.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\nbase_cmd\nlist[str]\nCommand without options.\nrequired\n\n\noptions\ndict[str, Any]\nOptions to parse and append to base command.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[str]\nList of strings giving shell command.\n\n\n\n\n\n\n\ncli.utils.train.generate_config_files(config, sweep)\nGenerate list of configuration files to process. Yields a tuple of the configuration file name and a boolean indicating\nwhether this is a group of configurations (i.e., a sweep).\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\nconfig\nstr\nBase configuration file\nrequired\n\n\nsweep\nstr | None\nSweep configuration file\nrequired\n\n\n\n\n\n\n\ncli.utils.train.launch_training(\n cfg_file,\n launcher,\n cloud,\n kwargs,\n launcher_args=None,\n use_exec=False,\n)\nExecute training with the given configuration." }, { "objectID": "docs/api/monkeypatch.llama_attn_hijack_xformers.html", @@ -2318,20 +2318,6 @@ "section": "Demo: Talk Like a Pirate", "text": "Demo: Talk Like a Pirate\nIn this demo, we are training the model to respond like a pirate. This was chosen as a way to easily show how to train a model to respond in a certain style of your choosing (without being prompted) and is quite easy to validate within the scope of a Colab.\n\nUpload your own dataset or use a Huggingface dataset\nYou can choose to use your own JSONL file from your own Google Drive; for example downloading the Pirate-Ultrachat JSONL to your Google Drive. JSONL datasets should be formatted similar to the OpenAI dataset format.\nYou can also simply use the winglian/pirate-ultrachat-10k dataset directly.\n\n# Default to HF dataset location\ndataset_id = \"winglian/pirate-ultrachat-10k\"\nuploaded = {}\n\n\nimport os\n# Optionally, upload your own JSONL to your Google Drive\nGOOGLE_DRIVE_PATH = \"\" # ex: \"MyDrive/Colab\\ Notebooks/train.jsonl\"\n\n# \"Select All\" permissions, or you may get the error:\n# \"MessageError: Error: credential propagation was unsuccessful\"\nif GOOGLE_DRIVE_PATH:\n from google.colab import drive\n # Mount your Google Drive\n GOOGLE_DRIVE_MNT = \"/content/drive/\"\n drive.mount(GOOGLE_DRIVE_MNT, force_remount=True)\n tmp_path = os.path.join(GOOGLE_DRIVE_MNT, GOOGLE_DRIVE_PATH.lstrip(\"/\"))\n # make sure file exists\n if not os.path.isfile(tmp_path):\n raise ValueError(f\"File {tmp_path} does not exist\")\n dataset_id = tmp_path" }, - { - "objectID": "TODO.html", - "href": "TODO.html", - "title": "todo list", - "section": "", - "text": "[] Validation of parameters for combinations that won’t work\n\n\n\n\nFSDP offload and gradient_checkpointing - https://github.com/pytorch/pytorch/issues/82203\nadamw_bnb_8bit doesn’t play well with FSDP offload" - }, - { - "objectID": "TODO.html#things-that-are-known-not-to-work", - "href": "TODO.html#things-that-are-known-not-to-work", - "title": "todo list", - "section": "", - "text": "FSDP offload and gradient_checkpointing - https://github.com/pytorch/pytorch/issues/82203\nadamw_bnb_8bit doesn’t play well with FSDP offload" - }, { "objectID": "FAQS.html", "href": "FAQS.html", @@ -2681,14 +2667,14 @@ "href": "docs/api/integrations.base.html", "title": "integrations.base", "section": "", - "text": "integrations.base\nBase class for all plugins.\nA plugin is a reusable, modular, and self-contained piece of code that extends the functionality of Axolotl.\nPlugins can be used to integrate third-party models, modify the training process, or add new features.\nTo create a new plugin, you need to inherit from the BasePlugin class and implement the required methods.\n\n\n\n\n\nName\nDescription\n\n\n\n\nBaseOptimizerFactory\nBase class for factories to create custom optimizers\n\n\nBasePlugin\nBase class for all plugins. Defines the interface for plugin methods.\n\n\nPluginManager\nThe PluginManager class is responsible for loading and managing plugins. It\n\n\n\n\n\nintegrations.base.BaseOptimizerFactory()\nBase class for factories to create custom optimizers\n\n\n\n\n\nName\nDescription\n\n\n\n\nget_decay_parameter_names\nGet all parameter names that weight decay will be applied to.\n\n\n\n\n\nintegrations.base.BaseOptimizerFactory.get_decay_parameter_names(model)\nGet all parameter names that weight decay will be applied to.\nThis function filters out parameters in two ways:\n1. By layer type (instances of layers specified in ALL_LAYERNORM_LAYERS)\n2. By parameter name patterns (containing ‘bias’, or variation of ‘norm’)\n\n\n\n\n\nintegrations.base.BasePlugin()\nBase class for all plugins. Defines the interface for plugin methods.\nA plugin is a reusable, modular, and self-contained piece of code that extends\nthe functionality of Axolotl. Plugins can be used to integrate third-party models,\nmodify the training process, or add new features.\nTo create a new plugin, you need to inherit from the BasePlugin class and\nimplement the required methods.\n\n\nPlugin methods include:\n- register(cfg): Registers the plugin with the given configuration.\n- load_datasets(cfg): Loads and preprocesses the dataset for training.\n- pre_model_load(cfg): Performs actions before the model is loaded.\n- post_model_build(cfg, model): Performs actions after the model is loaded, but\nbefore LoRA adapters are applied.\n- pre_lora_load(cfg, model): Performs actions before LoRA weights are loaded.\n- post_lora_load(cfg, model): Performs actions after LoRA weights are loaded.\n- post_model_load(cfg, model): Performs actions after the model is loaded,\ninclusive of any adapters.\n- post_trainer_create(cfg, trainer): Performs actions after the trainer is\ncreated.\n- create_optimizer(cfg, trainer): Creates and returns an optimizer for training.\n- create_lr_scheduler(cfg, trainer, optimizer, num_training_steps): Creates and\nreturns a learning rate scheduler.\n- add_callbacks_pre_trainer(cfg, model): Adds callbacks to the trainer before\ntraining.\n- add_callbacks_post_trainer(cfg, trainer): Adds callbacks to the trainer after\ntraining.\n\n\n\n\n\n\nName\nDescription\n\n\n\n\nadd_callbacks_post_trainer\nAdds callbacks to the trainer after creating the trainer. This is useful for\n\n\nadd_callbacks_pre_trainer\nSet up callbacks before creating the trainer.\n\n\ncreate_lr_scheduler\nCreates and returns a learning rate scheduler.\n\n\ncreate_optimizer\nCreates and returns an optimizer for training.\n\n\nget_collator_cls_and_kwargs\nReturns a custom class for the collator.\n\n\nget_input_args\nReturns a pydantic model for the plugin’s input arguments.\n\n\nget_trainer_cls\nReturns a custom class for the trainer.\n\n\nget_training_args\nReturns custom training arguments to set on TrainingArgs.\n\n\nget_training_args_mixin\nReturns a dataclass model for the plugin’s training arguments.\n\n\nload_datasets\nLoads and preprocesses the dataset for training.\n\n\npost_lora_load\nPerforms actions after LoRA weights are loaded.\n\n\npost_model_build\nPerforms actions after the model is built/loaded, but before any adapters are applied.\n\n\npost_model_load\nPerforms actions after the model is loaded.\n\n\npost_train\nPerforms actions after training is complete.\n\n\npost_train_unload\nPerforms actions after training is complete and the model is unloaded.\n\n\npost_trainer_create\nPerforms actions after the trainer is created.\n\n\npre_lora_load\nPerforms actions before LoRA weights are loaded.\n\n\npre_model_load\nPerforms actions before the model is loaded.\n\n\nregister\nRegisters the plugin with the given configuration.\n\n\n\n\n\nintegrations.base.BasePlugin.add_callbacks_post_trainer(cfg, trainer)\nAdds callbacks to the trainer after creating the trainer. This is useful for\ncallbacks that require access to the model or trainer.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[Callable]\nA list of callback functions to be added\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.add_callbacks_pre_trainer(cfg, model)\nSet up callbacks before creating the trainer.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[Callable]\nA list of callback functions to be added to the TrainingArgs.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.create_lr_scheduler(\n cfg,\n trainer,\n optimizer,\n num_training_steps,\n)\nCreates and returns a learning rate scheduler.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\noptimizer\nOptimizer\nThe optimizer for training.\nrequired\n\n\nnum_training_steps\nint\nTotal number of training steps\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nLRScheduler | None\nThe created learning rate scheduler.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.create_optimizer(cfg, trainer)\nCreates and returns an optimizer for training.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nOptimizer | None\nThe created optimizer.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.get_collator_cls_and_kwargs(cfg, is_eval=False)\nReturns a custom class for the collator.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe global axolotl configuration.\nrequired\n\n\nis_eval\nbool\nWhether this is an eval split.\nFalse\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\nclass\n\nThe class for the collator.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.get_input_args()\nReturns a pydantic model for the plugin’s input arguments.\n\n\n\nintegrations.base.BasePlugin.get_trainer_cls(cfg)\nReturns a custom class for the trainer.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe global axolotl configuration.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nTrainer | None\nThe first non-None trainer class returned by a plugin.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.get_training_args(cfg)\nReturns custom training arguments to set on TrainingArgs.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe global axolotl configuration.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\nobject\n\ndict containing the training arguments.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.get_training_args_mixin()\nReturns a dataclass model for the plugin’s training arguments.\n\n\n\nintegrations.base.BasePlugin.load_datasets(cfg, preprocess=False)\nLoads and preprocesses the dataset for training.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\npreprocess\nbool\nWhether this is the preprocess step of the datasets.\nFalse\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\ndataset_meta\nUnion['TrainDatasetMeta', None]\nThe metadata for the training dataset.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_lora_load(cfg, model)\nPerforms actions after LoRA weights are loaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_model_build(cfg, model)\nPerforms actions after the model is built/loaded, but before any adapters are applied.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_model_load(cfg, model)\nPerforms actions after the model is loaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_train(cfg, model)\nPerforms actions after training is complete.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe axolotl configuration.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_train_unload(cfg)\nPerforms actions after training is complete and the model is unloaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_trainer_create(cfg, trainer)\nPerforms actions after the trainer is created.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.pre_lora_load(cfg, model)\nPerforms actions before LoRA weights are loaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.pre_model_load(cfg)\nPerforms actions before the model is loaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.register(cfg)\nRegisters the plugin with the given configuration.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\n\n\n\n\n\n\n\nintegrations.base.PluginManager()\nThe PluginManager class is responsible for loading and managing plugins. It\nshould be a singleton so it can be accessed from anywhere in the codebase.\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\nplugins\nOrderedDict[str, BasePlugin]\nA list of loaded plugins.\n\n\n\n\n\n\nKey methods include:\n- get_instance(): Static method to get the singleton instance of PluginManager.\n- register(plugin_name: str): Registers a new plugin by its name.\n- pre_model_load(cfg): Calls the pre_model_load method of all registered plugins.\n\n\n\n\n\n\nName\nDescription\n\n\n\n\nadd_callbacks_post_trainer\nCalls the add_callbacks_post_trainer method of all registered plugins.\n\n\nadd_callbacks_pre_trainer\nCalls the add_callbacks_pre_trainer method of all registered plugins.\n\n\ncreate_lr_scheduler\nCalls the create_lr_scheduler method of all registered plugins and returns\n\n\ncreate_optimizer\nCalls the create_optimizer method of all registered plugins and returns\n\n\nget_collator_cls_and_kwargs\nCalls the get_collator_cls_and_kwargs method of all registered plugins and returns the first non-None collator class.\n\n\nget_input_args\nReturns a list of Pydantic classes for all registered plugins’ input arguments.’\n\n\nget_instance\nReturns the singleton instance of PluginManager. If the instance doesn’t\n\n\nget_trainer_cls\nCalls the get_trainer_cls method of all registered plugins and returns the\n\n\nget_training_args\nCalls the get_training_args method of all registered plugins and returns the combined training arguments.\n\n\nget_training_args_mixin\nReturns a list of dataclasses for all registered plugins’ training args mixins’\n\n\nload_datasets\nCalls the load_datasets method of each registered plugin.\n\n\npost_lora_load\nCalls the post_lora_load method of all registered plugins.\n\n\npost_model_build\nCalls the post_model_build method of all registered plugins after the\n\n\npost_model_load\nCalls the post_model_load method of all registered plugins after the model\n\n\npost_train\nCalls the post_train method of all registered plugins.\n\n\npost_train_unload\nCalls the post_train_unload method of all registered plugins.\n\n\npost_trainer_create\nCalls the post_trainer_create method of all registered plugins.\n\n\npre_lora_load\nCalls the pre_lora_load method of all registered plugins.\n\n\npre_model_load\nCalls the pre_model_load method of all registered plugins.\n\n\nregister\nRegisters a new plugin by its name.\n\n\n\n\n\nintegrations.base.PluginManager.add_callbacks_post_trainer(cfg, trainer)\nCalls the add_callbacks_post_trainer method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[Callable]\nA list of callback functions to be added to the TrainingArgs.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.add_callbacks_pre_trainer(cfg, model)\nCalls the add_callbacks_pre_trainer method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[Callable]\nA list of callback functions to be added to the TrainingArgs.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.create_lr_scheduler(\n trainer,\n optimizer,\n num_training_steps,\n)\nCalls the create_lr_scheduler method of all registered plugins and returns\nthe first non-None scheduler.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\noptimizer\nOptimizer\nThe optimizer for training.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nLRScheduler | None\nThe created learning rate scheduler, or None if not found.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.create_optimizer(trainer)\nCalls the create_optimizer method of all registered plugins and returns\nthe first non-None optimizer.\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nOptimizer | None\nThe created optimizer, or None if none was found.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.get_collator_cls_and_kwargs(cfg, is_eval=False)\nCalls the get_collator_cls_and_kwargs method of all registered plugins and returns the first non-None collator class.\nParameters:\ncfg (dict): The configuration for the plugins.\nis_eval (bool): Whether this is an eval split.\nReturns:\nobject: The collator class, or None if none was found.\n\n\n\nintegrations.base.PluginManager.get_input_args()\nReturns a list of Pydantic classes for all registered plugins’ input arguments.’\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[str]\nA list of Pydantic classes for all registered plugins’ input arguments.’\n\n\n\n\n\n\n\nintegrations.base.PluginManager.get_instance()\nReturns the singleton instance of PluginManager. If the instance doesn’t\nexist, it creates a new one.\n\n\n\nintegrations.base.PluginManager.get_trainer_cls(cfg)\nCalls the get_trainer_cls method of all registered plugins and returns the\nfirst non-None trainer class.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nTrainer | None\nThe first non-None trainer class returned by a plugin.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.get_training_args(cfg)\nCalls the get_training_args method of all registered plugins and returns the combined training arguments.\nParameters:\ncfg (dict): The configuration for the plugins.\nReturns:\nobject: The training arguments\n\n\n\nintegrations.base.PluginManager.get_training_args_mixin()\nReturns a list of dataclasses for all registered plugins’ training args mixins’\nReturns:\nlist[str]: A list of dataclsses\n\n\n\nintegrations.base.PluginManager.load_datasets(cfg, preprocess=False)\nCalls the load_datasets method of each registered plugin.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\npreprocess\nbool\nWhether this is preprocess step of the datasets.\nFalse\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nUnion['TrainDatasetMeta', None]\nThe dataset metadata loaded from all registered plugins.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_lora_load(cfg, model)\nCalls the post_lora_load method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_model_build(cfg, model)\nCalls the post_model_build method of all registered plugins after the\nmodel has been built / loaded, but before any adapters have been applied.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_model_load(cfg, model)\nCalls the post_model_load method of all registered plugins after the model\nhas been loaded inclusive of any adapters.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_train(cfg, model)\nCalls the post_train method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_train_unload(cfg)\nCalls the post_train_unload method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_trainer_create(cfg, trainer)\nCalls the post_trainer_create method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.pre_lora_load(cfg, model)\nCalls the pre_lora_load method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.pre_model_load(cfg)\nCalls the pre_model_load method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.register(plugin_name)\nRegisters a new plugin by its name.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\nplugin_name\nstr\nThe name of the plugin to be registered.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nImportError\nIf the plugin module cannot be imported.\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nDescription\n\n\n\n\nload_plugin\nLoads a plugin based on the given plugin name.\n\n\n\n\n\nintegrations.base.load_plugin(plugin_name)\nLoads a plugin based on the given plugin name.\nThe plugin name should be in the format “module_name.class_name”. This function\nsplits the plugin name into module and class, imports the module, retrieves the\nclass from the module, and creates an instance of the class.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\nplugin_name\nstr\nThe name of the plugin to be loaded. The name should be in the format “module_name.class_name”.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nBasePlugin\nAn instance of the loaded plugin.\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nImportError\nIf the plugin module cannot be imported." + "text": "integrations.base\nBase class for all plugins.\nA plugin is a reusable, modular, and self-contained piece of code that extends the functionality of Axolotl.\nPlugins can be used to integrate third-party models, modify the training process, or add new features.\nTo create a new plugin, you need to inherit from the BasePlugin class and implement the required methods.\n\n\n\n\n\nName\nDescription\n\n\n\n\nBaseOptimizerFactory\nBase class for factories to create custom optimizers\n\n\nBasePlugin\nBase class for all plugins. Defines the interface for plugin methods.\n\n\nPluginManager\nThe PluginManager class is responsible for loading and managing plugins. It\n\n\n\n\n\nintegrations.base.BaseOptimizerFactory()\nBase class for factories to create custom optimizers\n\n\n\n\n\nName\nDescription\n\n\n\n\nget_decay_parameter_names\nGet all parameter names that weight decay will be applied to.\n\n\n\n\n\nintegrations.base.BaseOptimizerFactory.get_decay_parameter_names(model)\nGet all parameter names that weight decay will be applied to.\nThis function filters out parameters in two ways:\n1. By layer type (instances of layers specified in ALL_LAYERNORM_LAYERS)\n2. By parameter name patterns (containing ‘bias’, or variation of ‘norm’)\n\n\n\n\n\nintegrations.base.BasePlugin()\nBase class for all plugins. Defines the interface for plugin methods.\nA plugin is a reusable, modular, and self-contained piece of code that extends\nthe functionality of Axolotl. Plugins can be used to integrate third-party models,\nmodify the training process, or add new features.\nTo create a new plugin, you need to inherit from the BasePlugin class and\nimplement the required methods.\n\n\nPlugin methods include:\n- register(cfg): Registers the plugin with the given configuration.\n- load_datasets(cfg): Loads and preprocesses the dataset for training.\n- pre_model_load(cfg): Performs actions before the model is loaded.\n- post_model_build(cfg, model): Performs actions after the model is loaded, but\nbefore LoRA adapters are applied.\n- pre_lora_load(cfg, model): Performs actions before LoRA weights are loaded.\n- post_lora_load(cfg, model): Performs actions after LoRA weights are loaded.\n- post_model_load(cfg, model): Performs actions after the model is loaded,\ninclusive of any adapters.\n- post_trainer_create(cfg, trainer): Performs actions after the trainer is\ncreated.\n- create_optimizer(cfg, trainer): Creates and returns an optimizer for training.\n- create_lr_scheduler(cfg, trainer, optimizer, num_training_steps): Creates and\nreturns a learning rate scheduler.\n- add_callbacks_pre_trainer(cfg, model): Adds callbacks to the trainer before\ntraining.\n- add_callbacks_post_trainer(cfg, trainer): Adds callbacks to the trainer after\ntraining.\n\n\n\n\n\n\nName\nDescription\n\n\n\n\nadd_callbacks_post_trainer\nAdds callbacks to the trainer after creating the trainer. This is useful for\n\n\nadd_callbacks_pre_trainer\nSet up callbacks before creating the trainer.\n\n\ncreate_lr_scheduler\nCreates and returns a learning rate scheduler.\n\n\ncreate_optimizer\nCreates and returns an optimizer for training.\n\n\nget_collator_cls_and_kwargs\nReturns a custom class for the collator.\n\n\nget_input_args\nReturns a pydantic model for the plugin’s input arguments.\n\n\nget_trainer_cls\nReturns a custom class for the trainer.\n\n\nget_training_args\nReturns custom training arguments to set on TrainingArgs.\n\n\nget_training_args_mixin\nReturns a dataclass model for the plugin’s training arguments.\n\n\nload_datasets\nLoads and preprocesses the dataset for training.\n\n\npost_lora_load\nPerforms actions after LoRA weights are loaded.\n\n\npost_model_build\nPerforms actions after the model is built/loaded, but before any adapters are applied.\n\n\npost_model_load\nPerforms actions after the model is loaded.\n\n\npost_train\nPerforms actions after training is complete.\n\n\npost_train_unload\nPerforms actions after training is complete and the model is unloaded.\n\n\npost_trainer_create\nPerforms actions after the trainer is created.\n\n\npre_lora_load\nPerforms actions before LoRA weights are loaded.\n\n\npre_model_load\nPerforms actions before the model is loaded.\n\n\nregister\nRegisters the plugin with the given configuration as an unparsed dict.\n\n\n\n\n\nintegrations.base.BasePlugin.add_callbacks_post_trainer(cfg, trainer)\nAdds callbacks to the trainer after creating the trainer. This is useful for\ncallbacks that require access to the model or trainer.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[Callable]\nA list of callback functions to be added\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.add_callbacks_pre_trainer(cfg, model)\nSet up callbacks before creating the trainer.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[Callable]\nA list of callback functions to be added to the TrainingArgs.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.create_lr_scheduler(\n cfg,\n trainer,\n optimizer,\n num_training_steps,\n)\nCreates and returns a learning rate scheduler.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\noptimizer\nOptimizer\nThe optimizer for training.\nrequired\n\n\nnum_training_steps\nint\nTotal number of training steps\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nLRScheduler | None\nThe created learning rate scheduler.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.create_optimizer(cfg, trainer)\nCreates and returns an optimizer for training.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nOptimizer | None\nThe created optimizer.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.get_collator_cls_and_kwargs(cfg, is_eval=False)\nReturns a custom class for the collator.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe global axolotl configuration.\nrequired\n\n\nis_eval\nbool\nWhether this is an eval split.\nFalse\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\nclass\n\nThe class for the collator.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.get_input_args()\nReturns a pydantic model for the plugin’s input arguments.\n\n\n\nintegrations.base.BasePlugin.get_trainer_cls(cfg)\nReturns a custom class for the trainer.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe global axolotl configuration.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nTrainer | None\nThe first non-None trainer class returned by a plugin.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.get_training_args(cfg)\nReturns custom training arguments to set on TrainingArgs.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe global axolotl configuration.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\nobject\n\ndict containing the training arguments.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.get_training_args_mixin()\nReturns a dataclass model for the plugin’s training arguments.\n\n\n\nintegrations.base.BasePlugin.load_datasets(cfg, preprocess=False)\nLoads and preprocesses the dataset for training.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\npreprocess\nbool\nWhether this is the preprocess step of the datasets.\nFalse\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\ndataset_meta\nUnion['TrainDatasetMeta', None]\nThe metadata for the training dataset.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_lora_load(cfg, model)\nPerforms actions after LoRA weights are loaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_model_build(cfg, model)\nPerforms actions after the model is built/loaded, but before any adapters are applied.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_model_load(cfg, model)\nPerforms actions after the model is loaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_train(cfg, model)\nPerforms actions after training is complete.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe axolotl configuration.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_train_unload(cfg)\nPerforms actions after training is complete and the model is unloaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_trainer_create(cfg, trainer)\nPerforms actions after the trainer is created.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.pre_lora_load(cfg, model)\nPerforms actions before LoRA weights are loaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.pre_model_load(cfg)\nPerforms actions before the model is loaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.register(cfg)\nRegisters the plugin with the given configuration as an unparsed dict.\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\ndict\nThe configuration for the plugin.\nrequired\n\n\n\n\n\n\n\n\n\nintegrations.base.PluginManager()\nThe PluginManager class is responsible for loading and managing plugins. It\nshould be a singleton so it can be accessed from anywhere in the codebase.\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\nplugins\nOrderedDict[str, BasePlugin]\nA list of loaded plugins.\n\n\n\n\n\n\nKey methods include:\n- get_instance(): Static method to get the singleton instance of PluginManager.\n- register(plugin_name: str): Registers a new plugin by its name.\n- pre_model_load(cfg): Calls the pre_model_load method of all registered plugins.\n\n\n\n\n\n\nName\nDescription\n\n\n\n\nadd_callbacks_post_trainer\nCalls the add_callbacks_post_trainer method of all registered plugins.\n\n\nadd_callbacks_pre_trainer\nCalls the add_callbacks_pre_trainer method of all registered plugins.\n\n\ncreate_lr_scheduler\nCalls the create_lr_scheduler method of all registered plugins and returns\n\n\ncreate_optimizer\nCalls the create_optimizer method of all registered plugins and returns\n\n\nget_collator_cls_and_kwargs\nCalls the get_collator_cls_and_kwargs method of all registered plugins and returns the first non-None collator class.\n\n\nget_input_args\nReturns a list of Pydantic classes for all registered plugins’ input arguments.’\n\n\nget_instance\nReturns the singleton instance of PluginManager. If the instance doesn’t\n\n\nget_trainer_cls\nCalls the get_trainer_cls method of all registered plugins and returns the\n\n\nget_training_args\nCalls the get_training_args method of all registered plugins and returns the combined training arguments.\n\n\nget_training_args_mixin\nReturns a list of dataclasses for all registered plugins’ training args mixins’\n\n\nload_datasets\nCalls the load_datasets method of each registered plugin.\n\n\npost_lora_load\nCalls the post_lora_load method of all registered plugins.\n\n\npost_model_build\nCalls the post_model_build method of all registered plugins after the\n\n\npost_model_load\nCalls the post_model_load method of all registered plugins after the model\n\n\npost_train\nCalls the post_train method of all registered plugins.\n\n\npost_train_unload\nCalls the post_train_unload method of all registered plugins.\n\n\npost_trainer_create\nCalls the post_trainer_create method of all registered plugins.\n\n\npre_lora_load\nCalls the pre_lora_load method of all registered plugins.\n\n\npre_model_load\nCalls the pre_model_load method of all registered plugins.\n\n\nregister\nRegisters a new plugin by its name.\n\n\n\n\n\nintegrations.base.PluginManager.add_callbacks_post_trainer(cfg, trainer)\nCalls the add_callbacks_post_trainer method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[Callable]\nA list of callback functions to be added to the TrainingArgs.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.add_callbacks_pre_trainer(cfg, model)\nCalls the add_callbacks_pre_trainer method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[Callable]\nA list of callback functions to be added to the TrainingArgs.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.create_lr_scheduler(\n trainer,\n optimizer,\n num_training_steps,\n)\nCalls the create_lr_scheduler method of all registered plugins and returns\nthe first non-None scheduler.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\noptimizer\nOptimizer\nThe optimizer for training.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nLRScheduler | None\nThe created learning rate scheduler, or None if not found.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.create_optimizer(trainer)\nCalls the create_optimizer method of all registered plugins and returns\nthe first non-None optimizer.\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nOptimizer | None\nThe created optimizer, or None if none was found.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.get_collator_cls_and_kwargs(cfg, is_eval=False)\nCalls the get_collator_cls_and_kwargs method of all registered plugins and returns the first non-None collator class.\nParameters:\ncfg (dict): The configuration for the plugins.\nis_eval (bool): Whether this is an eval split.\nReturns:\nobject: The collator class, or None if none was found.\n\n\n\nintegrations.base.PluginManager.get_input_args()\nReturns a list of Pydantic classes for all registered plugins’ input arguments.’\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[str]\nA list of Pydantic classes for all registered plugins’ input arguments.’\n\n\n\n\n\n\n\nintegrations.base.PluginManager.get_instance()\nReturns the singleton instance of PluginManager. If the instance doesn’t\nexist, it creates a new one.\n\n\n\nintegrations.base.PluginManager.get_trainer_cls(cfg)\nCalls the get_trainer_cls method of all registered plugins and returns the\nfirst non-None trainer class.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nTrainer | None\nThe first non-None trainer class returned by a plugin.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.get_training_args(cfg)\nCalls the get_training_args method of all registered plugins and returns the combined training arguments.\nParameters:\ncfg (dict): The configuration for the plugins.\nReturns:\nobject: The training arguments\n\n\n\nintegrations.base.PluginManager.get_training_args_mixin()\nReturns a list of dataclasses for all registered plugins’ training args mixins’\nReturns:\nlist[str]: A list of dataclsses\n\n\n\nintegrations.base.PluginManager.load_datasets(cfg, preprocess=False)\nCalls the load_datasets method of each registered plugin.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\npreprocess\nbool\nWhether this is preprocess step of the datasets.\nFalse\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nUnion['TrainDatasetMeta', None]\nThe dataset metadata loaded from all registered plugins.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_lora_load(cfg, model)\nCalls the post_lora_load method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_model_build(cfg, model)\nCalls the post_model_build method of all registered plugins after the\nmodel has been built / loaded, but before any adapters have been applied.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_model_load(cfg, model)\nCalls the post_model_load method of all registered plugins after the model\nhas been loaded inclusive of any adapters.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_train(cfg, model)\nCalls the post_train method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_train_unload(cfg)\nCalls the post_train_unload method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_trainer_create(cfg, trainer)\nCalls the post_trainer_create method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.pre_lora_load(cfg, model)\nCalls the pre_lora_load method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.pre_model_load(cfg)\nCalls the pre_model_load method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.register(plugin_name)\nRegisters a new plugin by its name.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\nplugin_name\nstr\nThe name of the plugin to be registered.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nImportError\nIf the plugin module cannot be imported.\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nDescription\n\n\n\n\nload_plugin\nLoads a plugin based on the given plugin name.\n\n\n\n\n\nintegrations.base.load_plugin(plugin_name)\nLoads a plugin based on the given plugin name.\nThe plugin name should be in the format “module_name.class_name”. This function\nsplits the plugin name into module and class, imports the module, retrieves the\nclass from the module, and creates an instance of the class.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\nplugin_name\nstr\nThe name of the plugin to be loaded. The name should be in the format “module_name.class_name”.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nBasePlugin\nAn instance of the loaded plugin.\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nImportError\nIf the plugin module cannot be imported." }, { "objectID": "docs/api/integrations.base.html#classes", "href": "docs/api/integrations.base.html#classes", "title": "integrations.base", "section": "", - "text": "Name\nDescription\n\n\n\n\nBaseOptimizerFactory\nBase class for factories to create custom optimizers\n\n\nBasePlugin\nBase class for all plugins. Defines the interface for plugin methods.\n\n\nPluginManager\nThe PluginManager class is responsible for loading and managing plugins. It\n\n\n\n\n\nintegrations.base.BaseOptimizerFactory()\nBase class for factories to create custom optimizers\n\n\n\n\n\nName\nDescription\n\n\n\n\nget_decay_parameter_names\nGet all parameter names that weight decay will be applied to.\n\n\n\n\n\nintegrations.base.BaseOptimizerFactory.get_decay_parameter_names(model)\nGet all parameter names that weight decay will be applied to.\nThis function filters out parameters in two ways:\n1. By layer type (instances of layers specified in ALL_LAYERNORM_LAYERS)\n2. By parameter name patterns (containing ‘bias’, or variation of ‘norm’)\n\n\n\n\n\nintegrations.base.BasePlugin()\nBase class for all plugins. Defines the interface for plugin methods.\nA plugin is a reusable, modular, and self-contained piece of code that extends\nthe functionality of Axolotl. Plugins can be used to integrate third-party models,\nmodify the training process, or add new features.\nTo create a new plugin, you need to inherit from the BasePlugin class and\nimplement the required methods.\n\n\nPlugin methods include:\n- register(cfg): Registers the plugin with the given configuration.\n- load_datasets(cfg): Loads and preprocesses the dataset for training.\n- pre_model_load(cfg): Performs actions before the model is loaded.\n- post_model_build(cfg, model): Performs actions after the model is loaded, but\nbefore LoRA adapters are applied.\n- pre_lora_load(cfg, model): Performs actions before LoRA weights are loaded.\n- post_lora_load(cfg, model): Performs actions after LoRA weights are loaded.\n- post_model_load(cfg, model): Performs actions after the model is loaded,\ninclusive of any adapters.\n- post_trainer_create(cfg, trainer): Performs actions after the trainer is\ncreated.\n- create_optimizer(cfg, trainer): Creates and returns an optimizer for training.\n- create_lr_scheduler(cfg, trainer, optimizer, num_training_steps): Creates and\nreturns a learning rate scheduler.\n- add_callbacks_pre_trainer(cfg, model): Adds callbacks to the trainer before\ntraining.\n- add_callbacks_post_trainer(cfg, trainer): Adds callbacks to the trainer after\ntraining.\n\n\n\n\n\n\nName\nDescription\n\n\n\n\nadd_callbacks_post_trainer\nAdds callbacks to the trainer after creating the trainer. This is useful for\n\n\nadd_callbacks_pre_trainer\nSet up callbacks before creating the trainer.\n\n\ncreate_lr_scheduler\nCreates and returns a learning rate scheduler.\n\n\ncreate_optimizer\nCreates and returns an optimizer for training.\n\n\nget_collator_cls_and_kwargs\nReturns a custom class for the collator.\n\n\nget_input_args\nReturns a pydantic model for the plugin’s input arguments.\n\n\nget_trainer_cls\nReturns a custom class for the trainer.\n\n\nget_training_args\nReturns custom training arguments to set on TrainingArgs.\n\n\nget_training_args_mixin\nReturns a dataclass model for the plugin’s training arguments.\n\n\nload_datasets\nLoads and preprocesses the dataset for training.\n\n\npost_lora_load\nPerforms actions after LoRA weights are loaded.\n\n\npost_model_build\nPerforms actions after the model is built/loaded, but before any adapters are applied.\n\n\npost_model_load\nPerforms actions after the model is loaded.\n\n\npost_train\nPerforms actions after training is complete.\n\n\npost_train_unload\nPerforms actions after training is complete and the model is unloaded.\n\n\npost_trainer_create\nPerforms actions after the trainer is created.\n\n\npre_lora_load\nPerforms actions before LoRA weights are loaded.\n\n\npre_model_load\nPerforms actions before the model is loaded.\n\n\nregister\nRegisters the plugin with the given configuration.\n\n\n\n\n\nintegrations.base.BasePlugin.add_callbacks_post_trainer(cfg, trainer)\nAdds callbacks to the trainer after creating the trainer. This is useful for\ncallbacks that require access to the model or trainer.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[Callable]\nA list of callback functions to be added\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.add_callbacks_pre_trainer(cfg, model)\nSet up callbacks before creating the trainer.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[Callable]\nA list of callback functions to be added to the TrainingArgs.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.create_lr_scheduler(\n cfg,\n trainer,\n optimizer,\n num_training_steps,\n)\nCreates and returns a learning rate scheduler.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\noptimizer\nOptimizer\nThe optimizer for training.\nrequired\n\n\nnum_training_steps\nint\nTotal number of training steps\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nLRScheduler | None\nThe created learning rate scheduler.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.create_optimizer(cfg, trainer)\nCreates and returns an optimizer for training.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nOptimizer | None\nThe created optimizer.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.get_collator_cls_and_kwargs(cfg, is_eval=False)\nReturns a custom class for the collator.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe global axolotl configuration.\nrequired\n\n\nis_eval\nbool\nWhether this is an eval split.\nFalse\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\nclass\n\nThe class for the collator.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.get_input_args()\nReturns a pydantic model for the plugin’s input arguments.\n\n\n\nintegrations.base.BasePlugin.get_trainer_cls(cfg)\nReturns a custom class for the trainer.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe global axolotl configuration.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nTrainer | None\nThe first non-None trainer class returned by a plugin.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.get_training_args(cfg)\nReturns custom training arguments to set on TrainingArgs.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe global axolotl configuration.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\nobject\n\ndict containing the training arguments.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.get_training_args_mixin()\nReturns a dataclass model for the plugin’s training arguments.\n\n\n\nintegrations.base.BasePlugin.load_datasets(cfg, preprocess=False)\nLoads and preprocesses the dataset for training.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\npreprocess\nbool\nWhether this is the preprocess step of the datasets.\nFalse\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\ndataset_meta\nUnion['TrainDatasetMeta', None]\nThe metadata for the training dataset.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_lora_load(cfg, model)\nPerforms actions after LoRA weights are loaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_model_build(cfg, model)\nPerforms actions after the model is built/loaded, but before any adapters are applied.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_model_load(cfg, model)\nPerforms actions after the model is loaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_train(cfg, model)\nPerforms actions after training is complete.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe axolotl configuration.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_train_unload(cfg)\nPerforms actions after training is complete and the model is unloaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_trainer_create(cfg, trainer)\nPerforms actions after the trainer is created.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.pre_lora_load(cfg, model)\nPerforms actions before LoRA weights are loaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.pre_model_load(cfg)\nPerforms actions before the model is loaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.register(cfg)\nRegisters the plugin with the given configuration.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\n\n\n\n\n\n\n\nintegrations.base.PluginManager()\nThe PluginManager class is responsible for loading and managing plugins. It\nshould be a singleton so it can be accessed from anywhere in the codebase.\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\nplugins\nOrderedDict[str, BasePlugin]\nA list of loaded plugins.\n\n\n\n\n\n\nKey methods include:\n- get_instance(): Static method to get the singleton instance of PluginManager.\n- register(plugin_name: str): Registers a new plugin by its name.\n- pre_model_load(cfg): Calls the pre_model_load method of all registered plugins.\n\n\n\n\n\n\nName\nDescription\n\n\n\n\nadd_callbacks_post_trainer\nCalls the add_callbacks_post_trainer method of all registered plugins.\n\n\nadd_callbacks_pre_trainer\nCalls the add_callbacks_pre_trainer method of all registered plugins.\n\n\ncreate_lr_scheduler\nCalls the create_lr_scheduler method of all registered plugins and returns\n\n\ncreate_optimizer\nCalls the create_optimizer method of all registered plugins and returns\n\n\nget_collator_cls_and_kwargs\nCalls the get_collator_cls_and_kwargs method of all registered plugins and returns the first non-None collator class.\n\n\nget_input_args\nReturns a list of Pydantic classes for all registered plugins’ input arguments.’\n\n\nget_instance\nReturns the singleton instance of PluginManager. If the instance doesn’t\n\n\nget_trainer_cls\nCalls the get_trainer_cls method of all registered plugins and returns the\n\n\nget_training_args\nCalls the get_training_args method of all registered plugins and returns the combined training arguments.\n\n\nget_training_args_mixin\nReturns a list of dataclasses for all registered plugins’ training args mixins’\n\n\nload_datasets\nCalls the load_datasets method of each registered plugin.\n\n\npost_lora_load\nCalls the post_lora_load method of all registered plugins.\n\n\npost_model_build\nCalls the post_model_build method of all registered plugins after the\n\n\npost_model_load\nCalls the post_model_load method of all registered plugins after the model\n\n\npost_train\nCalls the post_train method of all registered plugins.\n\n\npost_train_unload\nCalls the post_train_unload method of all registered plugins.\n\n\npost_trainer_create\nCalls the post_trainer_create method of all registered plugins.\n\n\npre_lora_load\nCalls the pre_lora_load method of all registered plugins.\n\n\npre_model_load\nCalls the pre_model_load method of all registered plugins.\n\n\nregister\nRegisters a new plugin by its name.\n\n\n\n\n\nintegrations.base.PluginManager.add_callbacks_post_trainer(cfg, trainer)\nCalls the add_callbacks_post_trainer method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[Callable]\nA list of callback functions to be added to the TrainingArgs.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.add_callbacks_pre_trainer(cfg, model)\nCalls the add_callbacks_pre_trainer method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[Callable]\nA list of callback functions to be added to the TrainingArgs.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.create_lr_scheduler(\n trainer,\n optimizer,\n num_training_steps,\n)\nCalls the create_lr_scheduler method of all registered plugins and returns\nthe first non-None scheduler.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\noptimizer\nOptimizer\nThe optimizer for training.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nLRScheduler | None\nThe created learning rate scheduler, or None if not found.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.create_optimizer(trainer)\nCalls the create_optimizer method of all registered plugins and returns\nthe first non-None optimizer.\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nOptimizer | None\nThe created optimizer, or None if none was found.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.get_collator_cls_and_kwargs(cfg, is_eval=False)\nCalls the get_collator_cls_and_kwargs method of all registered plugins and returns the first non-None collator class.\nParameters:\ncfg (dict): The configuration for the plugins.\nis_eval (bool): Whether this is an eval split.\nReturns:\nobject: The collator class, or None if none was found.\n\n\n\nintegrations.base.PluginManager.get_input_args()\nReturns a list of Pydantic classes for all registered plugins’ input arguments.’\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[str]\nA list of Pydantic classes for all registered plugins’ input arguments.’\n\n\n\n\n\n\n\nintegrations.base.PluginManager.get_instance()\nReturns the singleton instance of PluginManager. If the instance doesn’t\nexist, it creates a new one.\n\n\n\nintegrations.base.PluginManager.get_trainer_cls(cfg)\nCalls the get_trainer_cls method of all registered plugins and returns the\nfirst non-None trainer class.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nTrainer | None\nThe first non-None trainer class returned by a plugin.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.get_training_args(cfg)\nCalls the get_training_args method of all registered plugins and returns the combined training arguments.\nParameters:\ncfg (dict): The configuration for the plugins.\nReturns:\nobject: The training arguments\n\n\n\nintegrations.base.PluginManager.get_training_args_mixin()\nReturns a list of dataclasses for all registered plugins’ training args mixins’\nReturns:\nlist[str]: A list of dataclsses\n\n\n\nintegrations.base.PluginManager.load_datasets(cfg, preprocess=False)\nCalls the load_datasets method of each registered plugin.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\npreprocess\nbool\nWhether this is preprocess step of the datasets.\nFalse\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nUnion['TrainDatasetMeta', None]\nThe dataset metadata loaded from all registered plugins.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_lora_load(cfg, model)\nCalls the post_lora_load method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_model_build(cfg, model)\nCalls the post_model_build method of all registered plugins after the\nmodel has been built / loaded, but before any adapters have been applied.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_model_load(cfg, model)\nCalls the post_model_load method of all registered plugins after the model\nhas been loaded inclusive of any adapters.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_train(cfg, model)\nCalls the post_train method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_train_unload(cfg)\nCalls the post_train_unload method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_trainer_create(cfg, trainer)\nCalls the post_trainer_create method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.pre_lora_load(cfg, model)\nCalls the pre_lora_load method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.pre_model_load(cfg)\nCalls the pre_model_load method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.register(plugin_name)\nRegisters a new plugin by its name.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\nplugin_name\nstr\nThe name of the plugin to be registered.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nImportError\nIf the plugin module cannot be imported." + "text": "Name\nDescription\n\n\n\n\nBaseOptimizerFactory\nBase class for factories to create custom optimizers\n\n\nBasePlugin\nBase class for all plugins. Defines the interface for plugin methods.\n\n\nPluginManager\nThe PluginManager class is responsible for loading and managing plugins. It\n\n\n\n\n\nintegrations.base.BaseOptimizerFactory()\nBase class for factories to create custom optimizers\n\n\n\n\n\nName\nDescription\n\n\n\n\nget_decay_parameter_names\nGet all parameter names that weight decay will be applied to.\n\n\n\n\n\nintegrations.base.BaseOptimizerFactory.get_decay_parameter_names(model)\nGet all parameter names that weight decay will be applied to.\nThis function filters out parameters in two ways:\n1. By layer type (instances of layers specified in ALL_LAYERNORM_LAYERS)\n2. By parameter name patterns (containing ‘bias’, or variation of ‘norm’)\n\n\n\n\n\nintegrations.base.BasePlugin()\nBase class for all plugins. Defines the interface for plugin methods.\nA plugin is a reusable, modular, and self-contained piece of code that extends\nthe functionality of Axolotl. Plugins can be used to integrate third-party models,\nmodify the training process, or add new features.\nTo create a new plugin, you need to inherit from the BasePlugin class and\nimplement the required methods.\n\n\nPlugin methods include:\n- register(cfg): Registers the plugin with the given configuration.\n- load_datasets(cfg): Loads and preprocesses the dataset for training.\n- pre_model_load(cfg): Performs actions before the model is loaded.\n- post_model_build(cfg, model): Performs actions after the model is loaded, but\nbefore LoRA adapters are applied.\n- pre_lora_load(cfg, model): Performs actions before LoRA weights are loaded.\n- post_lora_load(cfg, model): Performs actions after LoRA weights are loaded.\n- post_model_load(cfg, model): Performs actions after the model is loaded,\ninclusive of any adapters.\n- post_trainer_create(cfg, trainer): Performs actions after the trainer is\ncreated.\n- create_optimizer(cfg, trainer): Creates and returns an optimizer for training.\n- create_lr_scheduler(cfg, trainer, optimizer, num_training_steps): Creates and\nreturns a learning rate scheduler.\n- add_callbacks_pre_trainer(cfg, model): Adds callbacks to the trainer before\ntraining.\n- add_callbacks_post_trainer(cfg, trainer): Adds callbacks to the trainer after\ntraining.\n\n\n\n\n\n\nName\nDescription\n\n\n\n\nadd_callbacks_post_trainer\nAdds callbacks to the trainer after creating the trainer. This is useful for\n\n\nadd_callbacks_pre_trainer\nSet up callbacks before creating the trainer.\n\n\ncreate_lr_scheduler\nCreates and returns a learning rate scheduler.\n\n\ncreate_optimizer\nCreates and returns an optimizer for training.\n\n\nget_collator_cls_and_kwargs\nReturns a custom class for the collator.\n\n\nget_input_args\nReturns a pydantic model for the plugin’s input arguments.\n\n\nget_trainer_cls\nReturns a custom class for the trainer.\n\n\nget_training_args\nReturns custom training arguments to set on TrainingArgs.\n\n\nget_training_args_mixin\nReturns a dataclass model for the plugin’s training arguments.\n\n\nload_datasets\nLoads and preprocesses the dataset for training.\n\n\npost_lora_load\nPerforms actions after LoRA weights are loaded.\n\n\npost_model_build\nPerforms actions after the model is built/loaded, but before any adapters are applied.\n\n\npost_model_load\nPerforms actions after the model is loaded.\n\n\npost_train\nPerforms actions after training is complete.\n\n\npost_train_unload\nPerforms actions after training is complete and the model is unloaded.\n\n\npost_trainer_create\nPerforms actions after the trainer is created.\n\n\npre_lora_load\nPerforms actions before LoRA weights are loaded.\n\n\npre_model_load\nPerforms actions before the model is loaded.\n\n\nregister\nRegisters the plugin with the given configuration as an unparsed dict.\n\n\n\n\n\nintegrations.base.BasePlugin.add_callbacks_post_trainer(cfg, trainer)\nAdds callbacks to the trainer after creating the trainer. This is useful for\ncallbacks that require access to the model or trainer.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[Callable]\nA list of callback functions to be added\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.add_callbacks_pre_trainer(cfg, model)\nSet up callbacks before creating the trainer.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[Callable]\nA list of callback functions to be added to the TrainingArgs.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.create_lr_scheduler(\n cfg,\n trainer,\n optimizer,\n num_training_steps,\n)\nCreates and returns a learning rate scheduler.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\noptimizer\nOptimizer\nThe optimizer for training.\nrequired\n\n\nnum_training_steps\nint\nTotal number of training steps\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nLRScheduler | None\nThe created learning rate scheduler.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.create_optimizer(cfg, trainer)\nCreates and returns an optimizer for training.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nOptimizer | None\nThe created optimizer.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.get_collator_cls_and_kwargs(cfg, is_eval=False)\nReturns a custom class for the collator.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe global axolotl configuration.\nrequired\n\n\nis_eval\nbool\nWhether this is an eval split.\nFalse\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\nclass\n\nThe class for the collator.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.get_input_args()\nReturns a pydantic model for the plugin’s input arguments.\n\n\n\nintegrations.base.BasePlugin.get_trainer_cls(cfg)\nReturns a custom class for the trainer.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe global axolotl configuration.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nTrainer | None\nThe first non-None trainer class returned by a plugin.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.get_training_args(cfg)\nReturns custom training arguments to set on TrainingArgs.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe global axolotl configuration.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\nobject\n\ndict containing the training arguments.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.get_training_args_mixin()\nReturns a dataclass model for the plugin’s training arguments.\n\n\n\nintegrations.base.BasePlugin.load_datasets(cfg, preprocess=False)\nLoads and preprocesses the dataset for training.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\npreprocess\nbool\nWhether this is the preprocess step of the datasets.\nFalse\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\ndataset_meta\nUnion['TrainDatasetMeta', None]\nThe metadata for the training dataset.\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_lora_load(cfg, model)\nPerforms actions after LoRA weights are loaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_model_build(cfg, model)\nPerforms actions after the model is built/loaded, but before any adapters are applied.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_model_load(cfg, model)\nPerforms actions after the model is loaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_train(cfg, model)\nPerforms actions after training is complete.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe axolotl configuration.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_train_unload(cfg)\nPerforms actions after training is complete and the model is unloaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.post_trainer_create(cfg, trainer)\nPerforms actions after the trainer is created.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.pre_lora_load(cfg, model)\nPerforms actions before LoRA weights are loaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.pre_model_load(cfg)\nPerforms actions before the model is loaded.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugin.\nrequired\n\n\n\n\n\n\n\nintegrations.base.BasePlugin.register(cfg)\nRegisters the plugin with the given configuration as an unparsed dict.\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\ndict\nThe configuration for the plugin.\nrequired\n\n\n\n\n\n\n\n\n\nintegrations.base.PluginManager()\nThe PluginManager class is responsible for loading and managing plugins. It\nshould be a singleton so it can be accessed from anywhere in the codebase.\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\nplugins\nOrderedDict[str, BasePlugin]\nA list of loaded plugins.\n\n\n\n\n\n\nKey methods include:\n- get_instance(): Static method to get the singleton instance of PluginManager.\n- register(plugin_name: str): Registers a new plugin by its name.\n- pre_model_load(cfg): Calls the pre_model_load method of all registered plugins.\n\n\n\n\n\n\nName\nDescription\n\n\n\n\nadd_callbacks_post_trainer\nCalls the add_callbacks_post_trainer method of all registered plugins.\n\n\nadd_callbacks_pre_trainer\nCalls the add_callbacks_pre_trainer method of all registered plugins.\n\n\ncreate_lr_scheduler\nCalls the create_lr_scheduler method of all registered plugins and returns\n\n\ncreate_optimizer\nCalls the create_optimizer method of all registered plugins and returns\n\n\nget_collator_cls_and_kwargs\nCalls the get_collator_cls_and_kwargs method of all registered plugins and returns the first non-None collator class.\n\n\nget_input_args\nReturns a list of Pydantic classes for all registered plugins’ input arguments.’\n\n\nget_instance\nReturns the singleton instance of PluginManager. If the instance doesn’t\n\n\nget_trainer_cls\nCalls the get_trainer_cls method of all registered plugins and returns the\n\n\nget_training_args\nCalls the get_training_args method of all registered plugins and returns the combined training arguments.\n\n\nget_training_args_mixin\nReturns a list of dataclasses for all registered plugins’ training args mixins’\n\n\nload_datasets\nCalls the load_datasets method of each registered plugin.\n\n\npost_lora_load\nCalls the post_lora_load method of all registered plugins.\n\n\npost_model_build\nCalls the post_model_build method of all registered plugins after the\n\n\npost_model_load\nCalls the post_model_load method of all registered plugins after the model\n\n\npost_train\nCalls the post_train method of all registered plugins.\n\n\npost_train_unload\nCalls the post_train_unload method of all registered plugins.\n\n\npost_trainer_create\nCalls the post_trainer_create method of all registered plugins.\n\n\npre_lora_load\nCalls the pre_lora_load method of all registered plugins.\n\n\npre_model_load\nCalls the pre_model_load method of all registered plugins.\n\n\nregister\nRegisters a new plugin by its name.\n\n\n\n\n\nintegrations.base.PluginManager.add_callbacks_post_trainer(cfg, trainer)\nCalls the add_callbacks_post_trainer method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[Callable]\nA list of callback functions to be added to the TrainingArgs.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.add_callbacks_pre_trainer(cfg, model)\nCalls the add_callbacks_pre_trainer method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[Callable]\nA list of callback functions to be added to the TrainingArgs.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.create_lr_scheduler(\n trainer,\n optimizer,\n num_training_steps,\n)\nCalls the create_lr_scheduler method of all registered plugins and returns\nthe first non-None scheduler.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\noptimizer\nOptimizer\nThe optimizer for training.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nLRScheduler | None\nThe created learning rate scheduler, or None if not found.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.create_optimizer(trainer)\nCalls the create_optimizer method of all registered plugins and returns\nthe first non-None optimizer.\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nOptimizer | None\nThe created optimizer, or None if none was found.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.get_collator_cls_and_kwargs(cfg, is_eval=False)\nCalls the get_collator_cls_and_kwargs method of all registered plugins and returns the first non-None collator class.\nParameters:\ncfg (dict): The configuration for the plugins.\nis_eval (bool): Whether this is an eval split.\nReturns:\nobject: The collator class, or None if none was found.\n\n\n\nintegrations.base.PluginManager.get_input_args()\nReturns a list of Pydantic classes for all registered plugins’ input arguments.’\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nlist[str]\nA list of Pydantic classes for all registered plugins’ input arguments.’\n\n\n\n\n\n\n\nintegrations.base.PluginManager.get_instance()\nReturns the singleton instance of PluginManager. If the instance doesn’t\nexist, it creates a new one.\n\n\n\nintegrations.base.PluginManager.get_trainer_cls(cfg)\nCalls the get_trainer_cls method of all registered plugins and returns the\nfirst non-None trainer class.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nTrainer | None\nThe first non-None trainer class returned by a plugin.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.get_training_args(cfg)\nCalls the get_training_args method of all registered plugins and returns the combined training arguments.\nParameters:\ncfg (dict): The configuration for the plugins.\nReturns:\nobject: The training arguments\n\n\n\nintegrations.base.PluginManager.get_training_args_mixin()\nReturns a list of dataclasses for all registered plugins’ training args mixins’\nReturns:\nlist[str]: A list of dataclsses\n\n\n\nintegrations.base.PluginManager.load_datasets(cfg, preprocess=False)\nCalls the load_datasets method of each registered plugin.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\npreprocess\nbool\nWhether this is preprocess step of the datasets.\nFalse\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nUnion['TrainDatasetMeta', None]\nThe dataset metadata loaded from all registered plugins.\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_lora_load(cfg, model)\nCalls the post_lora_load method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_model_build(cfg, model)\nCalls the post_model_build method of all registered plugins after the\nmodel has been built / loaded, but before any adapters have been applied.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_model_load(cfg, model)\nCalls the post_model_load method of all registered plugins after the model\nhas been loaded inclusive of any adapters.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_train(cfg, model)\nCalls the post_train method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel | PeftModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_train_unload(cfg)\nCalls the post_train_unload method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.post_trainer_create(cfg, trainer)\nCalls the post_trainer_create method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\ntrainer\nTrainer\nThe trainer object for training.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.pre_lora_load(cfg, model)\nCalls the pre_lora_load method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\nmodel\nPreTrainedModel\nThe loaded model.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.pre_model_load(cfg)\nCalls the pre_model_load method of all registered plugins.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\ncfg\nDictDefault\nThe configuration for the plugins.\nrequired\n\n\n\n\n\n\n\nintegrations.base.PluginManager.register(plugin_name)\nRegisters a new plugin by its name.\n\n\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\nDefault\n\n\n\n\nplugin_name\nstr\nThe name of the plugin to be registered.\nrequired\n\n\n\n\n\n\n\n\n\nName\nType\nDescription\n\n\n\n\n\nImportError\nIf the plugin module cannot be imported." }, { "objectID": "docs/api/integrations.base.html#functions", diff --git a/sitemap.xml b/sitemap.xml index 7c81f6bcc..1a5955e4e 100644 --- a/sitemap.xml +++ b/sitemap.xml @@ -2,794 +2,790 @@ https://docs.axolotl.ai/index.html - 2025-08-09T18:34:14.787Z + 2025-08-13T10:39:47.486Z https://docs.axolotl.ai/src/axolotl/integrations/LICENSE.html - 2025-08-09T18:34:14.791Z + 2025-08-13T10:39:47.490Z https://docs.axolotl.ai/docs/gradient_checkpointing.html - 2025-08-09T18:34:14.768Z + 2025-08-13T10:39:47.467Z https://docs.axolotl.ai/docs/mixed_precision.html - 2025-08-09T18:34:14.771Z + 2025-08-13T10:39:47.470Z https://docs.axolotl.ai/docs/sequence_parallelism.html - 2025-08-09T18:34:14.772Z + 2025-08-13T10:39:47.471Z https://docs.axolotl.ai/docs/docker.html - 2025-08-09T18:34:14.768Z + 2025-08-13T10:39:47.467Z https://docs.axolotl.ai/docs/torchao.html - 2025-08-09T18:34:14.772Z + 2025-08-13T10:39:47.471Z https://docs.axolotl.ai/docs/multi-gpu.html - 2025-08-09T18:34:14.771Z + 2025-08-13T10:39:47.470Z https://docs.axolotl.ai/docs/dataset_preprocessing.html - 2025-08-09T18:34:14.768Z + 2025-08-13T10:39:47.467Z https://docs.axolotl.ai/docs/debugging.html - 2025-08-09T18:34:14.768Z + 2025-08-13T10:39:47.467Z https://docs.axolotl.ai/docs/rlhf.html - 2025-08-09T18:34:14.772Z + 2025-08-13T10:39:47.470Z https://docs.axolotl.ai/docs/lr_groups.html - 2025-08-09T18:34:14.771Z + 2025-08-13T10:39:47.470Z https://docs.axolotl.ai/docs/multimodal.html - 2025-08-09T18:34:14.771Z + 2025-08-13T10:39:47.470Z https://docs.axolotl.ai/docs/ray-integration.html - 2025-08-09T18:34:14.772Z + 2025-08-13T10:39:47.470Z https://docs.axolotl.ai/docs/input_output.html - 2025-08-09T18:34:14.771Z + 2025-08-13T10:39:47.470Z https://docs.axolotl.ai/docs/inference.html - 2025-08-09T18:34:14.771Z + 2025-08-13T10:39:47.470Z https://docs.axolotl.ai/docs/fsdp_qlora.html - 2025-08-09T18:34:14.768Z + 2025-08-13T10:39:47.467Z https://docs.axolotl.ai/docs/multipack.html - 2025-08-09T18:34:14.771Z + 2025-08-13T10:39:47.470Z https://docs.axolotl.ai/docs/api/prompt_strategies.input_output.html - 2025-08-09T18:37:23.820Z + 2025-08-13T10:43:00.281Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_patch_multipack.html - 2025-08-09T18:37:24.083Z + 2025-08-13T10:43:00.544Z https://docs.axolotl.ai/docs/api/cli.art.html - 2025-08-09T18:37:23.459Z + 2025-08-13T10:42:59.920Z https://docs.axolotl.ai/docs/api/cli.quantize.html - 2025-08-09T18:37:23.534Z + 2025-08-13T10:42:59.996Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_attn_hijack_flash.html - 2025-08-09T18:37:24.035Z + 2025-08-13T10:43:00.496Z https://docs.axolotl.ai/docs/api/utils.callbacks.profiler.html - 2025-08-09T18:37:24.611Z + 2025-08-13T10:43:01.073Z https://docs.axolotl.ai/docs/api/prompt_strategies.stepwise_supervised.html - 2025-08-09T18:37:23.825Z + 2025-08-13T10:43:00.286Z https://docs.axolotl.ai/docs/api/integrations.cut_cross_entropy.args.html - 2025-08-09T18:37:24.496Z + 2025-08-13T10:43:00.957Z https://docs.axolotl.ai/docs/api/utils.data.sft.html - 2025-08-09T18:37:24.246Z + 2025-08-13T10:43:00.707Z https://docs.axolotl.ai/docs/api/monkeypatch.unsloth_.html - 2025-08-09T18:37:24.100Z + 2025-08-13T10:43:00.561Z https://docs.axolotl.ai/docs/api/kernels.geglu.html - 2025-08-09T18:37:24.010Z + 2025-08-13T10:43:00.471Z https://docs.axolotl.ai/docs/api/prompt_strategies.orpo.chat_template.html - 2025-08-09T18:37:23.910Z + 2025-08-13T10:43:00.370Z https://docs.axolotl.ai/docs/api/cli.utils.sweeps.html - 2025-08-09T18:37:23.580Z + 2025-08-13T10:43:00.041Z https://docs.axolotl.ai/docs/api/cli.delinearize_llama4.html - 2025-08-09T18:37:23.487Z + 2025-08-13T10:42:59.949Z https://docs.axolotl.ai/docs/api/prompt_strategies.pygmalion.html - 2025-08-09T18:37:23.842Z + 2025-08-13T10:43:00.302Z https://docs.axolotl.ai/docs/api/evaluate.html - 2025-08-09T18:37:23.245Z + 2025-08-13T10:42:59.702Z https://docs.axolotl.ai/docs/api/utils.data.pretraining.html - 2025-08-09T18:37:24.240Z + 2025-08-13T10:43:00.700Z https://docs.axolotl.ai/docs/api/index.html - 2025-08-09T18:37:23.175Z + 2025-08-13T10:42:59.631Z https://docs.axolotl.ai/docs/api/monkeypatch.stablelm_attn_hijack_flash.html - 2025-08-09T18:37:24.089Z + 2025-08-13T10:43:00.550Z https://docs.axolotl.ai/docs/api/monkeypatch.utils.html - 2025-08-09T18:37:24.080Z + 2025-08-13T10:43:00.541Z https://docs.axolotl.ai/docs/api/cli.checks.html - 2025-08-09T18:37:23.465Z + 2025-08-13T10:42:59.926Z https://docs.axolotl.ai/docs/api/utils.chat_templates.html - 2025-08-09T18:37:24.140Z + 2025-08-13T10:43:00.600Z https://docs.axolotl.ai/docs/api/core.builders.rl.html - 2025-08-09T18:37:23.335Z + 2025-08-13T10:42:59.791Z https://docs.axolotl.ai/docs/api/prompt_strategies.messages.chat.html - 2025-08-09T18:37:23.846Z + 2025-08-13T10:43:00.307Z https://docs.axolotl.ai/docs/api/core.trainers.mixins.optimizer.html - 2025-08-09T18:37:23.694Z + 2025-08-13T10:43:00.156Z https://docs.axolotl.ai/docs/api/prompt_strategies.orcamini.html - 2025-08-09T18:37:23.835Z + 2025-08-13T10:43:00.296Z https://docs.axolotl.ai/docs/api/core.trainers.mixins.scheduler.html - 2025-08-09T18:37:23.704Z + 2025-08-13T10:43:00.165Z https://docs.axolotl.ai/docs/api/cli.utils.fetch.html - 2025-08-09T18:37:23.568Z + 2025-08-13T10:43:00.030Z https://docs.axolotl.ai/docs/api/utils.schemas.datasets.html - 2025-08-09T18:37:24.312Z + 2025-08-13T10:43:00.773Z https://docs.axolotl.ai/docs/api/cli.cloud.base.html - 2025-08-09T18:37:23.544Z + 2025-08-13T10:43:00.006Z https://docs.axolotl.ai/docs/api/cli.utils.args.html - 2025-08-09T18:37:23.563Z + 2025-08-13T10:43:00.025Z https://docs.axolotl.ai/docs/api/utils.callbacks.comet_.html - 2025-08-09T18:37:24.620Z + 2025-08-13T10:43:01.081Z https://docs.axolotl.ai/docs/api/utils.callbacks.mlflow_.html - 2025-08-09T18:37:24.616Z + 2025-08-13T10:43:01.078Z https://docs.axolotl.ai/docs/api/core.builders.causal.html - 2025-08-09T18:37:23.330Z + 2025-08-13T10:42:59.787Z https://docs.axolotl.ai/docs/api/cli.train.html - 2025-08-09T18:37:23.429Z + 2025-08-13T10:42:59.890Z https://docs.axolotl.ai/docs/api/utils.schemas.integrations.html - 2025-08-09T18:37:24.342Z + 2025-08-13T10:43:00.802Z https://docs.axolotl.ai/docs/api/integrations.lm_eval.args.html - 2025-08-09T18:37:24.511Z + 2025-08-13T10:43:00.972Z https://docs.axolotl.ai/docs/api/cli.evaluate.html - 2025-08-09T18:37:23.437Z + 2025-08-13T10:42:59.898Z https://docs.axolotl.ai/docs/api/utils.trainer.html - 2025-08-09T18:37:24.178Z + 2025-08-13T10:43:00.639Z https://docs.axolotl.ai/docs/api/prompt_strategies.kto.llama3.html - 2025-08-09T18:37:23.880Z + 2025-08-13T10:43:00.341Z https://docs.axolotl.ai/docs/api/convert.html - 2025-08-09T18:37:23.269Z + 2025-08-13T10:42:59.726Z https://docs.axolotl.ai/docs/api/utils.schemas.multimodal.html - 2025-08-09T18:37:24.329Z + 2025-08-13T10:43:00.790Z https://docs.axolotl.ai/docs/api/loaders.patch_manager.html - 2025-08-09T18:37:23.687Z + 2025-08-13T10:43:00.149Z https://docs.axolotl.ai/docs/api/utils.schemas.training.html - 2025-08-09T18:37:24.295Z + 2025-08-13T10:43:00.755Z https://docs.axolotl.ai/docs/api/utils.schemas.config.html - 2025-08-09T18:37:24.281Z + 2025-08-13T10:43:00.741Z https://docs.axolotl.ai/docs/api/prompt_strategies.kto.user_defined.html - 2025-08-09T18:37:23.890Z + 2025-08-13T10:43:00.350Z https://docs.axolotl.ai/docs/api/prompt_strategies.bradley_terry.llama3.html - 2025-08-09T18:37:23.913Z + 2025-08-13T10:43:00.374Z https://docs.axolotl.ai/docs/api/cli.vllm_serve.html - 2025-08-09T18:37:23.541Z + 2025-08-13T10:43:00.002Z https://docs.axolotl.ai/docs/api/prompt_strategies.alpaca_w_system.html - 2025-08-09T18:37:23.788Z + 2025-08-13T10:43:00.249Z https://docs.axolotl.ai/docs/api/cli.merge_lora.html - 2025-08-09T18:37:23.509Z + 2025-08-13T10:42:59.971Z https://docs.axolotl.ai/docs/api/utils.ctx_managers.sequence_parallel.html - 2025-08-09T18:37:23.727Z + 2025-08-13T10:43:00.189Z https://docs.axolotl.ai/docs/api/prompt_strategies.alpaca_instruct.html - 2025-08-09T18:37:23.776Z + 2025-08-13T10:43:00.238Z https://docs.axolotl.ai/docs/api/utils.bench.html - 2025-08-09T18:37:24.154Z + 2025-08-13T10:43:00.614Z https://docs.axolotl.ai/docs/api/common.datasets.html - 2025-08-09T18:37:24.532Z + 2025-08-13T10:43:00.993Z https://docs.axolotl.ai/docs/api/cli.utils.train.html - 2025-08-09T18:37:23.590Z + 2025-08-13T10:43:00.053Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_attn_hijack_xformers.html - 2025-08-09T18:37:24.036Z + 2025-08-13T10:43:00.497Z https://docs.axolotl.ai/docs/api/core.chat.messages.html - 2025-08-09T18:37:23.370Z + 2025-08-13T10:42:59.827Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.chat_template.html - 2025-08-09T18:37:23.847Z + 2025-08-13T10:43:00.308Z https://docs.axolotl.ai/docs/api/core.trainers.trl.html - 2025-08-09T18:37:23.616Z + 2025-08-13T10:43:00.078Z https://docs.axolotl.ai/docs/api/cli.preprocess.html - 2025-08-09T18:37:23.529Z + 2025-08-13T10:42:59.991Z https://docs.axolotl.ai/docs/api/kernels.swiglu.html - 2025-08-09T18:37:24.020Z + 2025-08-13T10:43:00.481Z https://docs.axolotl.ai/docs/api/kernels.quantize.html - 2025-08-09T18:37:24.028Z + 2025-08-13T10:43:00.489Z https://docs.axolotl.ai/docs/api/prompt_strategies.chat_template.html - 2025-08-09T18:37:23.761Z + 2025-08-13T10:43:00.223Z https://docs.axolotl.ai/docs/api/prompt_strategies.kto.chatml.html - 2025-08-09T18:37:23.888Z + 2025-08-13T10:43:00.349Z https://docs.axolotl.ai/docs/api/core.trainers.grpo.trainer.html - 2025-08-09T18:37:23.639Z + 2025-08-13T10:43:00.101Z https://docs.axolotl.ai/docs/api/monkeypatch.mistral_attn_hijack_flash.html - 2025-08-09T18:37:24.038Z + 2025-08-13T10:43:00.499Z https://docs.axolotl.ai/docs/api/core.datasets.chat.html - 2025-08-09T18:37:23.380Z + 2025-08-13T10:42:59.836Z https://docs.axolotl.ai/docs/api/cli.args.html - 2025-08-09T18:37:23.456Z + 2025-08-13T10:42:59.916Z https://docs.axolotl.ai/docs/api/cli.main.html - 2025-08-09T18:37:23.420Z + 2025-08-13T10:42:59.882Z https://docs.axolotl.ai/docs/api/core.trainers.dpo.trainer.html - 2025-08-09T18:37:23.628Z + 2025-08-13T10:43:00.090Z https://docs.axolotl.ai/docs/api/utils.schemas.trl.html - 2025-08-09T18:37:24.324Z + 2025-08-13T10:43:00.785Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.passthrough.html - 2025-08-09T18:37:23.872Z + 2025-08-13T10:43:00.333Z https://docs.axolotl.ai/docs/api/prompt_tokenizers.html - 2025-08-09T18:37:23.310Z + 2025-08-13T10:42:59.767Z https://docs.axolotl.ai/docs/api/prompt_strategies.alpaca_chat.html - 2025-08-09T18:37:23.775Z + 2025-08-13T10:43:00.236Z https://docs.axolotl.ai/docs/api/logging_config.html - 2025-08-09T18:37:23.320Z + 2025-08-13T10:42:59.776Z https://docs.axolotl.ai/docs/dataset-formats/tokenized.html - 2025-08-09T18:34:14.768Z + 2025-08-13T10:39:47.467Z https://docs.axolotl.ai/docs/dataset-formats/index.html - 2025-08-09T18:34:14.768Z + 2025-08-13T10:39:47.466Z https://docs.axolotl.ai/docs/dataset-formats/pretraining.html - 2025-08-09T18:34:14.768Z + 2025-08-13T10:39:47.466Z https://docs.axolotl.ai/docs/dataset-formats/inst_tune.html - 2025-08-09T18:34:14.768Z + 2025-08-13T10:39:47.466Z https://docs.axolotl.ai/docs/qat.html - 2025-08-09T18:34:14.772Z + 2025-08-13T10:39:47.470Z https://docs.axolotl.ai/examples/colab-notebooks/colab-axolotl-example.html - 2025-08-09T18:34:14.776Z - - - https://docs.axolotl.ai/TODO.html - 2025-08-09T18:34:14.766Z + 2025-08-13T10:39:47.475Z https://docs.axolotl.ai/FAQS.html - 2025-08-09T18:34:14.766Z + 2025-08-13T10:39:47.465Z https://docs.axolotl.ai/docs/installation.html - 2025-08-09T18:34:14.771Z + 2025-08-13T10:39:47.470Z https://docs.axolotl.ai/docs/dataset-formats/stepwise_supervised.html - 2025-08-09T18:34:14.768Z + 2025-08-13T10:39:47.467Z https://docs.axolotl.ai/docs/dataset-formats/template_free.html - 2025-08-09T18:34:14.768Z + 2025-08-13T10:39:47.467Z https://docs.axolotl.ai/docs/dataset-formats/conversation.html - 2025-08-09T18:34:14.768Z + 2025-08-13T10:39:47.466Z https://docs.axolotl.ai/docs/api/utils.dict.html - 2025-08-09T18:37:24.231Z + 2025-08-13T10:43:00.691Z https://docs.axolotl.ai/docs/api/prompt_strategies.completion.html - 2025-08-09T18:37:23.814Z + 2025-08-13T10:43:00.275Z https://docs.axolotl.ai/docs/api/utils.collators.core.html - 2025-08-09T18:37:24.534Z + 2025-08-13T10:43:00.996Z https://docs.axolotl.ai/docs/api/cli.inference.html - 2025-08-09T18:37:23.501Z + 2025-08-13T10:42:59.963Z https://docs.axolotl.ai/docs/api/utils.freeze.html - 2025-08-09T18:37:24.162Z + 2025-08-13T10:43:00.622Z https://docs.axolotl.ai/docs/api/core.trainers.grpo.sampler.html - 2025-08-09T18:37:23.651Z + 2025-08-13T10:43:00.113Z https://docs.axolotl.ai/docs/api/core.trainers.mixins.rng_state_loader.html - 2025-08-09T18:37:23.697Z + 2025-08-13T10:43:00.159Z https://docs.axolotl.ai/docs/api/cli.utils.html - 2025-08-09T18:37:23.552Z + 2025-08-13T10:43:00.013Z https://docs.axolotl.ai/docs/api/core.chat.format.shared.html - 2025-08-09T18:37:23.375Z + 2025-08-13T10:42:59.831Z https://docs.axolotl.ai/docs/api/utils.callbacks.lisa.html - 2025-08-09T18:37:24.613Z + 2025-08-13T10:43:01.074Z https://docs.axolotl.ai/docs/api/utils.collators.mm_chat.html - 2025-08-09T18:37:24.561Z + 2025-08-13T10:43:01.022Z https://docs.axolotl.ai/docs/api/core.trainers.utils.html - 2025-08-09T18:37:23.652Z + 2025-08-13T10:43:00.115Z https://docs.axolotl.ai/docs/api/utils.optimizers.adopt.html - 2025-08-09T18:37:24.238Z + 2025-08-13T10:43:00.698Z https://docs.axolotl.ai/docs/api/integrations.base.html - 2025-08-09T18:37:24.492Z + 2025-08-13T10:43:00.954Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.user_defined.html - 2025-08-09T18:37:23.871Z + 2025-08-13T10:43:00.331Z https://docs.axolotl.ai/docs/api/monkeypatch.btlm_attn_hijack_flash.html - 2025-08-09T18:37:24.082Z + 2025-08-13T10:43:00.542Z https://docs.axolotl.ai/docs/api/utils.quantization.html - 2025-08-09T18:37:24.267Z + 2025-08-13T10:43:00.727Z https://docs.axolotl.ai/docs/api/utils.callbacks.qat.html - 2025-08-09T18:37:24.627Z + 2025-08-13T10:43:01.088Z https://docs.axolotl.ai/docs/api/core.builders.base.html - 2025-08-09T18:37:23.326Z + 2025-08-13T10:42:59.782Z https://docs.axolotl.ai/docs/api/monkeypatch.gradient_checkpointing.offload_cpu.html - 2025-08-09T18:37:24.106Z + 2025-08-13T10:43:00.567Z https://docs.axolotl.ai/docs/api/integrations.kd.trainer.html - 2025-08-09T18:37:24.504Z + 2025-08-13T10:43:00.966Z https://docs.axolotl.ai/docs/api/integrations.liger.args.html - 2025-08-09T18:37:24.508Z + 2025-08-13T10:43:00.969Z https://docs.axolotl.ai/docs/api/utils.collators.mamba.html - 2025-08-09T18:37:24.557Z + 2025-08-13T10:43:01.018Z https://docs.axolotl.ai/docs/api/loaders.model.html - 2025-08-09T18:37:23.662Z + 2025-08-13T10:43:00.124Z https://docs.axolotl.ai/docs/api/utils.schedulers.html - 2025-08-09T18:37:24.206Z + 2025-08-13T10:43:00.666Z https://docs.axolotl.ai/docs/api/kernels.lora.html - 2025-08-09T18:37:24.000Z + 2025-08-13T10:43:00.461Z https://docs.axolotl.ai/docs/api/utils.model_shard_quant.html - 2025-08-09T18:37:24.150Z + 2025-08-13T10:43:00.611Z https://docs.axolotl.ai/docs/api/core.chat.format.llama3x.html - 2025-08-09T18:37:23.373Z + 2025-08-13T10:42:59.830Z https://docs.axolotl.ai/docs/api/core.trainers.mamba.html - 2025-08-09T18:37:23.621Z + 2025-08-13T10:43:00.084Z https://docs.axolotl.ai/docs/api/utils.schemas.enums.html - 2025-08-09T18:37:24.352Z + 2025-08-13T10:43:00.812Z https://docs.axolotl.ai/docs/api/monkeypatch.mixtral.html - 2025-08-09T18:37:24.103Z + 2025-08-13T10:43:00.564Z https://docs.axolotl.ai/docs/api/kernels.utils.html - 2025-08-09T18:37:24.029Z + 2025-08-13T10:43:00.490Z https://docs.axolotl.ai/docs/api/core.training_args.html - 2025-08-09T18:37:23.348Z + 2025-08-13T10:42:59.804Z https://docs.axolotl.ai/docs/api/utils.callbacks.perplexity.html - 2025-08-09T18:37:24.608Z + 2025-08-13T10:43:01.069Z https://docs.axolotl.ai/docs/api/cli.cloud.modal_.html - 2025-08-09T18:37:23.550Z + 2025-08-13T10:43:00.012Z https://docs.axolotl.ai/docs/api/cli.utils.load.html - 2025-08-09T18:37:23.574Z + 2025-08-13T10:43:00.035Z https://docs.axolotl.ai/docs/api/train.html - 2025-08-09T18:37:23.235Z + 2025-08-13T10:42:59.692Z https://docs.axolotl.ai/docs/api/integrations.grokfast.optimizer.html - 2025-08-09T18:37:24.497Z + 2025-08-13T10:43:00.958Z https://docs.axolotl.ai/docs/api/utils.samplers.multipack.html - 2025-08-09T18:37:24.601Z + 2025-08-13T10:43:01.063Z https://docs.axolotl.ai/docs/api/prompt_strategies.metharme.html - 2025-08-09T18:37:23.831Z + 2025-08-13T10:43:00.292Z https://docs.axolotl.ai/docs/api/monkeypatch.llama_expand_mask.html - 2025-08-09T18:37:24.044Z + 2025-08-13T10:43:00.505Z https://docs.axolotl.ai/docs/api/monkeypatch.data.batch_dataset_fetcher.html - 2025-08-09T18:37:24.101Z + 2025-08-13T10:43:00.562Z https://docs.axolotl.ai/docs/api/utils.lora.html - 2025-08-09T18:37:24.145Z + 2025-08-13T10:43:00.605Z https://docs.axolotl.ai/docs/api/loaders.tokenizer.html - 2025-08-09T18:37:23.670Z + 2025-08-13T10:43:00.132Z https://docs.axolotl.ai/docs/api/core.chat.format.chatml.html - 2025-08-09T18:37:23.372Z + 2025-08-13T10:42:59.828Z https://docs.axolotl.ai/docs/api/utils.collators.batching.html - 2025-08-09T18:37:24.553Z + 2025-08-13T10:43:01.014Z https://docs.axolotl.ai/docs/api/cli.merge_sharded_fsdp_weights.html - 2025-08-09T18:37:23.521Z + 2025-08-13T10:42:59.983Z https://docs.axolotl.ai/docs/api/prompt_strategies.llama2_chat.html - 2025-08-09T18:37:23.808Z + 2025-08-13T10:43:00.269Z https://docs.axolotl.ai/docs/api/utils.tokenization.html - 2025-08-09T18:37:24.138Z + 2025-08-13T10:43:00.599Z https://docs.axolotl.ai/docs/api/common.architectures.html - 2025-08-09T18:37:24.516Z + 2025-08-13T10:43:00.977Z https://docs.axolotl.ai/docs/api/core.datasets.transforms.chat_builder.html - 2025-08-09T18:37:23.387Z + 2025-08-13T10:42:59.844Z https://docs.axolotl.ai/docs/api/core.trainers.base.html - 2025-08-09T18:37:23.601Z + 2025-08-13T10:43:00.063Z https://docs.axolotl.ai/docs/api/monkeypatch.lora_kernels.html - 2025-08-09T18:37:24.072Z + 2025-08-13T10:43:00.533Z https://docs.axolotl.ai/docs/api/utils.schemas.utils.html - 2025-08-09T18:37:24.357Z + 2025-08-13T10:43:00.818Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.llama3.html - 2025-08-09T18:37:23.857Z + 2025-08-13T10:43:00.318Z https://docs.axolotl.ai/docs/api/cli.config.html - 2025-08-09T18:37:23.483Z + 2025-08-13T10:42:59.944Z https://docs.axolotl.ai/docs/api/utils.schemas.peft.html - 2025-08-09T18:37:24.321Z + 2025-08-13T10:43:00.781Z https://docs.axolotl.ai/docs/api/prompt_strategies.user_defined.html - 2025-08-09T18:37:23.796Z + 2025-08-13T10:43:00.257Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.zephyr.html - 2025-08-09T18:37:23.869Z + 2025-08-13T10:43:00.330Z https://docs.axolotl.ai/docs/api/monkeypatch.multipack.html - 2025-08-09T18:37:24.039Z + 2025-08-13T10:43:00.500Z https://docs.axolotl.ai/docs/api/prompt_strategies.base.html - 2025-08-09T18:37:23.729Z + 2025-08-13T10:43:00.190Z https://docs.axolotl.ai/docs/api/models.mamba.modeling_mamba.html - 2025-08-09T18:37:24.533Z + 2025-08-13T10:43:00.994Z https://docs.axolotl.ai/docs/api/monkeypatch.relora.html - 2025-08-09T18:37:24.043Z + 2025-08-13T10:43:00.503Z https://docs.axolotl.ai/docs/api/common.const.html - 2025-08-09T18:37:24.517Z + 2025-08-13T10:43:00.978Z https://docs.axolotl.ai/docs/api/monkeypatch.trainer_fsdp_optim.html - 2025-08-09T18:37:24.092Z + 2025-08-13T10:43:00.553Z https://docs.axolotl.ai/docs/api/utils.distributed.html - 2025-08-09T18:37:24.225Z + 2025-08-13T10:43:00.685Z https://docs.axolotl.ai/docs/api/loaders.constants.html - 2025-08-09T18:37:23.688Z + 2025-08-13T10:43:00.150Z https://docs.axolotl.ai/docs/api/monkeypatch.gradient_checkpointing.offload_disk.html - 2025-08-09T18:37:24.132Z + 2025-08-13T10:43:00.592Z https://docs.axolotl.ai/docs/api/datasets.html - 2025-08-09T18:37:23.256Z + 2025-08-13T10:42:59.713Z https://docs.axolotl.ai/docs/api/monkeypatch.transformers_fa_utils.html - 2025-08-09T18:37:24.098Z + 2025-08-13T10:43:00.559Z https://docs.axolotl.ai/docs/api/loaders.processor.html - 2025-08-09T18:37:23.672Z + 2025-08-13T10:43:00.134Z https://docs.axolotl.ai/docs/api/integrations.spectrum.args.html - 2025-08-09T18:37:24.514Z + 2025-08-13T10:43:00.975Z https://docs.axolotl.ai/docs/api/loaders.adapter.html - 2025-08-09T18:37:23.677Z + 2025-08-13T10:43:00.139Z https://docs.axolotl.ai/docs/api/utils.schemas.model.html - 2025-08-09T18:37:24.288Z + 2025-08-13T10:43:00.748Z https://docs.axolotl.ai/docs/api/prompt_strategies.dpo.chatml.html - 2025-08-09T18:37:23.868Z + 2025-08-13T10:43:00.328Z https://docs.axolotl.ai/docs/batch_vs_grad.html - 2025-08-09T18:34:14.767Z + 2025-08-13T10:39:47.466Z https://docs.axolotl.ai/docs/mac.html - 2025-08-09T18:34:14.771Z + 2025-08-13T10:39:47.470Z https://docs.axolotl.ai/docs/nd_parallelism.html - 2025-08-09T18:34:14.771Z + 2025-08-13T10:39:47.470Z https://docs.axolotl.ai/docs/dataset_loading.html - 2025-08-09T18:34:14.768Z + 2025-08-13T10:39:47.467Z https://docs.axolotl.ai/docs/lora_optims.html - 2025-08-09T18:34:14.771Z + 2025-08-13T10:39:47.470Z https://docs.axolotl.ai/docs/unsloth.html - 2025-08-09T18:34:14.772Z + 2025-08-13T10:39:47.471Z https://docs.axolotl.ai/docs/config-reference.html - 2025-08-09T18:37:37.885Z + 2025-08-13T10:43:14.294Z https://docs.axolotl.ai/docs/custom_integrations.html - 2025-08-09T18:34:14.768Z + 2025-08-13T10:39:47.466Z https://docs.axolotl.ai/docs/faq.html - 2025-08-09T18:34:14.768Z + 2025-08-13T10:39:47.467Z https://docs.axolotl.ai/docs/amd_hpc.html - 2025-08-09T18:34:14.767Z + 2025-08-13T10:39:47.466Z https://docs.axolotl.ai/docs/multi-node.html - 2025-08-09T18:34:14.771Z + 2025-08-13T10:39:47.470Z https://docs.axolotl.ai/docs/cli.html - 2025-08-09T18:34:14.767Z + 2025-08-13T10:39:47.466Z https://docs.axolotl.ai/docs/nccl.html - 2025-08-09T18:34:14.771Z + 2025-08-13T10:39:47.470Z https://docs.axolotl.ai/docs/optimizers.html - 2025-08-09T18:34:14.772Z + 2025-08-13T10:39:47.470Z https://docs.axolotl.ai/docs/getting-started.html - 2025-08-09T18:34:14.768Z + 2025-08-13T10:39:47.467Z https://docs.axolotl.ai/docs/quantize.html - 2025-08-09T18:34:14.772Z + 2025-08-13T10:39:47.470Z https://docs.axolotl.ai/docs/reward_modelling.html - 2025-08-09T18:34:14.772Z + 2025-08-13T10:39:47.470Z https://docs.axolotl.ai/src/axolotl/integrations/cut_cross_entropy/ACKNOWLEDGEMENTS.html - 2025-08-09T18:34:14.791Z + 2025-08-13T10:39:47.491Z
Name
cfgDictDefaultdict The configuration for the plugin. required