From 55d1be2ae6f037081d7bcca55d93753fc2e10702 Mon Sep 17 00:00:00 2001 From: NanoCode012 Date: Tue, 23 Sep 2025 21:22:15 +0700 Subject: [PATCH] fix: unify default for conversations_field [skip-e2e] (#3070) * fix: unify default for conversations_field * fix: suggestion to remove defaults --- docs/multimodal.qmd | 1 - .../archived/deepcoder/deepcoder-14B-preview-lora.yml | 4 ---- .../deepcogito/cogito-v1-preview-llama-3B-lora.yml | 4 ---- .../deepcogito/cogito-v1-preview-qwen-14B-lora.yml | 4 ---- examples/gemma3/gemma-3-4b-vision-qlora.yml | 2 +- examples/llama-3/instruct-lora-8b.yml | 9 --------- .../llama-4/do-no-use-fa2/scout-vision-qlora-fsdp.yaml | 1 - examples/llama-4/scout-vision-qlora-fsdp2-flex.yaml | 1 - examples/phi/lora-3.5.yaml | 9 --------- examples/qwen2-vl/lora-7b.yaml | 2 +- examples/qwen2_5-vl/lora-7b.yaml | 2 +- src/axolotl/core/datasets/transforms/chat_builder.py | 10 +++++----- 12 files changed, 8 insertions(+), 41 deletions(-) diff --git a/docs/multimodal.qmd b/docs/multimodal.qmd index 413404195..3a28b579a 100644 --- a/docs/multimodal.qmd +++ b/docs/multimodal.qmd @@ -42,7 +42,6 @@ datasets: - path: HuggingFaceH4/llava-instruct-mix-vsft type: chat_template split: train[:1%] - field_messages: messages # (optional) if doing lora, only finetune the Language model, # leave the vision model and vision tower frozen diff --git a/examples/archived/deepcoder/deepcoder-14B-preview-lora.yml b/examples/archived/deepcoder/deepcoder-14B-preview-lora.yml index 2202091d5..3223ec19a 100644 --- a/examples/archived/deepcoder/deepcoder-14B-preview-lora.yml +++ b/examples/archived/deepcoder/deepcoder-14B-preview-lora.yml @@ -9,10 +9,6 @@ strict: false datasets: - path: fozziethebeat/alpaca_messages_2k_test type: chat_template - field_messages: messages - message_property_mappings: - role: role - content: content dataset_prepared_path: val_set_size: 0.05 diff --git a/examples/deepcogito/cogito-v1-preview-llama-3B-lora.yml b/examples/deepcogito/cogito-v1-preview-llama-3B-lora.yml index fc9a75e3f..97d1bb6b3 100644 --- a/examples/deepcogito/cogito-v1-preview-llama-3B-lora.yml +++ b/examples/deepcogito/cogito-v1-preview-llama-3B-lora.yml @@ -9,10 +9,6 @@ strict: false datasets: - path: fozziethebeat/alpaca_messages_2k_test type: chat_template - field_messages: messages - message_property_mappings: - role: role - content: content dataset_prepared_path: val_set_size: 0.05 diff --git a/examples/deepcogito/cogito-v1-preview-qwen-14B-lora.yml b/examples/deepcogito/cogito-v1-preview-qwen-14B-lora.yml index b527edc6f..b80cc5bc0 100644 --- a/examples/deepcogito/cogito-v1-preview-qwen-14B-lora.yml +++ b/examples/deepcogito/cogito-v1-preview-qwen-14B-lora.yml @@ -9,10 +9,6 @@ strict: false datasets: - path: fozziethebeat/alpaca_messages_2k_test type: chat_template - field_messages: messages - message_property_mappings: - role: role - content: content dataset_prepared_path: val_set_size: 0.05 diff --git a/examples/gemma3/gemma-3-4b-vision-qlora.yml b/examples/gemma3/gemma-3-4b-vision-qlora.yml index e9e606b69..b42b6b492 100644 --- a/examples/gemma3/gemma-3-4b-vision-qlora.yml +++ b/examples/gemma3/gemma-3-4b-vision-qlora.yml @@ -18,7 +18,7 @@ datasets: - path: HuggingFaceH4/llava-instruct-mix-vsft type: chat_template split: train[:1%] - field_messages: messages + dataset_prepared_path: last_run_prepared val_set_size: 0.01 output_dir: ./outputs/out diff --git a/examples/llama-3/instruct-lora-8b.yml b/examples/llama-3/instruct-lora-8b.yml index 69e17b9cf..401df1d72 100644 --- a/examples/llama-3/instruct-lora-8b.yml +++ b/examples/llama-3/instruct-lora-8b.yml @@ -12,15 +12,6 @@ chat_template: llama3 datasets: - path: fozziethebeat/alpaca_messages_2k_test type: chat_template - field_messages: messages - message_property_mappings: - role: role - content: content - roles: - user: - - user - assistant: - - assistant dataset_prepared_path: val_set_size: 0.05 diff --git a/examples/llama-4/do-no-use-fa2/scout-vision-qlora-fsdp.yaml b/examples/llama-4/do-no-use-fa2/scout-vision-qlora-fsdp.yaml index 9975949bb..4136dc14a 100644 --- a/examples/llama-4/do-no-use-fa2/scout-vision-qlora-fsdp.yaml +++ b/examples/llama-4/do-no-use-fa2/scout-vision-qlora-fsdp.yaml @@ -46,7 +46,6 @@ datasets: - path: HuggingFaceH4/llava-instruct-mix-vsft type: chat_template split: train[:1%] - field_messages: messages dataset_prepared_path: last_run_prepared val_set_size: 0.0 diff --git a/examples/llama-4/scout-vision-qlora-fsdp2-flex.yaml b/examples/llama-4/scout-vision-qlora-fsdp2-flex.yaml index ac7e05659..5972c2ae3 100644 --- a/examples/llama-4/scout-vision-qlora-fsdp2-flex.yaml +++ b/examples/llama-4/scout-vision-qlora-fsdp2-flex.yaml @@ -45,7 +45,6 @@ datasets: - path: HuggingFaceH4/llava-instruct-mix-vsft type: chat_template split: train[:1%] - field_messages: messages dataset_prepared_path: last_run_prepared val_set_size: 0.0 diff --git a/examples/phi/lora-3.5.yaml b/examples/phi/lora-3.5.yaml index a6fa15d98..c10014dab 100644 --- a/examples/phi/lora-3.5.yaml +++ b/examples/phi/lora-3.5.yaml @@ -12,15 +12,6 @@ chat_template: phi_3 datasets: - path: fozziethebeat/alpaca_messages_2k_test type: chat_template - field_messages: messages - message_property_mappings: - role: role - content: content - roles: - user: - - user - assistant: - - assistant dataset_prepared_path: val_set_size: 0.05 diff --git a/examples/qwen2-vl/lora-7b.yaml b/examples/qwen2-vl/lora-7b.yaml index 8ea608199..285a35cbb 100644 --- a/examples/qwen2-vl/lora-7b.yaml +++ b/examples/qwen2-vl/lora-7b.yaml @@ -11,7 +11,7 @@ datasets: - path: HuggingFaceH4/llava-instruct-mix-vsft type: chat_template split: train[:1%] - field_messages: messages + dataset_prepared_path: last_run_prepared val_set_size: 0.0 output_dir: ./outputs/out diff --git a/examples/qwen2_5-vl/lora-7b.yaml b/examples/qwen2_5-vl/lora-7b.yaml index 13a97dec3..7d499d841 100644 --- a/examples/qwen2_5-vl/lora-7b.yaml +++ b/examples/qwen2_5-vl/lora-7b.yaml @@ -11,7 +11,7 @@ datasets: - path: HuggingFaceH4/llava-instruct-mix-vsft type: chat_template split: train[:1%] - field_messages: messages + dataset_prepared_path: last_run_prepared val_set_size: 0.0 output_dir: ./outputs/out diff --git a/src/axolotl/core/datasets/transforms/chat_builder.py b/src/axolotl/core/datasets/transforms/chat_builder.py index 8f2013027..0de0ecb40 100644 --- a/src/axolotl/core/datasets/transforms/chat_builder.py +++ b/src/axolotl/core/datasets/transforms/chat_builder.py @@ -8,7 +8,7 @@ from typing import Any, Mapping def chat_message_transform_builder( train_on_inputs=False, - conversations_field: str = "conversations", + conversations_field: str = "messages", message_field_role: str | list[str] | None = None, # commonly "role" message_field_content: str | list[str] | None = None, # commonly "content" message_field_training: str | list[str] | None = None, # commonly "weight" @@ -20,13 +20,13 @@ def chat_message_transform_builder( If True, the transform will train on the inputs. If False, the transform will train on the targets. Defaults to False. conversations_field (str, optional): - The field name of the conversations. Defaults to "conversations". + The field name of the conversations. Defaults to "messages". message_field_role (str | list[str], optional): - The field name of the role. Defaults to "role". + The field name of the role. message_field_content (str | list[str], optional): - The field name of the message content. Defaults to "content". + The field name of the message content. message_field_training (str | list[str], optional): - The field name of the train/weight. Defaults to "weight". + The field name of the train/weight. Returns: Callable: