Update docs and docstrings related to Llama3VisionTransform (pytorch#2382)

Author: Ankur-singh

docs/source/basics/model_transforms.rst

@@ -32,8 +32,8 @@ These are intended to be drop-in replacements for tokenizers in multimodal datas
             Message(
                 role="user",
                 content=[
-                    {"type": "image", "content": Image.new(mode="RGB", size=(224, 224))},
-                    {"type": "image", "content": Image.new(mode="RGB", size=(224, 224))},
+                    {"type": "image", "content": Image.new(mode="RGB", size=(560, 560))},
+                    {"type": "image", "content": Image.new(mode="RGB", size=(560, 560))},
                     {"type": "text", "content": "What is common in these two images?"},
                 ],
             ),

docs/source/basics/multimodal_datasets.rst

@@ -45,7 +45,7 @@ in the text, ``"<image>"`` for where to place the image tokens. This will get re
     from torchtune.models.llama3_2_vision import llama3_2_vision_transform
     from torchtune.datasets.multimodal import multimodal_chat_dataset
 
-    model_transform = Llama3VisionTransform(
+    model_transform = llama3_2_vision_transform(
         path="/tmp/Meta-Llama-3-8B-Instruct/original/tokenizer.model",
         prompt_template="torchtune.data.QuestionAnswerTemplate",
         max_seq_len=8192,

torchtune/datasets/multimodal/_multimodal.py

@@ -120,10 +120,9 @@ def multimodal_chat_dataset(
 
     ::
 
-        >>> from torchtune.datasets.multimodal import multimodal_chat_dataset
         >>> from torchtune.models.llama3_2_vision import llama3_2_vision_transform
         >>> from torchtune.datasets.multimodal import multimodal_chat_dataset
-        >>> model_transform = Llama3VisionTransform(
+        >>> model_transform = llama3_2_vision_transform(
         >>>     path="/tmp/Meta-Llama-3-8B-Instruct/original/tokenizer.model",
         >>>     prompt_template="torchtune.data.QuestionAnswerTemplate",
         >>>     max_seq_len=8192,

torchtune/models/llama3_2_vision/_transform.py

@@ -194,7 +194,7 @@ def __call__(
 
         Args:
             sample (Mapping[str, Any]): A sample with a "messages" field.
-            inference (bool): Whether to run in inference mode. Default is True.
+            inference (bool): Whether to run in inference mode. Default is False.
 
         Returns:
             Mapping[str, Any]: The transformed sample with the following fields: