From 3595cb901fd0464c6d91d526ad8a35d0cf59add3 Mon Sep 17 00:00:00 2001
From: Quarto GHA Workflow Runner Overwrite the Perform a training step on a batch of inputs. Overrides the
-push_to_hub
Overwrite the
-push_to_hub method in order to force-add the tags when pushing the
-
training_step
-Perform a training step on a batch of inputs. Overrides the
-core.trainers.base.AxolotlTrainer.push_to_hub(*args, **kwargs)push_to_hub method in order to force-add the tags when pushing the
model on the Hub. Please refer to ~transformers.Trainer.push_to_hub for more details.training_step
-core.trainers.base.AxolotlTrainer.training_step(
- model,
- inputs,
- num_items_in_batch=None,
-)transformers.trainer.Trainer method to handle sequence parallelism if
-enabled.Parameters
-
-
-
-
-
-Name
-Type
-Description
-Default
-
-
-model
-nn.Module
-Model to perform training step for.
-required
-
-
-
-inputs
-dict[str, torch.Tensor | Any]
-Dictionary mapping.
-required
-