Run Inference

Run Inference

This section shows how to run inference-only workloads on Gaudi. For more advanced information about how to speed up inference, check out this guide.

With GaudiTrainer

You can find below a template to perform inference with a GaudiTrainer instance where we want to compute the accuracy over the given dataset:

Copied

import evaluate

metric = evaluate.load("accuracy")

# You can define your custom compute_metrics function. It takes an `EvalPrediction` object (a namedtuple with a
# predictions and label_ids field) and has to return a dictionary string to float.
def my_compute_metrics(p):
    return metric.compute(predictions=np.argmax(p.predictions, axis=1), references=p.label_ids)

# Trainer initialization
trainer = GaudiTrainer(
        model=my_model,
        gaudi_config=my_gaudi_config,
        args=my_args,
        train_dataset=None,
        eval_dataset=eval_dataset,
        compute_metrics=my_compute_metrics,
        tokenizer=my_tokenizer,
        data_collator=my_data_collator,
    )

# Run inference
metrics = trainer.evaluate()

The variable my_args should contain some inference-specific arguments, you can take a look here to see the arguments that can be interesting to set for inference.

In our Examples

All our examples contain instructions for running inference with a given model on a given dataset. The reasoning is the same for every example: run the example script with --do_eval and --per_device_eval_batch_size and without --do_train. A simple template is the following:

Copied

python path_to_the_example_script \
  --model_name_or_path my_model_name \
  --gaudi_config_name my_gaudi_config_name \
  --dataset_name my_dataset_name \
  --do_eval \
  --per_device_eval_batch_size my_batch_size \
  --output_dir path_to_my_output_dir \
  --use_habana \
  --use_lazy_mode \
  --use_hpu_graphs_for_inference