def run_eval()

in src/jobs/tune_t5.py [0:0]

• 43 lines of code
• 4 McCabe index (conditional complexity)

    def run_eval(self, eval_dataset, name="Eval", prefix=None, log_wandb=True):
        results_labels = []
        results_output = []
        losses = []
        for item in eval_dataset:
            input = item["input_text"]
            label = item['target_text']
            input_encodings = self.tokenizer(input, return_tensors="pt", truncation=True,
                                             padding=True)
            label_encodings = self.tokenizer(label, return_tensors="pt", truncation=True,
                                             padding=True)
            input_ids = input_encodings["input_ids"].to(self.device)
            attention_mask = input_encodings["attention_mask"].to(self.device)
            label_ids = label_encodings["input_ids"].to(self.device)

            with torch.no_grad():
                outputs = self.model(
                    input_ids=input_ids,
                    attention_mask=attention_mask,
                    labels=label_ids
                )
            loss = outputs.loss.item()
            outputs = self.model.generate(input_ids, max_length=30, num_return_sequences=1)
            response = self.tokenizer.decode(outputs[0], skip_special_tokens=True)
            results_output.append(response)
            results_labels.append(label)
            losses.append(loss)

        metrics = self.compute_metrics_text(results_output, results_labels, prefix=prefix)
        if (log_wandb):
            wandb.log(metrics)
            table = wandb.Table(
                columns=["input", "label", "prediction", "loss"],
                data=list(
                    zip(
                        [d["input_text"] for d in eval_dataset],
                        results_labels,
                        results_output,
                        losses
                    )
                ),
            )
            wandb.log({f"{name} Set": table})
        else:
            print(f"{name} stats {json.dumps(metrics)}")