Upload fine_tune_loop.py with huggingface_hub

fine_tune_loop.py

@@ -0,0 +1,103 @@
+from datasets import load_dataset
+from transformers import AutoTokenizer, DataCollatorWithPadding
+from torch.utils.data import DataLoader
+from transformers import AutoModelForSequenceClassification
+from transformers import AdamW
+from transformers import get_scheduler
+import torch
+from tqdm.auto import tqdm
+import evaluate
+
+raw_datasets = load_dataset("glue","mrpc")
+checkpoint = 'bert-base-cased'
+tokenizer = AutoTokenizer.from_pretrained(checkpoint)
+
+def tokenize_function(example):
+    return tokenizer(example['sentence1'], example['sentence2'], truncation=True)
+
+tokenized_dataset = raw_datasets.map(tokenize_function, batched=True)
+tokenized_dataset = tokenized_dataset.remove_columns(['sentence1', 'sentence2','idx'])
+tokenized_dataset = tokenized_dataset.rename_column('label','labels')
+#print(tokenized_dataset.column_names["train"])
+
+tokenized_dataset.set_format('torch')
+#print(tokenized_dataset)
+
+data_collator = DataCollatorWithPadding(tokenizer)
+
+train_dataloader = DataLoader(
+    tokenized_dataset['validation'], batch_size=8, collate_fn=data_collator
+)
+
+eval_dataloader = DataLoader(
+    tokenized_dataset['validation'], batch_size=8, collate_fn=data_collator
+)
+
+#for batch in train_dataloader:
+#    break
+#print({k: v.shape for k, v in batch.items()})
+#print()
+#print(batch)
+#print()
+
+model = AutoModelForSequenceClassification.from_pretrained(checkpoint, num_labels=2)
+
+#outputs = model(**batch)
+#print(outputs.loss, outputs.logits.shape)
+
+optimizer = AdamW(model.parameters(), lr=5e-5)
+
+#loss = outputs.loss
+#loss.backward()
+#optimizer.step()
+
+#optimizer.zero_grad()
+
+num_epochs = 3
+num_training_steps = num_epochs * len(train_dataloader)
+lr_scheduler = get_scheduler(
+    'linear',
+    optimizer=optimizer,
+    num_warmup_steps=0,
+    num_training_steps=num_training_steps
+)
+
+device = torch.device('mps') if torch.backends.mps.is_available() else torch.device('cpu')
+model.to(device)
+print(f'Using device: {device}')
+
+progress_bar = tqdm(range(num_training_steps))
+
+model.train()
+for epoch in range(num_epochs):
+    for batch in train_dataloader:
+        batch = {k: v.to(device) for k, v in batch.items()}
+        outputs = model(**batch)
+        loss = outputs.loss
+        loss.backward()
+
+        optimizer.step()
+        lr_scheduler.step()
+        optimizer.zero_grad()
+        progress_bar.update(1)
+
+metric= evaluate.load('glue','mrpc')
+model.eval()
+for batch in eval_dataloader:
+    batch = {k: v.to(device) for k, v in batch.items()}
+    with torch.no_grad():
+        outputs = model(**batch)
+
+    logits = outputs.logits
+    predictions = torch.argmax(logits, dim=-1)
+    metric.add_batch(predictions=predictions, references=batch['labels'])
+
+result = metric.compute()
+print(result)
+
+save_dir = "/Users/alexandr/Desktop/HUGGING_FACE/model"
+
+model.save_pretrained(save_dir)
+tokenizer.save_pretrained(save_dir)
+
+print(f"model and tokenizer saved to {save_dir}")