dev

app.py

@@ -6,7 +6,7 @@ from datasets import load_dataset, concatenate_datasets, load_from_disk
 import traceback
 from sklearn.metrics import accuracy_score
 import numpy as np
-#import torch
+import torch
 
 import os
 from huggingface_hub import login
@@ -17,7 +17,7 @@ os.environ['HF_HOME'] = '/data/.huggingface'
 @spaces.GPU(duration=120)
 def fine_tune_model(model_name, dataset_name, hub_id, api_key, num_epochs, batch_size, lr, grad):
     try:
-        #torch.cuda.empty_cache()
+        torch.cuda.empty_cache()
         def compute_metrics(eval_pred):
             logits, labels = eval_pred
             predictions = np.argmax(logits, axis=1)
@@ -35,7 +35,7 @@ def fine_tune_model(model_name, dataset_name, hub_id, api_key, num_epochs, batch
         )        
     
         # Load the model and tokenizer
-        model = AutoModelForSeq2SeqLM.from_pretrained(model_name.strip(), num_labels=2)
+        model = AutoModelForSeq2SeqLM.from_pretrained(model_name.strip(), num_labels=2, force_download=True)
         model.gradient_checkpointing_enable()
         #model = get_peft_model(model, lora_config)