moved filed to util

tasks/audio.py

@@ -3,17 +3,16 @@ from datetime import datetime
 from datasets import load_dataset
 from sklearn.metrics import accuracy_score
 import numpy as np
-import random
 import os
 import torch
 from torch.utils.data import DataLoader
 
 from .utils.evaluation import AudioEvaluationRequest
 from .utils.emissions import tracker, clean_emissions_data, get_space_info
-from data import FFTDataset
-from models import DualEncoder
-from train import Trainer
-from data_utils import collate_fn, Container
+from .utils.data import FFTDataset
+from .utils.models import DualEncoder
+from .utils.train import Trainer
+from .utils.data_utils import collate_fn, Container
 import yaml
 
 from dotenv import load_dotenv
@@ -61,7 +60,7 @@ async def evaluate_audio(request: AudioEvaluationRequest):
     # Update the code below to replace the random baseline by your model inference within the inference pass where the energy consumption and emissions are tracked.
     #--------------------------------------------------------------------------------------------
     device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-    args_path = 'config.yaml'
+    args_path = 'utils/config.yaml'
     data_args = Container(**yaml.safe_load(open(args_path, 'r'))['Data'])
     model_args = Container(**yaml.safe_load(open(args_path, 'r'))['CNNEncoder'])
     model_args_f = Container(**yaml.safe_load(open(args_path, 'r'))['CNNEncoder_f'])

tasks/{models.py → utils/models.py}

@@ -1,7 +1,7 @@
 import torch
 import torch.nn as nn
-from Modules.conformer import ConformerEncoder, ConformerDecoder
-from Modules.mhsa_pro import RotaryEmbedding, ContinuousRotaryEmbedding
+from .Modules.conformer import ConformerEncoder, ConformerDecoder
+from .Modules.mhsa_pro import RotaryEmbedding, ContinuousRotaryEmbedding
 
 class ConvBlock(nn.Module):
   def __init__(self, args, num_layer) -> None: