added parameters to adjust max size of inf_subset and to avoid duplic…

…ates in it

src/small_model_training/config.json

@@ -9,5 +9,7 @@
  "evaluation_strategy":"epoch",
  "save_strategy":"epoch",
  "load_best_model_at_end":true,
- "push_to_hub":false
+ "push_to_hub":false,
+ "avoid_duplicate_fewshots": true,
+ "max_inf_subset_size": 5
 }

src/small_model_training/text_classification.py

@@ -13,6 +13,15 @@ class ModelArguments:
  metadata={"help": "Path to pretrained model or model from huggingface.co/models"}
  )
 
+@dataclass
+class FewshotArguments:
+ avoid_duplicate_fewshots: bool = field(
+ metadata={"help": "Decides whether to avoid duplicate fewshot examples or not"}
+ )
+ max_inf_subset_size: int = field(
+ metadata={"help": "Limit for the size of the influential subset"}
+ )
+
 def get_influential_subset(dataset):
  id2label = {}
  label2id = {}
@@ -28,8 +37,8 @@ def get_influential_subset(dataset):
  counter += 1
 
  # get parameters from config
- parser = HfArgumentParser((ModelArguments, TrainingArguments))
- model_args, training_args = parser.parse_json_file('config.json')
+ parser = HfArgumentParser((ModelArguments, TrainingArguments, FewshotArguments))
+ model_args, training_args, fewshot_args = parser.parse_json_file('config.json')
 
  # setup preprocessing
  tokenizer = AutoTokenizer.from_pretrained(model_args.model_name_or_path)
@@ -78,13 +87,25 @@ def compute_metrics(eval_pred):
 
  inf_subset = []
 
- indices = calculate_candidates_entropy_knn(outputs, math.ceil(len(dataset["test"])/2))
+ indices = calculate_candidates_entropy_knn(outputs, min(math.ceil(len(dataset["test"])/2),fewshot_args.max_inf_subset_size))
 
  for elem in indices:
  inf_subset.append(dataset["test"][elem.item()])
 
  inf_subset = Dataset.from_list(inf_subset)
 
+ if fewshot_args.avoid_duplicate_fewshots:
+ unique_samples = set() # Set to store unique samples
+ deduplicated_inf_subset = [] # List to store deduplicated samples
+
+ # Iterate through the dataset and filter out duplicates
+ for example in inf_subset:
+ text = example['text']
+ if text not in unique_samples:
+ unique_samples.add(text)
+ deduplicated_inf_subset.append(example)
+ inf_subset = Dataset.from_list(deduplicated_inf_subset)
+
  return inf_subset
 
 #TO-DO: dynamic variant for calculating inf_subset