[Fix] Regular Linting

- Added missing comma in AVAILABLE_MODELS for consistency.
- Reordered import statements in vora.py for better readability.
- Simplified input_data generation by condensing method calls into a single line.
- Ensured default generation parameters are set correctly in the VoRA class.

Author: Luodian

lmms_eval/models/__init__.py

@@ -70,7 +70,7 @@
     "videochat_flash": "VideoChat_Flash",
     "whisper": "Whisper",
     "whisper_vllm": "WhisperVllm",
-    "vora": "VoRA"
+    "vora": "VoRA",
 }
 
 

lmms_eval/models/openai_compatible.py

@@ -189,12 +189,12 @@ def generate_until(self, requests) -> List[str]:
             payload["max_output_tokens"] = gen_kwargs["max_new_tokens"]
             payload["temperature"] = gen_kwargs["temperature"]
 
-            if 'o1' in self.model_version or 'o3' in self.model_version:
-                del payload['max_output_tokens']
-                del payload['temperature']
-                payload['reasoning_effort'] = 'medium'
-                payload['response_format'] = {'type': 'text'}
-            
+            if "o1" in self.model_version or "o3" in self.model_version:
+                del payload["max_output_tokens"]
+                del payload["temperature"]
+                payload["reasoning_effort"] = "medium"
+                payload["response_format"] = {"type": "text"}
+
             for attempt in range(self.max_retries):
                 try:
                     response = self.client.chat.completions.create(**payload)

lmms_eval/models/vora.py

@@ -1,19 +1,19 @@
 import base64
-from io import BytesIO
 import os
-from typing import List, Optional, Tuple, Union
 import uuid
 import warnings
+from io import BytesIO
+from typing import List, Optional, Tuple, Union
 
 warnings.simplefilter("ignore", category=DeprecationWarning)
 warnings.filterwarnings("ignore")
 
+import torch
 from accelerate import Accelerator, DistributedType
 from loguru import logger as eval_logger
 from PIL import Image
-import torch
 from tqdm import tqdm
-from transformers import AutoTokenizer, AutoModelForCausalLM, AutoProcessor
+from transformers import AutoModelForCausalLM, AutoProcessor, AutoTokenizer
 
 from lmms_eval import utils
 from lmms_eval.api.instance import Instance
@@ -208,16 +208,10 @@ def _collate(x):
                         message.append({"role": "user", "content": [{"type": "text", "text": context}]})
                 else:
                     message.append({"role": "user", "content": [{"type": "text", "text": context}]})
-                
+
                 messages.append(message)
 
-            input_data = self.processor.apply_chat_template(
-                messages,
-                add_generation_prompt=True,
-                tokenize=True,
-                return_tensors='pt',
-                return_dict=True
-            ).to(self.device).to(self.dtype)
+            input_data = self.processor.apply_chat_template(messages, add_generation_prompt=True, tokenize=True, return_tensors="pt", return_dict=True).to(self.device).to(self.dtype)
 
             # preconfigure gen_kwargs with defaults
             if "max_new_tokens" not in gen_kwargs:
@@ -228,21 +222,18 @@ def _collate(x):
                 gen_kwargs["top_p"] = None
             if "num_beams" not in gen_kwargs:
                 gen_kwargs["num_beams"] = 1
-            if 'do_sample' not in gen_kwargs:
-                gen_kwargs['do_sample'] = False
-            if 'repetition_penalty' not in gen_kwargs:
-                gen_kwargs['repetition_penalty'] = 1.0
-            if 'length_penalty' not in gen_kwargs:
-                gen_kwargs['length_penalty'] = 0.9
-            if 'min_length' not in gen_kwargs:
-                gen_kwargs['min_length'] = 4
-            if 'eos_token_id' not in gen_kwargs:
-                gen_kwargs['eos_token_id'] = self.tokenizer.eos_token_id
-
-            prompt = self.processor.apply_chat_template(
-                messages,
-                add_generation_prompt=True
-            )
+            if "do_sample" not in gen_kwargs:
+                gen_kwargs["do_sample"] = False
+            if "repetition_penalty" not in gen_kwargs:
+                gen_kwargs["repetition_penalty"] = 1.0
+            if "length_penalty" not in gen_kwargs:
+                gen_kwargs["length_penalty"] = 0.9
+            if "min_length" not in gen_kwargs:
+                gen_kwargs["min_length"] = 4
+            if "eos_token_id" not in gen_kwargs:
+                gen_kwargs["eos_token_id"] = self.tokenizer.eos_token_id
+
+            prompt = self.processor.apply_chat_template(messages, add_generation_prompt=True)
             cont = self.model.generate(
                 input_data,
                 **gen_kwargs,