Merge pull request #129 from Dannoopsy/mmbench_ru

add task MMBench-ru

lmms_eval/tasks/mmbench/_default_template_mmbench_ru_yaml

@@ -0,0 +1,24 @@
+dataset_path: deepvk/MMBench-ru
+dataset_kwargs:
+  token: True
+doc_to_target: "answer"
+model_specific_prompt_kwargs:
+  default:
+    pre_prompt: ""
+    post_prompt: "\nВыбери правильный вариант ответа буквой."
+doc_to_visual: !function ru_utils.mmbench_doc_to_visual
+doc_to_text: !function ru_utils.mmbench_doc_to_text
+doc_to_target: "answer"
+process_results: !function ru_utils.mmbench_process_results
+model_specific_generation_kwargs:
+  llava:
+    image_aspect_ratio: original
+output_type: generate_until
+generation_kwargs:
+  until:
+    - "ASSISTANT:"
+  max_new_tokens: 1024
+  temperature: 0
+  top_p: 1.0
+  num_beams: 1
+  do_sample: false

lmms_eval/tasks/mmbench/mmbench.yaml

@@ -5,6 +5,7 @@ task:
   - mmbench_cn_dev
   - mmbench_cn_test
   - mmbench_cn_cc
+  - mmbench_ru_dev
 metadata:
   version: 0.0
   sys_prompt: "There are several options:"

lmms_eval/tasks/mmbench/mmbench_ru_dev.yaml

@@ -0,0 +1,10 @@
+task: "mmbench_ru_dev"
+test_split: dev
+include: _default_template_mmbench_ru_yaml
+metric_list:
+  - metric: gpt_eval_score
+    aggregation: !function ru_utils.mmbench_aggregate_dev_results_eval
+    higher_is_better: true
+  - metric: submission
+    aggregation: !function ru_utils.mmbench_aggregate_dev_results_submission
+    higher_is_better: true

lmms_eval/tasks/mmbench/ru_utils.py

@@ -0,0 +1,128 @@
+import yaml
+import os
+from pathlib import Path
+import pandas as pd
+import json
+
+from loguru import logger as eval_logger
+from lmms_eval.tasks.mmbench.mmbench_evals import MMBench_Evaluator
+from lmms_eval.tasks._task_utils.file_utils import generate_submission_file
+
+with open(Path(__file__).parent / "mmbench.yaml", "r") as f:
+    raw_data = f.readlines()
+    safe_data = []
+    for i, line in enumerate(raw_data):
+        # remove function definition since yaml load cannot handle it
+        if "!function" not in line:
+            safe_data.append(line)
+
+    config = yaml.safe_load("".join(safe_data))
+
+GPT_EVAL_MODEL_NAME = config["metadata"]["gpt_eval_model_name"]
+API_TYPE = os.getenv("API_TYPE", "openai")
+
+if API_TYPE == "openai":
+    API_URL = os.getenv("OPENAI_API_URL", "https://api.openai.com/v1/chat/completions")
+    API_KEY = os.getenv("OPENAI_API_KEY", "YOUR_API_KEY")
+elif API_TYPE == "azure":
+    API_URL = os.getenv("AZURE_ENDPOINT", "https://api.cognitive.microsoft.com/sts/v1.0/issueToken")
+    API_KEY = os.getenv("AZURE_API_KEY", "YOUR_API_KEY")
+else:
+    API_URL = "YOUR_API_URL"
+    API_KEY = "YOUR_API_KEY"
+
+
+mmbench_evaluator = MMBench_Evaluator(sys_prompt=config["metadata"]["sys_prompt"], API_KEY=API_KEY, API_URL=API_URL, model_version=GPT_EVAL_MODEL_NAME)
+
+
+def mmbench_doc_to_visual(doc):
+    return [doc["image"].convert("RGB")]
+
+
+def mmbench_doc_to_text(doc, model_specific_prompt_kwargs=None):
+    option_candidate = ["A", "B", "C", "D", "E"]
+    options_prompt, options_dict = mmbench_evaluator.create_options_prompt(doc, option_candidate)
+
+    data = {
+        # "img": doc["image"],
+        "question": doc["question"],
+        "answer": doc.get("answer", None),
+        "options": options_prompt,
+        "category": doc["category"],
+        "L2-category": doc["l2-category"],
+        "options_dict": options_dict,
+        "index": doc["index"],
+        "hint": doc["hint"],
+        "source": doc["source"],
+        "split": doc["split"],
+    }
+
+    query_prompt = f"{data['hint']} {data['question']} {data['options']}" if pd.notna(data["hint"]) and data["hint"] != "nan" else f"{data['question']} {data['options']}"
+
+    if model_specific_prompt_kwargs:
+        query_prompt = f"{query_prompt}\n{model_specific_prompt_kwargs['post_prompt']}"
+
+    return query_prompt
+
+
+def mmbench_process_results(doc, results):
+    model_response = results[0].strip()
+    data = {
+        "gpt_eval_score": {
+            "index": doc["index"],
+            "question": doc["question"],
+            "answer": doc["answer"],
+            "prediction": model_response,
+            "hint": doc["hint"],
+            "source": doc["source"],
+            "split": doc["split"],
+            "category": doc["category"],
+            "L2-category": doc["l2-category"],
+        },
+        "submission": {
+            "index": doc["index"],
+            "question": doc["question"],
+            "answer": doc["answer"],
+            "prediction": model_response,
+            "hint": doc["hint"],
+            "source": doc["source"],
+            "split": doc["split"],
+            "category": doc["category"],
+            "L2-category": doc["l2-category"],
+        },
+    }
+    option_candidate = ["A", "B", "C", "D", "E"]
+    for c in option_candidate:
+        data["submission"][c] = doc.get(c, "nan")
+        data["gpt_eval_score"][c] = doc.get(c, "nan")
+    return data
+
+
+def mmbench_aggregate_dev_results_eval(results, args):
+    print(f"============= MMBench-RU(Dev) Detailed Results =============")
+    overall_acc, category_acc, l2_category_acc = mmbench_evaluator.eval_result(results, eval_method="openai")
+    file = generate_submission_file("mmbench_ru_dev_results.json", args)
+    details_info = {
+        "overall_acc": overall_acc,
+        "category_acc": category_acc,
+        "l2_category_acc": l2_category_acc,
+    }
+    with open(file, "w") as f:
+        json.dump(details_info, f)
+    return overall_acc * 100
+
+
+def mmbench_aggregate_dev_results_submission(results, args):
+    df = pd.DataFrame(results)
+    excel_write_path = generate_submission_file("mmbench_ru_dev_results.xlsx", args)
+    with pd.ExcelWriter(excel_write_path) as writer:
+        df.to_excel(writer, index=False)
+    eval_logger.info(f"Saved results to {excel_write_path}")
+
+
+def mmbench_aggregate_test_results(results, args):
+    df = pd.DataFrame(results)
+    excel_write_path = generate_submission_file("mmbench_ru_test_results.xlsx", args)
+    with pd.ExcelWriter(excel_write_path) as writer:
+        df.to_excel(writer, index=False)
+    eval_logger.info(f"Saved results to {excel_write_path}")