Upload folder using huggingface_hub

.gitattributes

@@ -267,3 +267,8 @@ sft/1M3/Full_smoe_sigmoidgating/checkpoint-20679/logs/0628_1717_llava...l_mme_ll
 sft/1M3/Full_smoe_sigmoidgating/checkpoint-20679/logs/0628_1717_llava...l_mme_llava_model_args_a39ef6/mmerealworld_lite.json filter=lfs diff=lfs merge=lfs -text
 sft/1M3/Full_smoe_sigmoidgating/checkpoint-20679/logs/0628_1717_llava...l_mme_llava_model_args_a39ef6/submissions/mmbench_en_dev_results.xlsx filter=lfs diff=lfs merge=lfs -text
 sft/1M3/Full_smoe_sigmoidgating/checkpoint-20679/logs/0628_1717_llava...l_mme_llava_model_args_a39ef6/textvqa_val.json filter=lfs diff=lfs merge=lfs -text
+sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2120_llava...l_mme_llava_model_args_467343/infovqa_val.json filter=lfs diff=lfs merge=lfs -text
+sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2120_llava...l_mme_llava_model_args_467343/mmbench_en_dev.json filter=lfs diff=lfs merge=lfs -text
+sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2120_llava...l_mme_llava_model_args_467343/mmerealworld_lite.json filter=lfs diff=lfs merge=lfs -text
+sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2120_llava...l_mme_llava_model_args_467343/submissions/mmbench_en_dev_results.xlsx filter=lfs diff=lfs merge=lfs -text
+sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2120_llava...l_mme_llava_model_args_467343/textvqa_val.json filter=lfs diff=lfs merge=lfs -text

sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_1805_llava...image_llava_model_args_467343/rank0_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 0 eval done

sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_1805_llava...image_llava_model_args_467343/results.json

@@ -0,0 +1,146 @@
+{
+    "results": {
+        "hallusion_bench_image": {
+            "aAcc,none": 49.1062,
+            "aAcc_stderr,none": "N/A",
+            "fAcc,none": 21.9653,
+            "fAcc_stderr,none": "N/A",
+            "qAcc,none": 19.1209,
+            "qAcc_stderr,none": "N/A",
+            "alias": "hallusion_bench_image"
+        },
+        "mathvista_testmini": {
+            "gpt_eval_score,none": 38.0,
+            "gpt_eval_score_stderr,none": "N/A",
+            "alias": "mathvista_testmini"
+        }
+    },
+    "configs": {
+        "hallusion_bench_image": {
+            "task": "hallusion_bench_image",
+            "dataset_path": "lmms-lab/HallusionBench",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "image",
+            "doc_to_visual": "<function hb_doc_to_visual at 0x7f378506f160>",
+            "doc_to_text": "<function hb_doc_to_text at 0x7f378506f8b0>",
+            "doc_to_target": "gt_answer_details",
+            "process_results": "<function hb_process_results at 0x7f3784fe60d0>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "aAcc",
+                    "aggregation": "<function hb_aggregation_result_aAcc at 0x7f3784fe6a60>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "qAcc",
+                    "aggregation": "<function hb_aggregation_result_qAcc at 0x7f3784ef30d0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "fAcc",
+                    "aggregation": "<function hb_aggregation_result_fAcc at 0x7f3784ef38b0>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 128,
+                "temperature": 0.0,
+                "top_p": 1.0,
+                "num_beams": 1,
+                "do_sample": false,
+                "until": [
+                    "\n\n"
+                ]
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ],
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "pre_prompt": "",
+                    "post_prompt": ""
+                }
+            }
+        },
+        "mathvista_testmini": {
+            "task": "mathvista_testmini",
+            "dataset_path": "AI4Math/MathVista",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "testmini",
+            "doc_to_visual": "<function mathvista_doc_to_visual at 0x7f37593e3c10>",
+            "doc_to_text": "<function mathvista_doc_to_text at 0x7f37593ee310>",
+            "doc_to_target": "answer",
+            "process_results": "<function mathvista_process_results at 0x7f37593f89d0>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "gpt_eval_score",
+                    "aggregation": "<function mathvista_aggregate_results at 0x7f3758ff20d0>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "until": [
+                    "ASSISTANT:"
+                ],
+                "max_new_tokens": 1024,
+                "temperature": 0.0,
+                "top_p": 1.0,
+                "num_beams": 1,
+                "do_sample": false,
+                "image_aspect_ratio": "original"
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "shot_type": "format-prompt",
+                    "shot": 0,
+                    "use_caption": false,
+                    "use_ocr": false
+                },
+                "phi3v": {
+                    "shot_type": "solution"
+                }
+            },
+            "model_specific_generation_kwargs": {
+                "llava": {
+                    "image_aspect_ratio": "original"
+                }
+            }
+        }
+    },
+    "versions": {
+        "hallusion_bench_image": "Yaml",
+        "mathvista_testmini": "Yaml"
+    },
+    "n-shot": {
+        "hallusion_bench_image": 0,
+        "mathvista_testmini": 0
+    },
+    "model_configs": {
+        "model": "llava",
+        "model_args": "pretrained=/cm/archive/namnv78/checkpoints/Xphi35-siglip224/SMOE/1M3/Full_xmoe/checkpoint-13786,conv_template=phi35",
+        "batch_size": "1",
+        "device": null,
+        "limit": null,
+        "bootstrap_iters": 100000,
+        "gen_kwargs": ""
+    },
+    "git_hash": "289c7fe5"
+}

sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2120_llava...l_mme_llava_model_args_467343/infovqa_val.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:faea45b46fc9c3adf8028f2bfca6a260cf274b36e95f8d5867fd68ca76e514f5
+size 576440422

sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2120_llava...l_mme_llava_model_args_467343/mmbench_en_dev.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9e47f86e17f0352af4ac9353a96d90808b2478c32c1f925325b4a4f7f76fa4e
+size 14568246

sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2120_llava...l_mme_llava_model_args_467343/mmerealworld_lite.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ec735a14ec3e2126dc0316d5613d3f8e01a8e9d06f53aaf773a5e261e48fb48
+size 1994104279

sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2120_llava...l_mme_llava_model_args_467343/rank0_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 0 eval done

sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2120_llava...l_mme_llava_model_args_467343/rank1_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 1 eval done

sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2120_llava...l_mme_llava_model_args_467343/rank2_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 2 eval done

sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2120_llava...l_mme_llava_model_args_467343/rank3_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 3 eval done

sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2120_llava...l_mme_llava_model_args_467343/results.json

@@ -0,0 +1,683 @@
+{
+    "results": {
+        "ai2d": {
+            "exact_match,flexible-extract": 0.6777849740932642,
+            "exact_match_stderr,flexible-extract": 0.008411056458256383,
+            "alias": "ai2d"
+        },
+        "infovqa_val": {
+            "anls,none": 0.29096751160299894,
+            "anls_stderr,none": 0.008583735009443328,
+            "alias": "infovqa_val"
+        },
+        "mmbench_en_dev": {
+            "gpt_eval_score,none": 68.47079037800687,
+            "gpt_eval_score_stderr,none": "N/A",
+            "submission,none": null,
+            "submission_stderr,none": "N/A",
+            "alias": "mmbench_en_dev"
+        },
+        "mme": {
+            "mme_cognition_score,none": 306.7857142857143,
+            "mme_cognition_score_stderr,none": "N/A",
+            "mme_percetion_score,none": 1222.2970188075228,
+            "mme_percetion_score_stderr,none": "N/A",
+            "alias": "mme"
+        },
+        "mmerealworld_lite": {
+            "mme_realworld_score,none": 0.3069306930693069,
+            "mme_realworld_score_stderr,none": "N/A",
+            "alias": "mmerealworld_lite"
+        },
+        "mmmu_val": {
+            "mmmu_acc,none": 0.42778,
+            "mmmu_acc_stderr,none": "N/A",
+            "alias": "mmmu_val"
+        },
+        "mmstar": {
+            "coarse perception,none": 0.6355847690782489,
+            "coarse perception_stderr,none": "N/A",
+            "fine-grained perception,none": 0.29908499140260375,
+            "fine-grained perception_stderr,none": "N/A",
+            "instance reasoning,none": 0.5314715354584872,
+            "instance reasoning_stderr,none": "N/A",
+            "logical reasoning,none": 0.3530776154538531,
+            "logical reasoning_stderr,none": "N/A",
+            "math,none": 0.39212443197006147,
+            "math_stderr,none": "N/A",
+            "science & technology,none": 0.3288526969392016,
+            "science & technology_stderr,none": "N/A",
+            "alias": "mmstar"
+        },
+        "scienceqa_img": {
+            "exact_match,none": 0.8086266732771443,
+            "exact_match_stderr,none": 0.008761318545158912,
+            "alias": "scienceqa_img"
+        },
+        "seedbench_2_plus": {
+            "seedbench_2_plus_Chart,none": 0.49506172839506174,
+            "seedbench_2_plus_Chart_stderr,none": "N/A",
+            "seedbench_2_plus_all,none": 0.49231444883618797,
+            "seedbench_2_plus_all_stderr,none": "N/A",
+            "seedbench_2_plus_Web,none": 0.5333333333333333,
+            "seedbench_2_plus_Web_stderr,none": "N/A",
+            "seedbench_2_plus_Map,none": 0.4560099132589839,
+            "seedbench_2_plus_Map_stderr,none": "N/A",
+            "alias": "seedbench_2_plus"
+        },
+        "textvqa_val": {
+            "exact_match,none": 0.39094000024795533,
+            "exact_match_stderr,none": 0.006696816605625882,
+            "submission,none": null,
+            "submission_stderr,none": "N/A",
+            "alias": "textvqa_val"
+        }
+    },
+    "configs": {
+        "ai2d": {
+            "task": "ai2d",
+            "dataset_path": "lmms-lab/ai2d",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "test",
+            "doc_to_visual": "<function ai2d_doc_to_visual at 0x7f44e7770d30>",
+            "doc_to_text": "<function ai2d_doc_to_text at 0x7f44e7778040>",
+            "doc_to_target": "<function ai2d_doc_to_target at 0x7f44e77785e0>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "exact_match",
+                    "aggregation": "mean",
+                    "higher_is_better": true,
+                    "ignore_case": true,
+                    "ignore_punctuation": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 512,
+                "temperature": 0.0,
+                "do_sample": false,
+                "until": [
+                    "\n\n"
+                ]
+            },
+            "repeats": 1,
+            "filter_list": [
+                {
+                    "name": "flexible-extract",
+                    "filter": [
+                        {
+                            "function": "<class 'utils.MultiChoiceRegexFilter'>",
+                            "group_select": 0,
+                            "ignore_case": true,
+                            "ignore_punctuation": true,
+                            "regex_pattern": "([A-Z])\\."
+                        }
+                    ]
+                }
+            ],
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ],
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "prompt_format": "mcq",
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer with the option's letter from the given choices directly."
+                },
+                "gpt4v": {
+                    "prompt_format": "mcq",
+                    "pre_prompt": "",
+                    "post_prompt": "\nAbove choices are given in {option}. {content} format.\nPlease answer with the option letter from the given choices directly."
+                },
+                "qwen_vl": {
+                    "prompt_format": "qa",
+                    "pre_prompt": "",
+                    "post_prompt": " Answer:"
+                },
+                "xcomposer2_4khd": {
+                    "prompt_format": "mcq_xcomposer",
+                    "pre_prompt": "[UNUSED_TOKEN_146]user\nQuestion: ",
+                    "post_prompt": "[UNUSED_TOKEN_145]\n[UNUSED_TOKEN_146]assistant\nThe answer is"
+                }
+            },
+            "model_specific_target_kwargs": {
+                "default": "mcq",
+                "qwen_vl": "qa"
+            }
+        },
+        "infovqa_val": {
+            "task": "infovqa_val",
+            "dataset_path": "lmms-lab/DocVQA",
+            "dataset_name": "InfographicVQA",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "validation",
+            "doc_to_visual": "<function infovqa_doc_to_visual at 0x7f451aef7c10>",
+            "doc_to_text": "<function infovqa_doc_to_text at 0x7f451aef7ee0>",
+            "doc_to_target": "answers",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "anls",
+                    "aggregation": "mean",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 32,
+                "temperature": 0.0,
+                "do_sample": false,
+                "until": [
+                    "\n\n"
+                ]
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer the question using a single word or phrase."
+                }
+            }
+        },
+        "mmbench_en_dev": {
+            "task": "mmbench_en_dev",
+            "dataset_path": "lmms-lab/MMBench",
+            "dataset_name": "en",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "dev",
+            "doc_to_visual": "<function mmbench_doc_to_visual at 0x7f45469985e0>",
+            "doc_to_text": "<function mmbench_doc_to_text at 0x7f4546998af0>",
+            "doc_to_target": "answer",
+            "process_results": "<function mmbench_process_results at 0x7f45469a3040>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "gpt_eval_score",
+                    "aggregation": "<function mmbench_aggregate_dev_results_eval at 0x7f4546a0ee50>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "submission",
+                    "aggregation": "<function mmbench_aggregate_dev_results_submission at 0x7f45469983a0>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "until": [
+                    "ASSISTANT:"
+                ],
+                "max_new_tokens": 1024,
+                "temperature": 0.0,
+                "top_p": 1.0,
+                "num_beams": 1,
+                "do_sample": false,
+                "image_aspect_ratio": "original"
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer with the option's letter from the given choices directly."
+                }
+            },
+            "model_specific_generation_kwargs": {
+                "llava": {
+                    "image_aspect_ratio": "original"
+                }
+            }
+        },
+        "mme": {
+            "task": "mme",
+            "dataset_path": "lmms-lab/MME",
+            "dataset_kwargs": {
+                "token": false
+            },
+            "test_split": "test",
+            "doc_to_visual": "<function mme_doc_to_visual at 0x7f4548475a60>",
+            "doc_to_text": "<function mme_doc_to_text at 0x7f4546a1a310>",
+            "doc_to_target": "answer",
+            "process_results": "<function mme_process_results at 0x7f4546a1a8b0>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "mme_percetion_score",
+                    "aggregation": "<function mme_aggregate_results at 0x7f4546a1adc0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "mme_cognition_score",
+                    "aggregation": "<function mme_aggregate_results at 0x7f4546a22280>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 16,
+                "temperature": 0.0,
+                "top_p": 1.0,
+                "num_beams": 1,
+                "do_sample": false,
+                "until": [
+                    "\n\n"
+                ]
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ],
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer the question using a single word or phrase."
+                },
+                "gpt4v": {
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer the question with Yes or No."
+                },
+                "qwen_vl": {
+                    "pre_prompt": "",
+                    "post_prompt": " Answer:"
+                },
+                "otterhd": {
+                    "pre_prompt": "",
+                    "post_prompt": " Answer:"
+                },
+                "xcomposer2_4khd": {
+                    "pre_prompt": "[UNUSED_TOKEN_146]user\n",
+                    "post_prompt": " Answer this question briefly[UNUSED_TOKEN_145]\n[UNUSED_TOKEN_146]assistant\n"
+                }
+            }
+        },
+        "mmerealworld_lite": {
+            "task": "mmerealworld_lite",
+            "dataset_path": "yifanzhang114/MME-RealWorld-lite-lmms-eval",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "train",
+            "doc_to_visual": "<function mme_realworld_doc_to_visual at 0x7f44e7f19ee0>",
+            "doc_to_text": "<function mme_realworld_doc_to_text at 0x7f44e7f1e9d0>",
+            "doc_to_target": "answer",
+            "process_results": "<function mme_realworld_process_results at 0x7f44e7f22670>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "mme_realworld_score",
+                    "aggregation": "<function mme_realworld_aggregate_results at 0x7f44e7f291f0>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 16,
+                "temperature": 0.0,
+                "top_p": 1.0,
+                "num_beams": 1,
+                "do_sample": false,
+                "until": [
+                    "\n\n"
+                ]
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ],
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "pre_prompt": "",
+                    "post_prompt": "\nSelect the best answer to the above multiple-choice question based on the image. Respond with only the letter (A, B, C, D, or E) of the correct option."
+                },
+                "gpt4v": {
+                    "pre_prompt": "",
+                    "post_prompt": "\nSelect the best answer to the above multiple-choice question based on the image. Respond with only the letter (A, B, C, D, or E) of the correct option."
+                },
+                "xcomposer2_4khd": {
+                    "pre_prompt": "[UNUSED_TOKEN_146]user\n",
+                    "post_prompt": " Answer this question with A, B, C, or D.[UNUSED_TOKEN_145]\n[UNUSED_TOKEN_146]assistant\n"
+                }
+            }
+        },
+        "mmmu_val": {
+            "task": "mmmu_val",
+            "dataset_path": "lmms-lab/MMMU",
+            "test_split": "validation",
+            "doc_to_visual": "<function mmmu_doc_to_visual at 0x7f44fa7ad940>",
+            "doc_to_text": "<function mmmu_doc_to_text at 0x7f44fa6d7670>",
+            "doc_to_target": "answer",
+            "process_results": "<function mmmu_process_results at 0x7f44fa66a550>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "mmmu_acc",
+                    "aggregation": "<function mmmu_aggregate_results at 0x7f44fa4e94c0>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 128,
+                "until": [
+                    "\n\n"
+                ],
+                "image_aspect_ratio": "original"
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ],
+            "model_specific_generation_kwargs": {
+                "llava": {
+                    "image_aspect_ratio": "original"
+                }
+            }
+        },
+        "mmstar": {
+            "task": "mmstar",
+            "dataset_path": "Lin-Chen/MMStar",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "val",
+            "doc_to_visual": "<function mmstar_doc_to_visual at 0x7f44e76ce3a0>",
+            "doc_to_text": "<function mmstar_doc_to_text at 0x7f44e76ce820>",
+            "doc_to_target": "answer",
+            "process_results": "<function mmstar_process_results at 0x7f44e76ced30>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "coarse perception",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7f44e76d81f0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "fine-grained perception",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7f44e76d85e0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "instance reasoning",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7f44e76d89d0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "logical reasoning",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7f44e76d8dc0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "science & technology",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7f44e76e01f0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "math",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7f44e76e05e0>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "until": [
+                    "\n\n"
+                ],
+                "do_sample": false
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ],
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer with the option's letter from the given choices directly"
+                }
+            }
+        },
+        "scienceqa_img": {
+            "task": "scienceqa_img",
+            "dataset_path": "lmms-lab/ScienceQA",
+            "dataset_name": "ScienceQA-IMG",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "test",
+            "doc_to_visual": "<function sqa_doc_to_visual at 0x7f4512cf1dc0>",
+            "doc_to_text": "<function sqa_doc_to_text at 0x7f4512cb9040>",
+            "doc_to_target": "<function sqa_doc_to_target at 0x7f4512cb9430>",
+            "process_results": "<function sqa_process_results at 0x7f4512cb9790>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "exact_match",
+                    "aggregation": "mean",
+                    "higher_is_better": true,
+                    "ignore_case": true,
+                    "ignore_punctuation": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 16,
+                "temperature": 0.0,
+                "do_sample": false,
+                "until": [
+                    "\n\n"
+                ],
+                "image_aspect_ratio": "original"
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ],
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "format": "default",
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer with the option's letter from the given choices directly."
+                },
+                "qwen_vl": {
+                    "format": "qwen_vl"
+                },
+                "idefics2": {
+                    "format": "default",
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer:"
+                }
+            },
+            "model_specific_generation_kwargs": {
+                "llava": {
+                    "image_aspect_ratio": "original"
+                }
+            }
+        },
+        "seedbench_2_plus": {
+            "task": "seedbench_2_plus",
+            "dataset_path": "doolayer/SEED-Bench-2-Plus",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "test",
+            "doc_to_visual": "<function seed_doc_to_visual at 0x7f4510087ee0>",
+            "doc_to_text": "<function seed_doc_to_text at 0x7f4510090550>",
+            "doc_to_target": "answer",
+            "process_results": "<function seed_process_result at 0x7f4510090a60>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "seedbench_2_plus_Chart",
+                    "aggregation": "<function seed_aggregation_result at 0x7f4510090f70>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "seedbench_2_plus_Map",
+                    "aggregation": "<function seed_aggregation_result at 0x7f450fee8430>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "seedbench_2_plus_Web",
+                    "aggregation": "<function seed_aggregation_result at 0x7f450fee88b0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "seedbench_2_plus_all",
+                    "aggregation": "<function seed_aggregation_result at 0x7f450fee8d30>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "until": [
+                    "ASSISTANT:"
+                ],
+                "max_new_tokens": 16,
+                "image_aspect_ratio": "original"
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ],
+            "model_specific_prompt_kwargs": {
+                "llava": {
+                    "img_token": "<image>",
+                    "post_prompt": "Answer with the option's letter from the given choices directly."
+                },
+                "gpt4V": {
+                    "img_token": "<image>",
+                    "post_prompt": "Answer with the option's letter from the given choices directly."
+                },
+                "default": {
+                    "img_token": "<image>",
+                    "post_prompt": "Answer with the option's letter from the given choices directly."
+                }
+            }
+        },
+        "textvqa_val": {
+            "task": "textvqa_val",
+            "dataset_path": "lmms-lab/textvqa",
+            "test_split": "validation",
+            "doc_to_visual": "<function textvqa_doc_to_visual at 0x7f4513fe49d0>",
+            "doc_to_text": "<function textvqa_doc_to_text at 0x7f4513fe4e50>",
+            "doc_to_target": "answer",
+            "process_results": "<function textvqa_process_results at 0x7f4513eaa160>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "exact_match",
+                    "aggregation": "mean",
+                    "higher_is_better": true,
+                    "ignore_case": true,
+                    "ignore_punctuation": true
+                },
+                {
+                    "metric": "submission",
+                    "aggregation": "<function textvqa_aggregate_submissions at 0x7f4513fe4790>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "until": [
+                    "ASSISTANT:"
+                ]
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "pre_prompt": "",
+                    "post_prompt": "\nAnswer the question using a single word or phrase.",
+                    "ocr": false
+                },
+                "qwen_vl": {
+                    "pre_prompt": "",
+                    "post_prompt": " Answer:"
+                }
+            }
+        }
+    },
+    "versions": {
+        "ai2d": "Yaml",
+        "infovqa_val": "Yaml",
+        "mmbench_en_dev": "Yaml",
+        "mme": "Yaml",
+        "mmerealworld_lite": "Yaml",
+        "mmmu_val": "Yaml",
+        "mmstar": "Yaml",
+        "scienceqa_img": "Yaml",
+        "seedbench_2_plus": "Yaml",
+        "textvqa_val": "Yaml"
+    },
+    "n-shot": {
+        "ai2d": 0,
+        "infovqa_val": 0,
+        "mmbench_en_dev": 0,
+        "mme": 0,
+        "mmerealworld_lite": 0,
+        "mmmu_val": 0,
+        "mmstar": 0,
+        "scienceqa_img": 0,
+        "seedbench_2_plus": 0,
+        "textvqa_val": 0
+    },
+    "model_configs": {
+        "model": "llava",
+        "model_args": "pretrained=/cm/archive/namnv78/checkpoints/Xphi35-siglip224/SMOE/1M3/Full_xmoe/checkpoint-13786,conv_template=phi35",
+        "batch_size": "1",
+        "device": null,
+        "limit": null,
+        "bootstrap_iters": 100000,
+        "gen_kwargs": ""
+    },
+    "git_hash": "289c7fe5"
+}

sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2120_llava...l_mme_llava_model_args_467343/submissions/mmbench_en_dev_results.json

@@ -0,0 +1 @@
+{"overall_acc": 0.6847079037800687, "category_acc": {"action_recognition": 0.9074074074074074, "attribute_comparison": 0.6363636363636364, "attribute_recognition": 0.8648648648648649, "celebrity_recognition": 0.7373737373737373, "function_reasoning": 0.8481012658227848, "future_prediction": 0.475, "identity_reasoning": 0.9555555555555556, "image_emotion": 0.78, "image_quality": 0.39622641509433965, "image_scene": 0.9519230769230769, "image_style": 0.8490566037735849, "image_topic": 0.8888888888888888, "nature_relation": 0.625, "object_localization": 0.37037037037037035, "ocr": 0.6666666666666666, "physical_property_reasoning": 0.4, "physical_relation": 0.5833333333333334, "social_relation": 0.8604651162790697, "spatial_relationship": 0.2222222222222222, "structuralized_imagetext_understanding": 0.5256410256410257}, "l2_category_acc": {"attribute_reasoning": 0.7035175879396985, "coarse_perception": 0.7972972972972973, "finegrained_perception (cross-instance)": 0.6083916083916084, "finegrained_perception (instance-level)": 0.658703071672355, "logic_reasoning": 0.5084745762711864, "relation_reasoning": 0.7043478260869566}}

sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2120_llava...l_mme_llava_model_args_467343/submissions/mmbench_en_dev_results.xlsx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eeb8b3a32d80b04f83cf3fd42c44bbeeb2fd3e13f3c9fb1fe77de31aed64e753
+size 865114

sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2120_llava...l_mme_llava_model_args_467343/textvqa_val.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:497d0eee2e25a6ff41b95df22c8101825d0e4a80821fd1bfde45914d651e6a7c
+size 13175441

sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2145_llava...bench_llava_model_args_467343/rank0_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 0 eval done

sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2145_llava...bench_llava_model_args_467343/rank1_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 1 eval done

sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2145_llava...bench_llava_model_args_467343/rank2_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 2 eval done

sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2145_llava...bench_llava_model_args_467343/rank3_metric_eval_done.txt

@@ -0,0 +1 @@
+rank 3 eval done

sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2145_llava...bench_llava_model_args_467343/results.json

@@ -0,0 +1,67 @@
+{
+    "results": {
+        "ocrbench": {
+            "ocrbench_accuracy,none": 0.379,
+            "ocrbench_accuracy_stderr,none": "N/A",
+            "alias": "ocrbench"
+        }
+    },
+    "configs": {
+        "ocrbench": {
+            "task": "ocrbench",
+            "dataset_path": "echo840/OCRBench",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "test",
+            "doc_to_visual": "<function ocrbench_doc_to_visual at 0x7f1f035f7dc0>",
+            "doc_to_text": "<function ocrbench_doc_to_text at 0x7f1f02df03a0>",
+            "doc_to_target": "answer",
+            "process_results": "<function ocrbench_process_results at 0x7f1f02df0700>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "ocrbench_accuracy",
+                    "aggregation": "<function ocrbench_aggregate_accuracy at 0x7f1f02df0a60>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 128,
+                "temperature": 0.0,
+                "top_p": 1.0,
+                "num_beams": 1,
+                "do_sample": false,
+                "until": [
+                    "\n\n"
+                ]
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "metadata": [
+                {
+                    "version": 0.0
+                }
+            ]
+        }
+    },
+    "versions": {
+        "ocrbench": "Yaml"
+    },
+    "n-shot": {
+        "ocrbench": 0
+    },
+    "model_configs": {
+        "model": "llava",
+        "model_args": "pretrained=/cm/archive/namnv78/checkpoints/Xphi35-siglip224/SMOE/1M3/Full_xmoe/checkpoint-13786,conv_template=phi35",
+        "batch_size": "1",
+        "device": null,
+        "limit": null,
+        "bootstrap_iters": 100000,
+        "gen_kwargs": ""
+    },
+    "git_hash": "289c7fe5"
+}

sft/1M3/Full_xmoe/checkpoint-13786/logs/0628_2145_llava...bench_llava_model_args_467343/results/ocrbench_results.txt

@@ -0,0 +1,18 @@
+######################### OCRBench #############################
+Text Recognition(Total 300): 206
+---------------- Details of Recognition Score ------------------
+Regular Text Recognition(Total 50): 42
+Irregular Text Recognition(Total 50): 44
+Artistic Text Recognition(Total 50): 45
+Handwriting Recognition(Total 50): 19
+Digit String Recognition(Total 50): 32
+Non-Semantic Text Recognition(Total 50): 24
+----------------------------------------------------------------
+Scene Text-centric VQA(Total 200): 108
+----------------------------------------------------------------
+Doc-oriented VQA(Total 200): 20
+----------------------------------------------------------------
+Key Information Extraction(Total 200): 16
+Handwritten Mathematical Expression Recognition(Total 100): 29
+--------------------- Final Score ------------------------------
+Final Score(Total 1000): 379