dddc916d0af8ce5501874b0e9bcdd410372d15780b8254e327e14723fd9da0d7

.gitattributes

@@ -33,3 +33,8 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1803_llava..._pope_llava_model_args_7bd459/mmerealworld_lite.json filter=lfs diff=lfs merge=lfs -text
+sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1803_llava..._pope_llava_model_args_7bd459/pope.json filter=lfs diff=lfs merge=lfs -text
+sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1803_llava..._pope_llava_model_args_7bd459/textvqa_val.json filter=lfs diff=lfs merge=lfs -text
+sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1833_llava...mstar_llava_model_args_7bd459/mmbench_en_dev.json filter=lfs diff=lfs merge=lfs -text
+sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1833_llava...mstar_llava_model_args_7bd459/submissions/mmbench_en_dev_results.xlsx filter=lfs diff=lfs merge=lfs -text

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1803_llava..._pope_llava_model_args_7bd459/mmerealworld_lite.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b26e5e78b76b521492611ba5cfdc56ad379e4dfc7c5a5930f496708bdc63556
+size 1994104210

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1803_llava..._pope_llava_model_args_7bd459/pope.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4daf799d1b7ec517a5af1073d1049ec7bee5b95cdd694cf06778b0d0efbbfa9f
+size 17577558

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1803_llava..._pope_llava_model_args_7bd459/rank0_metric_eval_done.txt

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

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1803_llava..._pope_llava_model_args_7bd459/rank1_metric_eval_done.txt

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

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1803_llava..._pope_llava_model_args_7bd459/rank2_metric_eval_done.txt

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

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1803_llava..._pope_llava_model_args_7bd459/rank3_metric_eval_done.txt

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

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1803_llava..._pope_llava_model_args_7bd459/results.json

@@ -0,0 +1,257 @@
+{
+    "results": {
+        "mmerealworld_lite": {
+            "mme_realworld_score,none": 0.3324648254299114,
+            "mme_realworld_score_stderr,none": "N/A",
+            "alias": "mmerealworld_lite"
+        },
+        "ocrbench": {
+            "ocrbench_accuracy,none": 0.319,
+            "ocrbench_accuracy_stderr,none": "N/A",
+            "alias": "ocrbench"
+        },
+        "pope": {
+            "pope_accuracy,none": 0.8682222222222222,
+            "pope_accuracy_stderr,none": "N/A",
+            "pope_precision,none": 0.9297199170124482,
+            "pope_precision_stderr,none": "N/A",
+            "pope_recall,none": 0.7966666666666666,
+            "pope_recall_stderr,none": "N/A",
+            "pope_f1_score,none": 0.8580660603159406,
+            "pope_f1_score_stderr,none": "N/A",
+            "pope_yes_ratio,none": 0.5,
+            "pope_yes_ratio_stderr,none": "N/A",
+            "alias": "pope"
+        },
+        "textvqa_val": {
+            "exact_match,none": 0.41270000022649767,
+            "exact_match_stderr,none": 0.006748425752359104,
+            "submission,none": null,
+            "submission_stderr,none": "N/A",
+            "alias": "textvqa_val"
+        }
+    },
+    "configs": {
+        "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 0x7fe9371bc9d0>",
+            "doc_to_text": "<function mme_realworld_doc_to_text at 0x7fe9371b65e0>",
+            "doc_to_target": "answer",
+            "process_results": "<function mme_realworld_process_results at 0x7fe9371453a0>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "mme_realworld_score",
+                    "aggregation": "<function mme_realworld_aggregate_results at 0x7fe937146040>",
+                    "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"
+                }
+            }
+        },
+        "ocrbench": {
+            "task": "ocrbench",
+            "dataset_path": "echo840/OCRBench",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "test",
+            "doc_to_visual": "<function ocrbench_doc_to_visual at 0x7fe95037f040>",
+            "doc_to_text": "<function ocrbench_doc_to_text at 0x7fe95f79f670>",
+            "doc_to_target": "answer",
+            "process_results": "<function ocrbench_process_results at 0x7fe95f79f9d0>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "ocrbench_accuracy",
+                    "aggregation": "<function ocrbench_aggregate_accuracy at 0x7fe95f79fd30>",
+                    "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
+                }
+            ]
+        },
+        "pope": {
+            "task": "pope",
+            "dataset_path": "lmms-lab/POPE",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "test",
+            "doc_to_visual": "<function pope_doc_to_visual at 0x7fe9371cfd30>",
+            "doc_to_text": "<function pope_doc_to_text at 0x7fe9371d6280>",
+            "doc_to_target": "answer",
+            "process_results": "<function pope_process_results at 0x7fe9371d6790>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "pope_accuracy",
+                    "aggregation": "<function pope_aggregate_accuracy at 0x7fe9371d6ca0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "pope_precision",
+                    "aggregation": "<function pope_aggregate_precision at 0x7fe9371db1f0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "pope_recall",
+                    "aggregation": "<function pope_aggregate_recall at 0x7fe9371db700>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "pope_f1_score",
+                    "aggregation": "<function pope_aggregate_f1_score at 0x7fe9371dbc10>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "pope_yes_ratio",
+                    "aggregation": "<function pope_aggregate_yes_ratio at 0x7fe9371e0160>",
+                    "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
+                }
+            ]
+        },
+        "textvqa_val": {
+            "task": "textvqa_val",
+            "dataset_path": "lmms-lab/textvqa",
+            "test_split": "validation",
+            "doc_to_visual": "<function textvqa_doc_to_visual at 0x7fe96b6a60d0>",
+            "doc_to_text": "<function textvqa_doc_to_text at 0x7fe96b6a6550>",
+            "doc_to_target": "answer",
+            "process_results": "<function textvqa_process_results at 0x7fe96b6a6820>",
+            "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 0x7fe96b708ee0>",
+                    "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": {
+        "mmerealworld_lite": "Yaml",
+        "ocrbench": "Yaml",
+        "pope": "Yaml",
+        "textvqa_val": "Yaml"
+    },
+    "n-shot": {
+        "mmerealworld_lite": 0,
+        "ocrbench": 0,
+        "pope": 0,
+        "textvqa_val": 0
+    },
+    "model_configs": {
+        "model": "llava",
+        "model_args": "pretrained=/cm/archive/namnv78/checkpoints/Xphi35-siglip224/sft/SMOE/Full_smoe/checkpoint-16632,conv_template=phi35",
+        "batch_size": "1",
+        "device": null,
+        "limit": null,
+        "bootstrap_iters": 100000,
+        "gen_kwargs": ""
+    },
+    "git_hash": "bf33b90"
+}

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1803_llava..._pope_llava_model_args_7bd459/results/ocrbench_results.txt

@@ -0,0 +1,18 @@
+######################### OCRBench #############################
+Text Recognition(Total 300): 182
+---------------- Details of Recognition Score ------------------
+Regular Text Recognition(Total 50): 46
+Irregular Text Recognition(Total 50): 42
+Artistic Text Recognition(Total 50): 44
+Handwriting Recognition(Total 50): 26
+Digit String Recognition(Total 50): 11
+Non-Semantic Text Recognition(Total 50): 13
+----------------------------------------------------------------
+Scene Text-centric VQA(Total 200): 107
+----------------------------------------------------------------
+Doc-oriented VQA(Total 200): 22
+----------------------------------------------------------------
+Key Information Extraction(Total 200): 8
+Handwritten Mathematical Expression Recognition(Total 100): 0
+--------------------- Final Score ------------------------------
+Final Score(Total 1000): 319

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1803_llava..._pope_llava_model_args_7bd459/textvqa_val.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02035d62838017ab756a6df07b88060c358076cdf3927fca0f0048c65b88a09d
+size 13150501

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1833_llava...mstar_llava_model_args_7bd459/mmbench_en_dev.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4dcfddcb36e62b42fba763210b96466f64630d467f8192bc4803317fc5f3db0
+size 14568169

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1833_llava...mstar_llava_model_args_7bd459/rank0_metric_eval_done.txt

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

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1833_llava...mstar_llava_model_args_7bd459/rank1_metric_eval_done.txt

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

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1833_llava...mstar_llava_model_args_7bd459/rank2_metric_eval_done.txt

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

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1833_llava...mstar_llava_model_args_7bd459/rank3_metric_eval_done.txt

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

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1833_llava...mstar_llava_model_args_7bd459/results.json

@@ -0,0 +1,298 @@
+{
+    "results": {
+        "ai2d": {
+            "exact_match,flexible-extract": 0.6492875647668394,
+            "exact_match_stderr,flexible-extract": 0.008588667905633653,
+            "alias": "ai2d"
+        },
+        "mmbench_en_dev": {
+            "gpt_eval_score,none": 71.99312714776632,
+            "gpt_eval_score_stderr,none": "N/A",
+            "submission,none": null,
+            "submission_stderr,none": "N/A",
+            "alias": "mmbench_en_dev"
+        },
+        "mmmu_val": {
+            "mmmu_acc,none": 0.42222,
+            "mmmu_acc_stderr,none": "N/A",
+            "alias": "mmmu_val"
+        },
+        "mmstar": {
+            "coarse perception,none": 0.7030779789873819,
+            "coarse perception_stderr,none": "N/A",
+            "fine-grained perception,none": 0.34155612871530333,
+            "fine-grained perception_stderr,none": "N/A",
+            "instance reasoning,none": 0.53201826163044,
+            "instance reasoning_stderr,none": "N/A",
+            "logical reasoning,none": 0.36875764499526875,
+            "logical reasoning_stderr,none": "N/A",
+            "math,none": 0.2920175086875167,
+            "math_stderr,none": "N/A",
+            "science & technology,none": 0.25190144653700547,
+            "science & technology_stderr,none": "N/A",
+            "alias": "mmstar"
+        }
+    },
+    "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 0x7ef90acbe8b0>",
+            "doc_to_text": "<function ai2d_doc_to_text at 0x7ef90acbeb80>",
+            "doc_to_target": "<function ai2d_doc_to_target at 0x7ef90ac48040>",
+            "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"
+            }
+        },
+        "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 0x7ef969d64e50>",
+            "doc_to_text": "<function mmbench_doc_to_text at 0x7ef969d67430>",
+            "doc_to_target": "answer",
+            "process_results": "<function mmbench_process_results at 0x7ef969d679d0>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "gpt_eval_score",
+                    "aggregation": "<function mmbench_aggregate_dev_results_eval at 0x7ef969d645e0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "submission",
+                    "aggregation": "<function mmbench_aggregate_dev_results_submission at 0x7ef969d64b80>",
+                    "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"
+                }
+            }
+        },
+        "mmmu_val": {
+            "task": "mmmu_val",
+            "dataset_path": "lmms-lab/MMMU",
+            "test_split": "validation",
+            "doc_to_visual": "<function mmmu_doc_to_visual at 0x7ef932f079d0>",
+            "doc_to_text": "<function mmmu_doc_to_text at 0x7ef932e1a670>",
+            "doc_to_target": "answer",
+            "process_results": "<function mmmu_process_results at 0x7ef932cf84c0>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "mmmu_acc",
+                    "aggregation": "<function mmmu_aggregate_results at 0x7ef932c263a0>",
+                    "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 0x7ef90ab97a60>",
+            "doc_to_text": "<function mmstar_doc_to_text at 0x7ef90ab97e50>",
+            "doc_to_target": "answer",
+            "process_results": "<function mmstar_process_results at 0x7ef90aba1310>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "coarse perception",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7ef90aba1700>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "fine-grained perception",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7ef90aba1a60>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "instance reasoning",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7ef90aba1dc0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "logical reasoning",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7ef90aba9160>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "science & technology",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7ef90aba94c0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "math",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7ef90aba9820>",
+                    "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"
+                }
+            }
+        }
+    },
+    "versions": {
+        "ai2d": "Yaml",
+        "mmbench_en_dev": "Yaml",
+        "mmmu_val": "Yaml",
+        "mmstar": "Yaml"
+    },
+    "n-shot": {
+        "ai2d": 0,
+        "mmbench_en_dev": 0,
+        "mmmu_val": 0,
+        "mmstar": 0
+    },
+    "model_configs": {
+        "model": "llava",
+        "model_args": "pretrained=/cm/archive/namnv78/checkpoints/Xphi35-siglip224/sft/SMOE/Full_smoe/checkpoint-16632,conv_template=phi35",
+        "batch_size": "1",
+        "device": null,
+        "limit": null,
+        "bootstrap_iters": 100000,
+        "gen_kwargs": ""
+    },
+    "git_hash": "bf33b90"
+}

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1833_llava...mstar_llava_model_args_7bd459/submissions/mmbench_en_dev_results.json

@@ -0,0 +1 @@
+{"overall_acc": 0.7199312714776632, "category_acc": {"action_recognition": 0.9259259259259259, "attribute_comparison": 0.6818181818181818, "attribute_recognition": 0.8648648648648649, "celebrity_recognition": 0.8080808080808081, "function_reasoning": 0.8734177215189873, "future_prediction": 0.5, "identity_reasoning": 0.9777777777777777, "image_emotion": 0.84, "image_quality": 0.5849056603773585, "image_scene": 0.9423076923076923, "image_style": 0.7924528301886793, "image_topic": 0.8888888888888888, "nature_relation": 0.5208333333333334, "object_localization": 0.5432098765432098, "ocr": 0.6666666666666666, "physical_property_reasoning": 0.5866666666666667, "physical_relation": 0.4583333333333333, "social_relation": 0.8372093023255814, "spatial_relationship": 0.3111111111111111, "structuralized_imagetext_understanding": 0.46153846153846156}, "l2_category_acc": {"attribute_reasoning": 0.7889447236180904, "coarse_perception": 0.8277027027027027, "finegrained_perception (cross-instance)": 0.6573426573426573, "finegrained_perception (instance-level)": 0.7303754266211604, "logic_reasoning": 0.4745762711864407, "relation_reasoning": 0.6260869565217392}}

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1833_llava...mstar_llava_model_args_7bd459/submissions/mmbench_en_dev_results.xlsx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8e1bf82d81bdc3d83cc5a496f96c0d8c23b646e46eebc33d1afa187aa952791
+size 865173

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1843_llava...image_llava_model_args_7bd459/rank1_metric_eval_done.txt

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

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1843_llava...image_llava_model_args_7bd459/rank2_metric_eval_done.txt

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

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1843_llava...image_llava_model_args_7bd459/rank3_metric_eval_done.txt

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

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1848_llava...tmini_llava_model_args_7bd459/rank0_metric_eval_done.txt

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

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1848_llava...tmini_llava_model_args_7bd459/rank1_metric_eval_done.txt

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

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1848_llava...tmini_llava_model_args_7bd459/rank2_metric_eval_done.txt

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

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1848_llava...tmini_llava_model_args_7bd459/rank3_metric_eval_done.txt

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

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1848_llava...tmini_llava_model_args_7bd459/results.json

@@ -0,0 +1,79 @@
+{
+    "results": {
+        "mathvista_testmini": {
+            "gpt_eval_score,none": 31.9,
+            "gpt_eval_score_stderr,none": "N/A",
+            "alias": "mathvista_testmini"
+        }
+    },
+    "configs": {
+        "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 0x7facdb3ecc10>",
+            "doc_to_text": "<function mathvista_doc_to_text at 0x7facdb3f8310>",
+            "doc_to_target": "answer",
+            "process_results": "<function mathvista_process_results at 0x7facdb4029d0>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "gpt_eval_score",
+                    "aggregation": "<function mathvista_aggregate_results at 0x7facdaca90d0>",
+                    "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": {
+        "mathvista_testmini": "Yaml"
+    },
+    "n-shot": {
+        "mathvista_testmini": 0
+    },
+    "model_configs": {
+        "model": "llava",
+        "model_args": "pretrained=/cm/archive/namnv78/checkpoints/Xphi35-siglip224/sft/SMOE/Full_smoe/checkpoint-16632,conv_template=phi35",
+        "batch_size": "1",
+        "device": null,
+        "limit": null,
+        "bootstrap_iters": 100000,
+        "gen_kwargs": ""
+    },
+    "git_hash": "bf33b90"
+}

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1908_llava...image_llava_model_args_7bd459/rank1_metric_eval_done.txt

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

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1908_llava...image_llava_model_args_7bd459/rank2_metric_eval_done.txt

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

sft/smoe_refine_665k_llava/checkpoint-16632/logs/0409_1908_llava...image_llava_model_args_7bd459/rank3_metric_eval_done.txt

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