Upload folder using huggingface_hub

.gitattributes

@@ -202,3 +202,10 @@ sft/1M3/revise_Full_remoe/logs/0626_1437_llava_v1.5_gqa_llava_model_args_179bff/
 sft/1M3/revise_Full_remoe/checkpoint-27572/trainer_state.json filter=lfs diff=lfs merge=lfs -text
 sft/1M3/revise_Full_remoe/checkpoint-34462/trainer_state.json filter=lfs diff=lfs merge=lfs -text
 sft/1M3/revise_Full_remoe/checkpoint-20679/trainer_state.json filter=lfs diff=lfs merge=lfs -text
+sft/1M3/Full_smoe_tcmoe/logs/0626_0357_llava..._pope_llava_model_args_59313d/mmbench_en_dev.json filter=lfs diff=lfs merge=lfs -text
+sft/1M3/Full_smoe_tcmoe/logs/0626_0357_llava..._pope_llava_model_args_59313d/mmerealworld_lite.json filter=lfs diff=lfs merge=lfs -text
+sft/1M3/Full_smoe_tcmoe/logs/0626_0357_llava..._pope_llava_model_args_59313d/pope.json filter=lfs diff=lfs merge=lfs -text
+sft/1M3/Full_smoe_tcmoe/logs/0626_0357_llava..._pope_llava_model_args_59313d/submissions/mmbench_en_dev_results.xlsx filter=lfs diff=lfs merge=lfs -text
+sft/1M3/Full_smoe_tcmoe/logs/0626_0357_llava..._pope_llava_model_args_59313d/textvqa_val.json filter=lfs diff=lfs merge=lfs -text
+sft/1M3/Full_smoe_tcmoe/logs/0626_0430_llava...a_val_llava_model_args_59313d/infovqa_val.json filter=lfs diff=lfs merge=lfs -text
+sft/1M3/Full_smoe_tcmoe/logs/0626_1139_llava_v1.5_gqa_llava_model_args_59313d/gqa.json filter=lfs diff=lfs merge=lfs -text

sft/1M3/Full_smoe_tcmoe/logs/0626_0357_llava..._pope_llava_model_args_59313d/mmbench_en_dev.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d380c3f002ec948ee21ee60010216ba74afd3294c59b7157e0b3c91e2b1f6c24
+size 14568182

sft/1M3/Full_smoe_tcmoe/logs/0626_0357_llava..._pope_llava_model_args_59313d/mmerealworld_lite.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:685b6be963788b6a14951cba2339ca3d109bc4802d7f516f3b7ad2215daaec4d
+size 1994104215

sft/1M3/Full_smoe_tcmoe/logs/0626_0357_llava..._pope_llava_model_args_59313d/pope.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3292cb696bd3bdc7d08e6f50e1abd53b27659af4ad57fdfeaabd83b900709339
+size 17593236

sft/1M3/Full_smoe_tcmoe/logs/0626_0357_llava..._pope_llava_model_args_59313d/rank0_metric_eval_done.txt

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

sft/1M3/Full_smoe_tcmoe/logs/0626_0357_llava..._pope_llava_model_args_59313d/rank1_metric_eval_done.txt

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

sft/1M3/Full_smoe_tcmoe/logs/0626_0357_llava..._pope_llava_model_args_59313d/rank2_metric_eval_done.txt

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

sft/1M3/Full_smoe_tcmoe/logs/0626_0357_llava..._pope_llava_model_args_59313d/rank3_metric_eval_done.txt

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

sft/1M3/Full_smoe_tcmoe/logs/0626_0357_llava..._pope_llava_model_args_59313d/results.json

@@ -0,0 +1,488 @@
+{
+    "results": {
+        "ai2d": {
+            "exact_match,flexible-extract": 0.6784326424870466,
+            "exact_match_stderr,flexible-extract": 0.00840661254403524,
+            "alias": "ai2d"
+        },
+        "mmbench_en_dev": {
+            "gpt_eval_score,none": 70.10309278350515,
+            "gpt_eval_score_stderr,none": "N/A",
+            "submission,none": null,
+            "submission_stderr,none": "N/A",
+            "alias": "mmbench_en_dev"
+        },
+        "mmerealworld_lite": {
+            "mme_realworld_score,none": 0.31005732152162585,
+            "mme_realworld_score_stderr,none": "N/A",
+            "alias": "mmerealworld_lite"
+        },
+        "mmmu_val": {
+            "mmmu_acc,none": 0.42889,
+            "mmmu_acc_stderr,none": "N/A",
+            "alias": "mmmu_val"
+        },
+        "mmstar": {
+            "coarse perception,none": 0.6904363843279423,
+            "coarse perception_stderr,none": "N/A",
+            "fine-grained perception,none": 0.31350405305821666,
+            "fine-grained perception_stderr,none": "N/A",
+            "instance reasoning,none": 0.5281045320914839,
+            "instance reasoning_stderr,none": "N/A",
+            "logical reasoning,none": 0.3833729526798834,
+            "logical reasoning_stderr,none": "N/A",
+            "math,none": 0.42606700525706137,
+            "math_stderr,none": "N/A",
+            "science & technology,none": 0.26584995173646053,
+            "science & technology_stderr,none": "N/A",
+            "alias": "mmstar"
+        },
+        "pope": {
+            "pope_accuracy,none": 0.6772222222222222,
+            "pope_accuracy_stderr,none": "N/A",
+            "pope_precision,none": 0.8657068680670286,
+            "pope_precision_stderr,none": "N/A",
+            "pope_recall,none": 0.8612350316975816,
+            "pope_recall_stderr,none": "N/A",
+            "pope_f1_score,none": 0.8634651600753297,
+            "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": 0.37152000020742415,
+            "exact_match_stderr,none": 0.0066345294644140785,
+            "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 0x7fc147970c10>",
+            "doc_to_text": "<function ai2d_doc_to_text at 0x7fc147970ee0>",
+            "doc_to_target": "<function ai2d_doc_to_target at 0x7fc1479784c0>",
+            "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 0x7fc1a6bc34c0>",
+            "doc_to_text": "<function mmbench_doc_to_text at 0x7fc1a6bc39d0>",
+            "doc_to_target": "answer",
+            "process_results": "<function mmbench_process_results at 0x7fc1a6bc3ee0>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "gpt_eval_score",
+                    "aggregation": "<function mmbench_aggregate_dev_results_eval at 0x7fc1a6c38d30>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "submission",
+                    "aggregation": "<function mmbench_aggregate_dev_results_submission at 0x7fc1a6bc3280>",
+                    "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"
+                }
+            }
+        },
+        "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 0x7fc148119dc0>",
+            "doc_to_text": "<function mme_realworld_doc_to_text at 0x7fc14811e8b0>",
+            "doc_to_target": "answer",
+            "process_results": "<function mme_realworld_process_results at 0x7fc148124550>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "mme_realworld_score",
+                    "aggregation": "<function mme_realworld_aggregate_results at 0x7fc1481280d0>",
+                    "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 0x7fc15a5dd820>",
+            "doc_to_text": "<function mmmu_doc_to_text at 0x7fc15a4f2550>",
+            "doc_to_target": "answer",
+            "process_results": "<function mmmu_process_results at 0x7fc15a416430>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "mmmu_acc",
+                    "aggregation": "<function mmmu_aggregate_results at 0x7fc15a32e3a0>",
+                    "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 0x7fc1478d0280>",
+            "doc_to_text": "<function mmstar_doc_to_text at 0x7fc1478d0700>",
+            "doc_to_target": "answer",
+            "process_results": "<function mmstar_process_results at 0x7fc1478d0c10>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "coarse perception",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7fc1478d70d0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "fine-grained perception",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7fc1478d74c0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "instance reasoning",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7fc1478d78b0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "logical reasoning",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7fc1478d7ca0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "science & technology",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7fc1478df0d0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "math",
+                    "aggregation": "<function mmstar_aggregate_results at 0x7fc1478df4c0>",
+                    "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"
+                }
+            }
+        },
+        "pope": {
+            "task": "pope",
+            "dataset_path": "lmms-lab/POPE",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "test",
+            "doc_to_visual": "<function pope_doc_to_visual at 0x7fc1481b38b0>",
+            "doc_to_text": "<function pope_doc_to_text at 0x7fc1481b3e50>",
+            "doc_to_target": "answer",
+            "process_results": "<function pope_process_results at 0x7fc1481b9430>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "pope_accuracy",
+                    "aggregation": "<function pope_aggregate_accuracy at 0x7fc1481b99d0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "pope_precision",
+                    "aggregation": "<function pope_aggregate_precision at 0x7fc1481b9f70>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "pope_recall",
+                    "aggregation": "<function pope_aggregate_recall at 0x7fc1481bc550>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "pope_f1_score",
+                    "aggregation": "<function pope_aggregate_f1_score at 0x7fc1481bcaf0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "pope_yes_ratio",
+                    "aggregation": "<function pope_aggregate_yes_ratio at 0x7fc1481be0d0>",
+                    "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 0x7fc1740658b0>",
+            "doc_to_text": "<function textvqa_doc_to_text at 0x7fc174065d30>",
+            "doc_to_target": "answer",
+            "process_results": "<function textvqa_process_results at 0x7fc17401c040>",
+            "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 0x7fc174065670>",
+                    "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",
+        "mmbench_en_dev": "Yaml",
+        "mmerealworld_lite": "Yaml",
+        "mmmu_val": "Yaml",
+        "mmstar": "Yaml",
+        "pope": "Yaml",
+        "textvqa_val": "Yaml"
+    },
+    "n-shot": {
+        "ai2d": 0,
+        "mmbench_en_dev": 0,
+        "mmerealworld_lite": 0,
+        "mmmu_val": 0,
+        "mmstar": 0,
+        "pope": 0,
+        "textvqa_val": 0
+    },
+    "model_configs": {
+        "model": "llava",
+        "model_args": "pretrained=/cm/archive/namnv78/checkpoints/Xphi35-siglip224/SMOE/1M3/Full_smoe_tcmoe,conv_template=phi35",
+        "batch_size": "1",
+        "device": null,
+        "limit": null,
+        "bootstrap_iters": 100000,
+        "gen_kwargs": ""
+    },
+    "git_hash": "289c7fe5"
+}

sft/1M3/Full_smoe_tcmoe/logs/0626_0357_llava..._pope_llava_model_args_59313d/submissions/mmbench_en_dev_results.json

@@ -0,0 +1 @@
+{"overall_acc": 0.7010309278350515, "category_acc": {"action_recognition": 0.8703703703703703, "attribute_comparison": 0.6363636363636364, "attribute_recognition": 0.8648648648648649, "celebrity_recognition": 0.7373737373737373, "function_reasoning": 0.8860759493670886, "future_prediction": 0.575, "identity_reasoning": 0.9777777777777777, "image_emotion": 0.8, "image_quality": 0.3584905660377358, "image_scene": 0.9423076923076923, "image_style": 0.9433962264150944, "image_topic": 0.8888888888888888, "nature_relation": 0.6666666666666666, "object_localization": 0.3950617283950617, "ocr": 0.717948717948718, "physical_property_reasoning": 0.4533333333333333, "physical_relation": 0.5833333333333334, "social_relation": 0.8372093023255814, "spatial_relationship": 0.2222222222222222, "structuralized_imagetext_understanding": 0.5384615384615384}, "l2_category_acc": {"attribute_reasoning": 0.7437185929648241, "coarse_perception": 0.8074324324324325, "finegrained_perception (cross-instance)": 0.5944055944055944, "finegrained_perception (instance-level)": 0.6723549488054608, "logic_reasoning": 0.5508474576271186, "relation_reasoning": 0.7130434782608696}}

sft/1M3/Full_smoe_tcmoe/logs/0626_0357_llava..._pope_llava_model_args_59313d/submissions/mmbench_en_dev_results.xlsx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:745d85523cf69e3b24c40c5541afd98c2264f6db90fdc753189e3fc074950446
+size 865055

sft/1M3/Full_smoe_tcmoe/logs/0626_0357_llava..._pope_llava_model_args_59313d/textvqa_val.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:087f877a71b7fa8f8206e68cfb6549dc4d3a061058d1538e7950d34116fabf44
+size 13200758

sft/1M3/Full_smoe_tcmoe/logs/0626_0419_llava...l_mme_llava_model_args_59313d/rank0_metric_eval_done.txt

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

sft/1M3/Full_smoe_tcmoe/logs/0626_0419_llava...l_mme_llava_model_args_59313d/rank1_metric_eval_done.txt

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

sft/1M3/Full_smoe_tcmoe/logs/0626_0419_llava...l_mme_llava_model_args_59313d/rank2_metric_eval_done.txt

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

sft/1M3/Full_smoe_tcmoe/logs/0626_0419_llava...l_mme_llava_model_args_59313d/rank3_metric_eval_done.txt

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

sft/1M3/Full_smoe_tcmoe/logs/0626_0419_llava...l_mme_llava_model_args_59313d/results.json

@@ -0,0 +1,215 @@
+{
+    "results": {
+        "docvqa_val": {
+            "anls,none": 0.2570573939054029,
+            "anls_stderr,none": 0.005975812787495763,
+            "alias": "docvqa_val"
+        },
+        "mme": {
+            "mme_cognition_score,none": 347.8571428571429,
+            "mme_cognition_score_stderr,none": "N/A",
+            "mme_percetion_score,none": 1235.8730492196878,
+            "mme_percetion_score_stderr,none": "N/A",
+            "alias": "mme"
+        },
+        "scienceqa_img": {
+            "exact_match,none": 0.7987109568666336,
+            "exact_match_stderr,none": 0.008930167911772436,
+            "alias": "scienceqa_img"
+        }
+    },
+    "configs": {
+        "docvqa_val": {
+            "task": "docvqa_val",
+            "dataset_path": "lmms-lab/DocVQA",
+            "dataset_name": "DocVQA",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "validation",
+            "doc_to_visual": "<function docvqa_doc_to_visual at 0x7f2a812c14c0>",
+            "doc_to_text": "<function docvqa_doc_to_text at 0x7f2a812c1820>",
+            "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."
+                },
+                "qwen_vl": {
+                    "pre_prompt": "",
+                    "post_prompt": " Answer:"
+                }
+            }
+        },
+        "mme": {
+            "task": "mme",
+            "dataset_path": "lmms-lab/MME",
+            "dataset_kwargs": {
+                "token": false
+            },
+            "test_split": "test",
+            "doc_to_visual": "<function mme_doc_to_visual at 0x7f2aab92a8b0>",
+            "doc_to_text": "<function mme_doc_to_text at 0x7f2aa9ecc160>",
+            "doc_to_target": "answer",
+            "process_results": "<function mme_process_results at 0x7f2aa9ecc700>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "mme_percetion_score",
+                    "aggregation": "<function mme_aggregate_results at 0x7f2aa9eccc10>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "mme_cognition_score",
+                    "aggregation": "<function mme_aggregate_results at 0x7f2aa9ed40d0>",
+                    "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"
+                }
+            }
+        },
+        "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 0x7f2a75c8ac10>",
+            "doc_to_text": "<function sqa_doc_to_text at 0x7f2a75c8ae50>",
+            "doc_to_target": "<function sqa_doc_to_target at 0x7f2a75ada280>",
+            "process_results": "<function sqa_process_results at 0x7f2a75ada5e0>",
+            "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"
+                }
+            }
+        }
+    },
+    "versions": {
+        "docvqa_val": "Yaml",
+        "mme": "Yaml",
+        "scienceqa_img": "Yaml"
+    },
+    "n-shot": {
+        "docvqa_val": 0,
+        "mme": 0,
+        "scienceqa_img": 0
+    },
+    "model_configs": {
+        "model": "llava",
+        "model_args": "pretrained=/cm/archive/namnv78/checkpoints/Xphi35-siglip224/SMOE/1M3/Full_smoe_tcmoe,conv_template=phi35",
+        "batch_size": "1",
+        "device": null,
+        "limit": null,
+        "bootstrap_iters": 100000,
+        "gen_kwargs": ""
+    },
+    "git_hash": "289c7fe5"
+}

sft/1M3/Full_smoe_tcmoe/logs/0626_0430_llava...a_val_llava_model_args_59313d/infovqa_val.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dfa57a8969ea335579f36927fa1b04d59c4ada57f64ee5c223cf86770a85275d
+size 576440094

sft/1M3/Full_smoe_tcmoe/logs/0626_0430_llava...a_val_llava_model_args_59313d/rank0_metric_eval_done.txt

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

sft/1M3/Full_smoe_tcmoe/logs/0626_0430_llava...a_val_llava_model_args_59313d/rank1_metric_eval_done.txt

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

sft/1M3/Full_smoe_tcmoe/logs/0626_0430_llava...a_val_llava_model_args_59313d/rank2_metric_eval_done.txt

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

sft/1M3/Full_smoe_tcmoe/logs/0626_0430_llava...a_val_llava_model_args_59313d/rank3_metric_eval_done.txt

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

sft/1M3/Full_smoe_tcmoe/logs/0626_0430_llava...a_val_llava_model_args_59313d/results.json

@@ -0,0 +1,245 @@
+{
+    "results": {
+        "infovqa_val": {
+            "anls,none": 0.2995358800428418,
+            "anls_stderr,none": 0.00865642075096775,
+            "alias": "infovqa_val"
+        },
+        "mmmu_pro": {
+            "mmmu_acc,none": 0.18931,
+            "mmmu_acc_stderr,none": 0.03685532629908108,
+            "alias": "mmmu_pro"
+        },
+        "mmmu_pro_standard": {
+            "mmmu_acc,none": 0.26301,
+            "mmmu_acc_stderr,none": "N/A",
+            "alias": " - mmmu_pro_standard"
+        },
+        "mmmu_pro_vision": {
+            "mmmu_acc,none": 0.11561,
+            "mmmu_acc_stderr,none": "N/A",
+            "alias": " - mmmu_pro_vision"
+        },
+        "realworldqa": {
+            "exact_match,flexible-extract": 0.515032679738562,
+            "exact_match_stderr,flexible-extract": 0.018081187561622823,
+            "alias": "realworldqa"
+        }
+    },
+    "groups": {
+        "mmmu_pro": {
+            "mmmu_acc,none": 0.18931,
+            "mmmu_acc_stderr,none": 0.03685532629908108,
+            "alias": "mmmu_pro"
+        }
+    },
+    "configs": {
+        "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 0x7fc9fa5d69d0>",
+            "doc_to_text": "<function infovqa_doc_to_text at 0x7fc9fa5d6ca0>",
+            "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."
+                }
+            }
+        },
+        "mmmu_pro_standard": {
+            "task": "mmmu_pro_standard",
+            "dataset_path": "MMMU/MMMU_Pro",
+            "dataset_name": "standard (10 options)",
+            "test_split": "test",
+            "doc_to_visual": "<function mmmu_pro_doc_to_visual at 0x7fca25805790>",
+            "doc_to_text": "<function mmmu_pro_doc_to_text at 0x7fca25810550>",
+            "doc_to_target": "{{answer}}",
+            "process_results": "<function mmmu_pro_process_results at 0x7fca258194c0>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "mmmu_acc",
+                    "aggregation": "<function mmmu_pro_aggregate_results at 0x7fca25823430>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 256,
+                "until": [
+                    "\n\n"
+                ]
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "metadata": {
+                "version": 0.0,
+                "interleaved_format": false
+            },
+            "model_specific_prompt_kwargs": {
+                "default": {
+                    "pre_prompt": "",
+                    "post_prompt": "Answer with the option letter from the given choices directly."
+                }
+            }
+        },
+        "mmmu_pro_vision": {
+            "task": "mmmu_pro_vision",
+            "dataset_path": "MMMU/MMMU_Pro",
+            "dataset_name": "vision",
+            "test_split": "test",
+            "doc_to_visual": "<function mmmu_pro_doc_to_visual at 0x7fca257f2ca0>",
+            "doc_to_text": "Answer with the option letter from the given choices directly.",
+            "doc_to_target": "{{answer}}",
+            "process_results": "<function mmmu_pro_process_results at 0x7fca257f9b80>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "mmmu_acc",
+                    "aggregation": "<function mmmu_pro_aggregate_results at 0x7fca25800af0>",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 256,
+                "until": [
+                    "\n\n"
+                ]
+            },
+            "repeats": 1,
+            "should_decontaminate": false,
+            "metadata": {
+                "version": 0.0,
+                "interleaved_format": false
+            }
+        },
+        "realworldqa": {
+            "task": "realworldqa",
+            "dataset_path": "lmms-lab/RealWorldQA",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "test",
+            "doc_to_visual": "<function realworldqa_doc_to_visual at 0x7fc9f1671a60>",
+            "doc_to_text": "<function realworldqa_doc_to_text at 0x7fc9f168a0d0>",
+            "doc_to_target": "answer",
+            "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,
+                "top_p": 1.0,
+                "num_beams": 1,
+                "do_sample": false,
+                "until": [
+                    "\n\n"
+                ]
+            },
+            "repeats": 1,
+            "filter_list": [
+                {
+                    "name": "flexible-extract",
+                    "filter": [
+                        {
+                            "function": "<class 'utils.NumberWordsToDigitsFilter'>"
+                        },
+                        {
+                            "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": {
+                    "pre_prompt": "",
+                    "post_prompt": ""
+                },
+                "gpt4v": {
+                    "pre_prompt": "",
+                    "post_prompt": ""
+                },
+                "xcomposer2_4khd": {
+                    "pre_prompt": "[UNUSED_TOKEN_146]user\nQuestion: ",
+                    "post_prompt": "[UNUSED_TOKEN_145]\n[UNUSED_TOKEN_146]assistant\nThe answer is"
+                }
+            }
+        }
+    },
+    "versions": {
+        "infovqa_val": "Yaml",
+        "mmmu_pro": "N/A",
+        "mmmu_pro_standard": "Yaml",
+        "mmmu_pro_vision": "Yaml",
+        "realworldqa": "Yaml"
+    },
+    "n-shot": {
+        "infovqa_val": 0,
+        "mmmu_pro": 0,
+        "mmmu_pro_standard": 0,
+        "mmmu_pro_vision": 0,
+        "realworldqa": 0
+    },
+    "model_configs": {
+        "model": "llava",
+        "model_args": "pretrained=/cm/archive/namnv78/checkpoints/Xphi35-siglip224/SMOE/1M3/Full_smoe_tcmoe,conv_template=phi35",
+        "batch_size": "1",
+        "device": null,
+        "limit": null,
+        "bootstrap_iters": 100000,
+        "gen_kwargs": ""
+    },
+    "git_hash": "289c7fe5"
+}

sft/1M3/Full_smoe_tcmoe/logs/0626_1139_llava_v1.5_gqa_llava_model_args_59313d/gqa.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dcc4c5e650a5f87ecc3e3934c40f696d319a0a49ed56237d9b4a66604c28b234
+size 38569629

sft/1M3/Full_smoe_tcmoe/logs/0626_1139_llava_v1.5_gqa_llava_model_args_59313d/rank0_metric_eval_done.txt

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

sft/1M3/Full_smoe_tcmoe/logs/0626_1139_llava_v1.5_gqa_llava_model_args_59313d/results.json

@@ -0,0 +1,79 @@
+{
+    "results": {
+        "gqa": {
+            "exact_match,none": 0.3945778343138814,
+            "exact_match_stderr,none": 0.004358198135704839,
+            "alias": "gqa"
+        }
+    },
+    "configs": {
+        "gqa": {
+            "task": "gqa",
+            "dataset_path": "lmms-lab/GQA",
+            "dataset_name": "testdev_balanced_instructions",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "testdev",
+            "doc_to_visual": "<function gqa_doc_to_visual at 0x7ff78ec2f8b0>",
+            "doc_to_text": "<function gqa_doc_to_text at 0x7ff78eb83b80>",
+            "doc_to_target": "answer",
+            "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,
+                "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."
+                },
+                "qwen_vl": {
+                    "pre_prompt": "",
+                    "post_prompt": " Answer:"
+                }
+            }
+        }
+    },
+    "versions": {
+        "gqa": "Yaml"
+    },
+    "n-shot": {
+        "gqa": 0
+    },
+    "model_configs": {
+        "model": "llava",
+        "model_args": "pretrained=/cm/archive/namnv78/checkpoints/Xphi35-siglip224/SMOE/1M3/Full_smoe_tcmoe,conv_template=phi35",
+        "batch_size": "1",
+        "device": null,
+        "limit": null,
+        "bootstrap_iters": 100000,
+        "gen_kwargs": ""
+    },
+    "git_hash": "289c7fe5"
+}

sft/1M3/Full_smoe_tcmoe/logs/0626_1211_llava...image_llava_model_args_59313d/rank0_metric_eval_done.txt

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

sft/1M3/Full_smoe_tcmoe/logs/0626_1211_llava...image_llava_model_args_59313d/rank1_metric_eval_done.txt

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

sft/1M3/Full_smoe_tcmoe/logs/0626_1211_llava...image_llava_model_args_59313d/results.json

@@ -0,0 +1,146 @@
+{
+    "results": {
+        "hallusion_bench_image": {
+            "aAcc,none": 45.0053,
+            "aAcc_stderr,none": "N/A",
+            "fAcc,none": 17.6301,
+            "fAcc_stderr,none": "N/A",
+            "qAcc,none": 14.9451,
+            "qAcc_stderr,none": "N/A",
+            "alias": "hallusion_bench_image"
+        },
+        "mathvista_testmini": {
+            "gpt_eval_score,none": 37.8,
+            "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 0x7fdbc65ef0d0>",
+            "doc_to_text": "<function hb_doc_to_text at 0x7fdbc65ef820>",
+            "doc_to_target": "gt_answer_details",
+            "process_results": "<function hb_process_results at 0x7fdbc6563040>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "aAcc",
+                    "aggregation": "<function hb_aggregation_result_aAcc at 0x7fdbc65639d0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "qAcc",
+                    "aggregation": "<function hb_aggregation_result_qAcc at 0x7fdbc6467040>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "fAcc",
+                    "aggregation": "<function hb_aggregation_result_fAcc at 0x7fdbc6467820>",
+                    "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 0x7fdb9bbb0af0>",
+            "doc_to_text": "<function mathvista_doc_to_text at 0x7fdb9b7a91f0>",
+            "doc_to_target": "answer",
+            "process_results": "<function mathvista_process_results at 0x7fdb9b7b18b0>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "gpt_eval_score",
+                    "aggregation": "<function mathvista_aggregate_results at 0x7fdb9b7bbf70>",
+                    "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_smoe_tcmoe,conv_template=phi35",
+        "batch_size": "1",
+        "device": null,
+        "limit": null,
+        "bootstrap_iters": 100000,
+        "gen_kwargs": ""
+    },
+    "git_hash": "289c7fe5"
+}

sft/1M3/Full_smoe_tcmoe/logs/0626_1233_llava..._plus_llava_model_args_59313d/rank0_metric_eval_done.txt

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

sft/1M3/Full_smoe_tcmoe/logs/0626_1233_llava..._plus_llava_model_args_59313d/rank1_metric_eval_done.txt

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

sft/1M3/Full_smoe_tcmoe/logs/0626_1233_llava..._plus_llava_model_args_59313d/results.json

@@ -0,0 +1,168 @@
+{
+    "results": {
+        "chartqa": {
+            "relaxed_overall,none": 0.1644,
+            "relaxed_overall_stderr,none": 0.007414246045132662,
+            "relaxed_human_split,none": 0.1864,
+            "relaxed_human_split_stderr,none": 0.011019127412601738,
+            "relaxed_augmented_split,none": 0.1424,
+            "relaxed_augmented_split_stderr,none": 0.009888175403493184,
+            "alias": "chartqa"
+        },
+        "seedbench_2_plus": {
+            "seedbench_2_plus_Chart,none": 0.5049382716049383,
+            "seedbench_2_plus_Chart_stderr,none": "N/A",
+            "seedbench_2_plus_all,none": 0.5072463768115942,
+            "seedbench_2_plus_all_stderr,none": "N/A",
+            "seedbench_2_plus_Web,none": 0.5545454545454546,
+            "seedbench_2_plus_Web_stderr,none": "N/A",
+            "seedbench_2_plus_Map,none": 0.4708798017348203,
+            "seedbench_2_plus_Map_stderr,none": "N/A",
+            "alias": "seedbench_2_plus"
+        }
+    },
+    "configs": {
+        "chartqa": {
+            "task": "chartqa",
+            "dataset_path": "lmms-lab/ChartQA",
+            "dataset_kwargs": {
+                "token": true
+            },
+            "test_split": "test",
+            "doc_to_visual": "<function chartqa_doc_to_visual at 0x7fa788965d30>",
+            "doc_to_text": "<function chartqa_doc_to_text at 0x7fa78896c790>",
+            "doc_to_target": "answer",
+            "process_results": "<function chartqa_process_results at 0x7fa78896ca60>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "relaxed_overall",
+                    "aggregation": "mean",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "relaxed_human_split",
+                    "aggregation": "mean",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "relaxed_augmented_split",
+                    "aggregation": "mean",
+                    "higher_is_better": true
+                }
+            ],
+            "output_type": "generate_until",
+            "generation_kwargs": {
+                "max_new_tokens": 16,
+                "temperature": 0.0,
+                "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 with a single word."
+                },
+                "qwen_vl": {
+                    "pre_prompt": "",
+                    "post_prompt": " Answer:"
+                }
+            }
+        },
+        "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 0x7fa7b10d8ca0>",
+            "doc_to_text": "<function seed_doc_to_text at 0x7fa7b1101310>",
+            "doc_to_target": "answer",
+            "process_results": "<function seed_process_result at 0x7fa7b1101820>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "seedbench_2_plus_Chart",
+                    "aggregation": "<function seed_aggregation_result at 0x7fa7b1101d30>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "seedbench_2_plus_Map",
+                    "aggregation": "<function seed_aggregation_result at 0x7fa7b0ea31f0>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "seedbench_2_plus_Web",
+                    "aggregation": "<function seed_aggregation_result at 0x7fa7b0ea3670>",
+                    "higher_is_better": true
+                },
+                {
+                    "metric": "seedbench_2_plus_all",
+                    "aggregation": "<function seed_aggregation_result at 0x7fa7b0ea3af0>",
+                    "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."
+                }
+            }
+        }
+    },
+    "versions": {
+        "chartqa": "Yaml",
+        "seedbench_2_plus": "Yaml"
+    },
+    "n-shot": {
+        "chartqa": 0,
+        "seedbench_2_plus": 0
+    },
+    "model_configs": {
+        "model": "llava",
+        "model_args": "pretrained=/cm/archive/namnv78/checkpoints/Xphi35-siglip224/SMOE/1M3/Full_smoe_tcmoe,conv_template=phi35",
+        "batch_size": "1",
+        "device": null,
+        "limit": null,
+        "bootstrap_iters": 100000,
+        "gen_kwargs": ""
+    },
+    "git_hash": "289c7fe5"
+}