5da582ad10fdabc083e14c315b828322bc348f47d5eaaf44d3a16a6b5d728b0e

.gitattributes

@@ -198,3 +198,4 @@ sft/1M3/revise_Full_remoe/logs/0625_1719_llava..._pope_llava_model_args_179bff/p
 sft/1M3/revise_Full_remoe/logs/0625_1719_llava..._pope_llava_model_args_179bff/submissions/mmbench_en_dev_results.xlsx filter=lfs diff=lfs merge=lfs -text
 sft/1M3/revise_Full_remoe/logs/0625_1719_llava..._pope_llava_model_args_179bff/textvqa_val.json filter=lfs diff=lfs merge=lfs -text
 sft/1M3/revise_Full_remoe/logs/0625_1751_llava...a_val_llava_model_args_179bff/infovqa_val.json filter=lfs diff=lfs merge=lfs -text
+sft/1M3/revise_Full_remoe/logs/0626_1437_llava_v1.5_gqa_llava_model_args_179bff/gqa.json filter=lfs diff=lfs merge=lfs -text

sft/1M3/revise_Full_remoe/logs/0626_1433_llava...bench_llava_model_args_179bff/rank0_metric_eval_done.txt

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

sft/1M3/revise_Full_remoe/logs/0626_1433_llava...bench_llava_model_args_179bff/rank1_metric_eval_done.txt

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

sft/1M3/revise_Full_remoe/logs/0626_1433_llava...bench_llava_model_args_179bff/rank2_metric_eval_done.txt

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

sft/1M3/revise_Full_remoe/logs/0626_1433_llava...bench_llava_model_args_179bff/rank3_metric_eval_done.txt

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

sft/1M3/revise_Full_remoe/logs/0626_1433_llava...bench_llava_model_args_179bff/results.json

@@ -0,0 +1,67 @@
+{
+    "results": {
+        "ocrbench": {
+            "ocrbench_accuracy,none": 0.152,
+            "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 0x7f748f49dc10>",
+            "doc_to_text": "<function ocrbench_doc_to_text at 0x7f748ed7e1f0>",
+            "doc_to_target": "answer",
+            "process_results": "<function ocrbench_process_results at 0x7f748ed7e550>",
+            "description": "",
+            "target_delimiter": " ",
+            "fewshot_delimiter": "\n\n",
+            "metric_list": [
+                {
+                    "metric": "ocrbench_accuracy",
+                    "aggregation": "<function ocrbench_aggregate_accuracy at 0x7f748ed7e8b0>",
+                    "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/revise_Full_remoe,conv_template=phi35",
+        "batch_size": "1",
+        "device": null,
+        "limit": null,
+        "bootstrap_iters": 100000,
+        "gen_kwargs": ""
+    },
+    "git_hash": "289c7fe5"
+}

sft/1M3/revise_Full_remoe/logs/0626_1433_llava...bench_llava_model_args_179bff/results/ocrbench_results.txt

@@ -0,0 +1,18 @@
+######################### OCRBench #############################
+Text Recognition(Total 300): 87
+---------------- Details of Recognition Score ------------------
+Regular Text Recognition(Total 50): 14
+Irregular Text Recognition(Total 50): 18
+Artistic Text Recognition(Total 50): 34
+Handwriting Recognition(Total 50): 14
+Digit String Recognition(Total 50): 4
+Non-Semantic Text Recognition(Total 50): 3
+----------------------------------------------------------------
+Scene Text-centric VQA(Total 200): 45
+----------------------------------------------------------------
+Doc-oriented VQA(Total 200): 19
+----------------------------------------------------------------
+Key Information Extraction(Total 200): 1
+Handwritten Mathematical Expression Recognition(Total 100): 0
+--------------------- Final Score ------------------------------
+Final Score(Total 1000): 152

sft/1M3/revise_Full_remoe/logs/0626_1437_llava_v1.5_gqa_llava_model_args_179bff/gqa.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32e91639bb5de8bf0769f5cc6b51f75b93551c5a1f7fe606a984a67040a07978
+size 38394149

sft/1M3/revise_Full_remoe/logs/0626_1437_llava_v1.5_gqa_llava_model_args_179bff/rank0_metric_eval_done.txt

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

sft/1M3/revise_Full_remoe/logs/0626_1437_llava_v1.5_gqa_llava_model_args_179bff/rank1_metric_eval_done.txt

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

sft/1M3/revise_Full_remoe/logs/0626_1437_llava_v1.5_gqa_llava_model_args_179bff/rank2_metric_eval_done.txt

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

sft/1M3/revise_Full_remoe/logs/0626_1437_llava_v1.5_gqa_llava_model_args_179bff/rank3_metric_eval_done.txt

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

sft/1M3/revise_Full_remoe/logs/0626_1437_llava_v1.5_gqa_llava_model_args_179bff/results.json

@@ -0,0 +1,79 @@
+{
+    "results": {
+        "gqa": {
+            "exact_match,none": 0.39704245508029895,
+            "exact_match_stderr,none": 0.004362880425338431,
+            "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 0x7fef7d5f2af0>",
+            "doc_to_text": "<function gqa_doc_to_text at 0x7fef7d546dc0>",
+            "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/revise_Full_remoe,conv_template=phi35",
+        "batch_size": "1",
+        "device": null,
+        "limit": null,
+        "bootstrap_iters": 100000,
+        "gen_kwargs": ""
+    },
+    "git_hash": "289c7fe5"
+}