Upload folder using huggingface_hub

#118
sft/1M3/Full_remoe/checkpoint-27572/logs/0627_1510_llava...image_llava_model_args_543ab0/gpt_response/hallusion_output_vd_model.json ADDED
The diff for this file is too large to render. See raw diff
 
sft/1M3/Full_remoe/checkpoint-27572/logs/0627_1510_llava...image_llava_model_args_543ab0/gpt_response/hallusion_output_vs_model.json ADDED
The diff for this file is too large to render. See raw diff
 
sft/1M3/Full_remoe/checkpoint-27572/logs/0627_1510_llava...image_llava_model_args_543ab0/hallusion_bench_image.json ADDED
The diff for this file is too large to render. See raw diff
 
sft/1M3/Full_remoe/checkpoint-27572/logs/0627_1510_llava...image_llava_model_args_543ab0/mathvista_testmini.json ADDED
The diff for this file is too large to render. See raw diff
 
sft/1M3/Full_remoe/checkpoint-27572/logs/0627_1510_llava...image_llava_model_args_543ab0/rank0_metric_eval_done.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ rank 0 eval done
sft/1M3/Full_remoe/checkpoint-27572/logs/0627_1510_llava...image_llava_model_args_543ab0/results.json ADDED
@@ -0,0 +1,146 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "results": {
3
+ "hallusion_bench_image": {
4
+ "aAcc,none": 42.7971,
5
+ "aAcc_stderr,none": "N/A",
6
+ "fAcc,none": 15.0289,
7
+ "fAcc_stderr,none": "N/A",
8
+ "qAcc,none": 9.4505,
9
+ "qAcc_stderr,none": "N/A",
10
+ "alias": "hallusion_bench_image"
11
+ },
12
+ "mathvista_testmini": {
13
+ "gpt_eval_score,none": 34.1,
14
+ "gpt_eval_score_stderr,none": "N/A",
15
+ "alias": "mathvista_testmini"
16
+ }
17
+ },
18
+ "configs": {
19
+ "hallusion_bench_image": {
20
+ "task": "hallusion_bench_image",
21
+ "dataset_path": "lmms-lab/HallusionBench",
22
+ "dataset_kwargs": {
23
+ "token": true
24
+ },
25
+ "test_split": "image",
26
+ "doc_to_visual": "<function hb_doc_to_visual at 0x7f3fdadc5ee0>",
27
+ "doc_to_text": "<function hb_doc_to_text at 0x7f3fdadd0670>",
28
+ "doc_to_target": "gt_answer_details",
29
+ "process_results": "<function hb_process_results at 0x7f3fdadd0e50>",
30
+ "description": "",
31
+ "target_delimiter": " ",
32
+ "fewshot_delimiter": "\n\n",
33
+ "metric_list": [
34
+ {
35
+ "metric": "aAcc",
36
+ "aggregation": "<function hb_aggregation_result_aAcc at 0x7f3fdacc6820>",
37
+ "higher_is_better": true
38
+ },
39
+ {
40
+ "metric": "qAcc",
41
+ "aggregation": "<function hb_aggregation_result_qAcc at 0x7f3fdacc6e50>",
42
+ "higher_is_better": true
43
+ },
44
+ {
45
+ "metric": "fAcc",
46
+ "aggregation": "<function hb_aggregation_result_fAcc at 0x7f3fdab19670>",
47
+ "higher_is_better": true
48
+ }
49
+ ],
50
+ "output_type": "generate_until",
51
+ "generation_kwargs": {
52
+ "max_new_tokens": 128,
53
+ "temperature": 0.0,
54
+ "top_p": 1.0,
55
+ "num_beams": 1,
56
+ "do_sample": false,
57
+ "until": [
58
+ "\n\n"
59
+ ]
60
+ },
61
+ "repeats": 1,
62
+ "should_decontaminate": false,
63
+ "metadata": [
64
+ {
65
+ "version": 0.0
66
+ }
67
+ ],
68
+ "model_specific_prompt_kwargs": {
69
+ "default": {
70
+ "pre_prompt": "",
71
+ "post_prompt": ""
72
+ }
73
+ }
74
+ },
75
+ "mathvista_testmini": {
76
+ "task": "mathvista_testmini",
77
+ "dataset_path": "AI4Math/MathVista",
78
+ "dataset_kwargs": {
79
+ "token": true
80
+ },
81
+ "test_split": "testmini",
82
+ "doc_to_visual": "<function mathvista_doc_to_visual at 0x7f3faeadd9d0>",
83
+ "doc_to_text": "<function mathvista_doc_to_text at 0x7f3faeae90d0>",
84
+ "doc_to_target": "answer",
85
+ "process_results": "<function mathvista_process_results at 0x7f3fae6e0790>",
86
+ "description": "",
87
+ "target_delimiter": " ",
88
+ "fewshot_delimiter": "\n\n",
89
+ "metric_list": [
90
+ {
91
+ "metric": "gpt_eval_score",
92
+ "aggregation": "<function mathvista_aggregate_results at 0x7f3fae6e8e50>",
93
+ "higher_is_better": true
94
+ }
95
+ ],
96
+ "output_type": "generate_until",
97
+ "generation_kwargs": {
98
+ "until": [
99
+ "ASSISTANT:"
100
+ ],
101
+ "max_new_tokens": 1024,
102
+ "temperature": 0.0,
103
+ "top_p": 1.0,
104
+ "num_beams": 1,
105
+ "do_sample": false,
106
+ "image_aspect_ratio": "original"
107
+ },
108
+ "repeats": 1,
109
+ "should_decontaminate": false,
110
+ "model_specific_prompt_kwargs": {
111
+ "default": {
112
+ "shot_type": "format-prompt",
113
+ "shot": 0,
114
+ "use_caption": false,
115
+ "use_ocr": false
116
+ },
117
+ "phi3v": {
118
+ "shot_type": "solution"
119
+ }
120
+ },
121
+ "model_specific_generation_kwargs": {
122
+ "llava": {
123
+ "image_aspect_ratio": "original"
124
+ }
125
+ }
126
+ }
127
+ },
128
+ "versions": {
129
+ "hallusion_bench_image": "Yaml",
130
+ "mathvista_testmini": "Yaml"
131
+ },
132
+ "n-shot": {
133
+ "hallusion_bench_image": 0,
134
+ "mathvista_testmini": 0
135
+ },
136
+ "model_configs": {
137
+ "model": "llava",
138
+ "model_args": "pretrained=/cm/archive/namnv78/checkpoints/Xphi35-siglip224/SMOE/1M3/Full_remoe/checkpoint-27572,conv_template=phi35",
139
+ "batch_size": "1",
140
+ "device": null,
141
+ "limit": null,
142
+ "bootstrap_iters": 100000,
143
+ "gen_kwargs": ""
144
+ },
145
+ "git_hash": "289c7fe5"
146
+ }
sft/1M3/Full_remoe/checkpoint-27572/logs/0627_1510_llava...image_llava_model_args_543ab0/submissions/mathvista_testmini_scores.json ADDED
The diff for this file is too large to render. See raw diff