Hugging Face's logo

Wendy-Fly
/
ACL-2025

Model card Files Community
2
Wendy-Fly commited on
Commit
51b2cb5
·
verified ·
1 Parent(s): 55b233e

Upload mme_all.sh with huggingface_hub

Browse files
Files changed (1) hide show
  1. mme_all.sh +262 -0
mme_all.sh ADDED
@@ -0,0 +1,262 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ #!/bin/bash
2
+
3
+ python -m llava.eval.model_vqa_loader \
4
+ --model-path checkpoints/llava-v1.5-7b-task-llava_image_tune_logits_NImg_Gas_5P_0203 \
5
+ --question-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/llava_mme.jsonl \
6
+ --image-folder /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/MME_Benchmark_release_version \
7
+ --answers-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/answers/llava_image_tune_logits_NImg_Gas_5P_0203.jsonl \
8
+ --temperature 0 \
9
+ --conv-mode vicuna_v1
10
+
11
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME
12
+
13
+ python convert_answer_to_mme.py --experiment llava_image_tune_logits_NImg_Gas_5P_0203
14
+
15
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/eval_tool
16
+
17
+ python calculation.py --results_dir answers/llava_image_tune_logits_NImg_Gas_5P_0203
18
+
19
+
20
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/NIPS_2025/LLaVA
21
+ #!/bin/bash
22
+
23
+ python -m llava.eval.model_vqa_loader \
24
+ --model-path checkpoints/llava-v1.5-7b-task-llava_image_tune_logits_NImg_Gas_10P_0203 \
25
+ --question-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/llava_mme.jsonl \
26
+ --image-folder /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/MME_Benchmark_release_version \
27
+ --answers-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/answers/llava_image_tune_logits_NImg_Gas_10P_0203.jsonl \
28
+ --temperature 0 \
29
+ --conv-mode vicuna_v1
30
+
31
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME
32
+
33
+ python convert_answer_to_mme.py --experiment llava_image_tune_logits_NImg_Gas_10P_0203
34
+
35
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/eval_tool
36
+
37
+ python calculation.py --results_dir answers/llava_image_tune_logits_NImg_Gas_10P_0203
38
+
39
+
40
+
41
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/NIPS_2025/LLaVA
42
+ #!/bin/bash
43
+
44
+ python -m llava.eval.model_vqa_loader \
45
+ --model-path checkpoints/llava-v1.5-7b-task-llava_image_tune_logits_NImg_Gas_20P_0203 \
46
+ --question-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/llava_mme.jsonl \
47
+ --image-folder /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/MME_Benchmark_release_version \
48
+ --answers-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/answers/llava_image_tune_logits_NImg_Gas_20P_0203.jsonl \
49
+ --temperature 0 \
50
+ --conv-mode vicuna_v1
51
+
52
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME
53
+
54
+ python convert_answer_to_mme.py --experiment llava_image_tune_logits_NImg_Gas_20P_0203
55
+
56
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/eval_tool
57
+
58
+ python calculation.py --results_dir answers/llava_image_tune_logits_NImg_Gas_20P_0203
59
+
60
+
61
+
62
+
63
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/NIPS_2025/LLaVA
64
+ #!/bin/bash
65
+
66
+ python -m llava.eval.model_vqa_loader \
67
+ --model-path checkpoints/llava-v1.5-7b-task-llava_image_tune_logits_NImg_Gas_40P_0203 \
68
+ --question-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/llava_mme.jsonl \
69
+ --image-folder /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/MME_Benchmark_release_version \
70
+ --answers-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/answers/llava_image_tune_logits_NImg_Gas_40P_0203.jsonl \
71
+ --temperature 0 \
72
+ --conv-mode vicuna_v1
73
+
74
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME
75
+
76
+ python convert_answer_to_mme.py --experiment llava_image_tune_logits_NImg_Gas_40P_0203
77
+
78
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/eval_tool
79
+
80
+ python calculation.py --results_dir answers/llava_image_tune_logits_NImg_Gas_40P_0203
81
+
82
+
83
+
84
+
85
+ ##################################################################################################################
86
+
87
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/NIPS_2025/LLaVA
88
+ #!/bin/bash
89
+
90
+ python -m llava.eval.model_vqa_loader \
91
+ --model-path checkpoints/llava-v1.5-7b-task-llava_image_tune_logits_NImg_Rand_5P_0203 \
92
+ --question-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/llava_mme.jsonl \
93
+ --image-folder /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/MME_Benchmark_release_version \
94
+ --answers-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/answers/llava_image_tune_logits_NImg_Rand_5P_0203.jsonl \
95
+ --temperature 0 \
96
+ --conv-mode vicuna_v1
97
+
98
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME
99
+
100
+ python convert_answer_to_mme.py --experiment llava_image_tune_logits_NImg_Rand_5P_0203
101
+
102
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/eval_tool
103
+
104
+ python calculation.py --results_dir answers/llava_image_tune_logits_NImg_Rand_5P_0203
105
+
106
+
107
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/NIPS_2025/LLaVA
108
+ #!/bin/bash
109
+
110
+ python -m llava.eval.model_vqa_loader \
111
+ --model-path checkpoints/llava-v1.5-7b-task-llava_image_tune_logits_NImg_Rand_10P_0203 \
112
+ --question-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/llava_mme.jsonl \
113
+ --image-folder /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/MME_Benchmark_release_version \
114
+ --answers-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/answers/llava_image_tune_logits_NImg_Rand_10P_0203.jsonl \
115
+ --temperature 0 \
116
+ --conv-mode vicuna_v1
117
+
118
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME
119
+
120
+ python convert_answer_to_mme.py --experiment llava_image_tune_logits_NImg_Rand_10P_0203
121
+
122
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/eval_tool
123
+
124
+ python calculation.py --results_dir answers/llava_image_tune_logits_NImg_Rand_10P_0203
125
+
126
+
127
+
128
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/NIPS_2025/LLaVA
129
+ #!/bin/bash
130
+
131
+ python -m llava.eval.model_vqa_loader \
132
+ --model-path checkpoints/llava-v1.5-7b-task-llava_image_tune_logits_NImg_Rand_20P_0203 \
133
+ --question-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/llava_mme.jsonl \
134
+ --image-folder /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/MME_Benchmark_release_version \
135
+ --answers-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/answers/llava_image_tune_logits_NImg_Rand_20P_0203.jsonl \
136
+ --temperature 0 \
137
+ --conv-mode vicuna_v1
138
+
139
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME
140
+
141
+ python convert_answer_to_mme.py --experiment llava_image_tune_logits_NImg_Rand_20P_0203
142
+
143
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/eval_tool
144
+
145
+ python calculation.py --results_dir answers/llava_image_tune_logits_NImg_Rand_20P_0203
146
+
147
+
148
+
149
+
150
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/NIPS_2025/LLaVA
151
+ #!/bin/bash
152
+
153
+ python -m llava.eval.model_vqa_loader \
154
+ --model-path checkpoints/llava-v1.5-7b-task-llava_image_tune_logits_NImg_Rand_40P_0203 \
155
+ --question-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/llava_mme.jsonl \
156
+ --image-folder /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/MME_Benchmark_release_version \
157
+ --answers-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/answers/llava_image_tune_logits_NImg_Rand_40P_0203.jsonl \
158
+ --temperature 0 \
159
+ --conv-mode vicuna_v1
160
+
161
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME
162
+
163
+ python convert_answer_to_mme.py --experiment llava_image_tune_logits_NImg_Rand_40P_0203
164
+
165
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/eval_tool
166
+
167
+ python calculation.py --results_dir answers/llava_image_tune_logits_NImg_Rand_40P_0203
168
+
169
+
170
+
171
+
172
+
173
+
174
+ ############################################################################################################################
175
+
176
+
177
+
178
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/NIPS_2025/LLaVA
179
+ #!/bin/bash
180
+
181
+ python -m llava.eval.model_vqa_loader \
182
+ --model-path checkpoints/llava-v1.5-7b-task-llava_image_tune_logits_NImg_High_5P_0203 \
183
+ --question-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/llava_mme.jsonl \
184
+ --image-folder /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/MME_Benchmark_release_version \
185
+ --answers-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/answers/llava_image_tune_logits_NImg_High_5P_0203.jsonl \
186
+ --temperature 0 \
187
+ --conv-mode vicuna_v1
188
+
189
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME
190
+
191
+ python convert_answer_to_mme.py --experiment llava_image_tune_logits_NImg_High_5P_0203
192
+
193
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/eval_tool
194
+
195
+ python calculation.py --results_dir answers/llava_image_tune_logits_NImg_High_5P_0203
196
+
197
+
198
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/NIPS_2025/LLaVA
199
+ #!/bin/bash
200
+
201
+ python -m llava.eval.model_vqa_loader \
202
+ --model-path checkpoints/llava-v1.5-7b-task-llava_image_tune_logits_NImg_High_10P_0203 \
203
+ --question-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/llava_mme.jsonl \
204
+ --image-folder /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/MME_Benchmark_release_version \
205
+ --answers-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/answers/llava_image_tune_logits_NImg_High_10P_0203.jsonl \
206
+ --temperature 0 \
207
+ --conv-mode vicuna_v1
208
+
209
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME
210
+
211
+ python convert_answer_to_mme.py --experiment llava_image_tune_logits_NImg_High_10P_0203
212
+
213
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/eval_tool
214
+
215
+ python calculation.py --results_dir answers/llava_image_tune_logits_NImg_High_10P_0203
216
+
217
+
218
+
219
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/NIPS_2025/LLaVA
220
+ #!/bin/bash
221
+
222
+ python -m llava.eval.model_vqa_loader \
223
+ --model-path checkpoints/llava-v1.5-7b-task-llava_image_tune_logits_NImg_High_20P_0203 \
224
+ --question-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/llava_mme.jsonl \
225
+ --image-folder /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/MME_Benchmark_release_version \
226
+ --answers-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/answers/llava_image_tune_logits_NImg_High_20P_0203.jsonl \
227
+ --temperature 0 \
228
+ --conv-mode vicuna_v1
229
+
230
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME
231
+
232
+ python convert_answer_to_mme.py --experiment llava_image_tune_logits_NImg_High_20P_0203
233
+
234
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/eval_tool
235
+
236
+ python calculation.py --results_dir answers/llava_image_tune_logits_NImg_High_20P_0203
237
+
238
+
239
+
240
+
241
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/NIPS_2025/LLaVA
242
+ #!/bin/bash
243
+
244
+ python -m llava.eval.model_vqa_loader \
245
+ --model-path checkpoints/llava-v1.5-7b-task-llava_image_tune_logits_NImg_High_40P_0203 \
246
+ --question-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/llava_mme.jsonl \
247
+ --image-folder /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/MME_Benchmark_release_version \
248
+ --answers-file /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/answers/llava_image_tune_logits_NImg_High_40P_0203.jsonl \
249
+ --temperature 0 \
250
+ --conv-mode vicuna_v1
251
+
252
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME
253
+
254
+ python convert_answer_to_mme.py --experiment llava_image_tune_logits_NImg_High_40P_0203
255
+
256
+ cd /inspire/hdd/ws-ba572160-47f8-4ca1-984e-d6bcdeb95dbb/a100-maybe/albus/DataSet/LLaVA-VQA-Eval/MME/eval_tool
257
+
258
+ python calculation.py --results_dir answers/llava_image_tune_logits_NImg_High_40P_0203
259
+
260
+
261
+
262
+