chuuhtetnaing commited on
Commit
eda9b27
·
verified ·
1 Parent(s): 61f713b

fine-tuned for 1 epoch

Browse files
Files changed (4) hide show
  1. README.md +245 -0
  2. adapter_config.json +41 -0
  3. adapter_model.safetensors +3 -0
  4. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,245 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ license: apache-2.0
4
+ base_model: HuggingFaceTB/SmolVLM-Instruct
5
+ tags:
6
+ - generated_from_trainer
7
+ model-index:
8
+ - name: smolvlm-mmocr-sft-round-2
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ # smolvlm-mmocr-sft-round-2
16
+
17
+ This model is a fine-tuned version of [HuggingFaceTB/SmolVLM-Instruct](https://huggingface.co/HuggingFaceTB/SmolVLM-Instruct) on an unknown dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 0.0196
20
+
21
+ ## Model description
22
+
23
+ More information needed
24
+
25
+ ## Intended uses & limitations
26
+
27
+ More information needed
28
+
29
+ ## Training and evaluation data
30
+
31
+ More information needed
32
+
33
+ ## Training procedure
34
+
35
+ ### Training hyperparameters
36
+
37
+ The following hyperparameters were used during training:
38
+ - learning_rate: 1e-05
39
+ - train_batch_size: 96
40
+ - eval_batch_size: 96
41
+ - seed: 42
42
+ - gradient_accumulation_steps: 8
43
+ - total_train_batch_size: 768
44
+ - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
45
+ - lr_scheduler_type: linear
46
+ - lr_scheduler_warmup_steps: 1000
47
+ - num_epochs: 1
48
+
49
+ ### Training results
50
+
51
+ | Training Loss | Epoch | Step | Validation Loss |
52
+ |:-------------:|:------:|:----:|:---------------:|
53
+ | 0.5099 | 0.0054 | 20 | 0.4944 |
54
+ | 0.4426 | 0.0109 | 40 | 0.4381 |
55
+ | 0.4117 | 0.0163 | 60 | 0.3822 |
56
+ | 0.3378 | 0.0217 | 80 | 0.3352 |
57
+ | 0.3107 | 0.0271 | 100 | 0.2984 |
58
+ | 0.2839 | 0.0326 | 120 | 0.2696 |
59
+ | 0.2629 | 0.0380 | 140 | 0.2478 |
60
+ | 0.2467 | 0.0434 | 160 | 0.2305 |
61
+ | 0.2306 | 0.0489 | 180 | 0.2160 |
62
+ | 0.2283 | 0.0543 | 200 | 0.2037 |
63
+ | 0.1958 | 0.0597 | 220 | 0.1927 |
64
+ | 0.1887 | 0.0651 | 240 | 0.1830 |
65
+ | 0.1699 | 0.0706 | 260 | 0.1738 |
66
+ | 0.155 | 0.0760 | 280 | 0.1655 |
67
+ | 0.1579 | 0.0814 | 300 | 0.1575 |
68
+ | 0.1437 | 0.0869 | 320 | 0.1506 |
69
+ | 0.1549 | 0.0923 | 340 | 0.1437 |
70
+ | 0.1336 | 0.0977 | 360 | 0.1371 |
71
+ | 0.1268 | 0.1032 | 380 | 0.1314 |
72
+ | 0.139 | 0.1086 | 400 | 0.1258 |
73
+ | 0.1215 | 0.1140 | 420 | 0.1205 |
74
+ | 0.1103 | 0.1194 | 440 | 0.1156 |
75
+ | 0.1141 | 0.1249 | 460 | 0.1111 |
76
+ | 0.1175 | 0.1303 | 480 | 0.1066 |
77
+ | 0.0978 | 0.1357 | 500 | 0.1023 |
78
+ | 0.1016 | 0.1412 | 520 | 0.0986 |
79
+ | 0.0899 | 0.1466 | 540 | 0.0947 |
80
+ | 0.0952 | 0.1520 | 560 | 0.0909 |
81
+ | 0.0775 | 0.1574 | 580 | 0.0876 |
82
+ | 0.0862 | 0.1629 | 600 | 0.0845 |
83
+ | 0.0777 | 0.1683 | 620 | 0.0809 |
84
+ | 0.077 | 0.1737 | 640 | 0.0779 |
85
+ | 0.0757 | 0.1792 | 660 | 0.0752 |
86
+ | 0.0627 | 0.1846 | 680 | 0.0724 |
87
+ | 0.0635 | 0.1900 | 700 | 0.0701 |
88
+ | 0.0619 | 0.1954 | 720 | 0.0674 |
89
+ | 0.0593 | 0.2009 | 740 | 0.0651 |
90
+ | 0.0686 | 0.2063 | 760 | 0.0633 |
91
+ | 0.0626 | 0.2117 | 780 | 0.0613 |
92
+ | 0.0596 | 0.2172 | 800 | 0.0591 |
93
+ | 0.0447 | 0.2226 | 820 | 0.0575 |
94
+ | 0.0548 | 0.2280 | 840 | 0.0556 |
95
+ | 0.0555 | 0.2334 | 860 | 0.0546 |
96
+ | 0.0547 | 0.2389 | 880 | 0.0536 |
97
+ | 0.0461 | 0.2443 | 900 | 0.0512 |
98
+ | 0.0445 | 0.2497 | 920 | 0.0501 |
99
+ | 0.0386 | 0.2552 | 940 | 0.0491 |
100
+ | 0.0334 | 0.2606 | 960 | 0.0475 |
101
+ | 0.0498 | 0.2660 | 980 | 0.0458 |
102
+ | 0.0365 | 0.2715 | 1000 | 0.0452 |
103
+ | 0.0364 | 0.2769 | 1020 | 0.0442 |
104
+ | 0.0351 | 0.2823 | 1040 | 0.0433 |
105
+ | 0.0393 | 0.2878 | 1060 | 0.0418 |
106
+ | 0.0359 | 0.2932 | 1080 | 0.0413 |
107
+ | 0.0347 | 0.2986 | 1100 | 0.0405 |
108
+ | 0.0351 | 0.3040 | 1120 | 0.0397 |
109
+ | 0.0386 | 0.3095 | 1140 | 0.0385 |
110
+ | 0.04 | 0.3149 | 1160 | 0.0380 |
111
+ | 0.0288 | 0.3203 | 1180 | 0.0375 |
112
+ | 0.0279 | 0.3258 | 1200 | 0.0370 |
113
+ | 0.0285 | 0.3312 | 1220 | 0.0362 |
114
+ | 0.0308 | 0.3366 | 1240 | 0.0357 |
115
+ | 0.0302 | 0.3420 | 1260 | 0.0351 |
116
+ | 0.026 | 0.3475 | 1280 | 0.0349 |
117
+ | 0.0262 | 0.3529 | 1300 | 0.0345 |
118
+ | 0.0295 | 0.3583 | 1320 | 0.0341 |
119
+ | 0.0315 | 0.3638 | 1340 | 0.0335 |
120
+ | 0.0325 | 0.3692 | 1360 | 0.0327 |
121
+ | 0.0264 | 0.3746 | 1380 | 0.0324 |
122
+ | 0.0321 | 0.3800 | 1400 | 0.0319 |
123
+ | 0.0268 | 0.3855 | 1420 | 0.0312 |
124
+ | 0.0248 | 0.3909 | 1440 | 0.0313 |
125
+ | 0.0361 | 0.3963 | 1460 | 0.0312 |
126
+ | 0.0277 | 0.4018 | 1480 | 0.0303 |
127
+ | 0.0274 | 0.4072 | 1500 | 0.0300 |
128
+ | 0.028 | 0.4126 | 1520 | 0.0294 |
129
+ | 0.0223 | 0.4181 | 1540 | 0.0290 |
130
+ | 0.0247 | 0.4235 | 1560 | 0.0289 |
131
+ | 0.0265 | 0.4289 | 1580 | 0.0289 |
132
+ | 0.0285 | 0.4343 | 1600 | 0.0285 |
133
+ | 0.0284 | 0.4398 | 1620 | 0.0284 |
134
+ | 0.0255 | 0.4452 | 1640 | 0.0279 |
135
+ | 0.0246 | 0.4506 | 1660 | 0.0273 |
136
+ | 0.0196 | 0.4561 | 1680 | 0.0274 |
137
+ | 0.0272 | 0.4615 | 1700 | 0.0270 |
138
+ | 0.0206 | 0.4669 | 1720 | 0.0268 |
139
+ | 0.02 | 0.4723 | 1740 | 0.0268 |
140
+ | 0.0229 | 0.4778 | 1760 | 0.0268 |
141
+ | 0.0214 | 0.4832 | 1780 | 0.0260 |
142
+ | 0.0246 | 0.4886 | 1800 | 0.0261 |
143
+ | 0.0189 | 0.4941 | 1820 | 0.0254 |
144
+ | 0.0199 | 0.4995 | 1840 | 0.0253 |
145
+ | 0.0168 | 0.5049 | 1860 | 0.0251 |
146
+ | 0.0297 | 0.5103 | 1880 | 0.0250 |
147
+ | 0.0237 | 0.5158 | 1900 | 0.0245 |
148
+ | 0.0189 | 0.5212 | 1920 | 0.0250 |
149
+ | 0.0274 | 0.5266 | 1940 | 0.0243 |
150
+ | 0.0143 | 0.5321 | 1960 | 0.0243 |
151
+ | 0.0201 | 0.5375 | 1980 | 0.0241 |
152
+ | 0.0258 | 0.5429 | 2000 | 0.0241 |
153
+ | 0.0256 | 0.5484 | 2020 | 0.0239 |
154
+ | 0.018 | 0.5538 | 2040 | 0.0235 |
155
+ | 0.019 | 0.5592 | 2060 | 0.0236 |
156
+ | 0.0252 | 0.5646 | 2080 | 0.0233 |
157
+ | 0.0215 | 0.5701 | 2100 | 0.0234 |
158
+ | 0.0204 | 0.5755 | 2120 | 0.0235 |
159
+ | 0.0165 | 0.5809 | 2140 | 0.0230 |
160
+ | 0.018 | 0.5864 | 2160 | 0.0230 |
161
+ | 0.0236 | 0.5918 | 2180 | 0.0228 |
162
+ | 0.0199 | 0.5972 | 2200 | 0.0228 |
163
+ | 0.0153 | 0.6026 | 2220 | 0.0226 |
164
+ | 0.0166 | 0.6081 | 2240 | 0.0227 |
165
+ | 0.0212 | 0.6135 | 2260 | 0.0225 |
166
+ | 0.0213 | 0.6189 | 2280 | 0.0224 |
167
+ | 0.0165 | 0.6244 | 2300 | 0.0223 |
168
+ | 0.0197 | 0.6298 | 2320 | 0.0224 |
169
+ | 0.0219 | 0.6352 | 2340 | 0.0223 |
170
+ | 0.0277 | 0.6407 | 2360 | 0.0222 |
171
+ | 0.015 | 0.6461 | 2380 | 0.0220 |
172
+ | 0.0191 | 0.6515 | 2400 | 0.0218 |
173
+ | 0.0204 | 0.6569 | 2420 | 0.0220 |
174
+ | 0.0162 | 0.6624 | 2440 | 0.0219 |
175
+ | 0.0174 | 0.6678 | 2460 | 0.0219 |
176
+ | 0.0144 | 0.6732 | 2480 | 0.0217 |
177
+ | 0.0219 | 0.6787 | 2500 | 0.0215 |
178
+ | 0.0152 | 0.6841 | 2520 | 0.0217 |
179
+ | 0.0107 | 0.6895 | 2540 | 0.0215 |
180
+ | 0.0173 | 0.6949 | 2560 | 0.0213 |
181
+ | 0.0173 | 0.7004 | 2580 | 0.0214 |
182
+ | 0.0147 | 0.7058 | 2600 | 0.0214 |
183
+ | 0.015 | 0.7112 | 2620 | 0.0211 |
184
+ | 0.0158 | 0.7167 | 2640 | 0.0211 |
185
+ | 0.0128 | 0.7221 | 2660 | 0.0211 |
186
+ | 0.0185 | 0.7275 | 2680 | 0.0210 |
187
+ | 0.0214 | 0.7329 | 2700 | 0.0210 |
188
+ | 0.0162 | 0.7384 | 2720 | 0.0209 |
189
+ | 0.017 | 0.7438 | 2740 | 0.0210 |
190
+ | 0.0124 | 0.7492 | 2760 | 0.0211 |
191
+ | 0.0187 | 0.7547 | 2780 | 0.0207 |
192
+ | 0.0166 | 0.7601 | 2800 | 0.0209 |
193
+ | 0.0204 | 0.7655 | 2820 | 0.0207 |
194
+ | 0.0168 | 0.7710 | 2840 | 0.0207 |
195
+ | 0.0209 | 0.7764 | 2860 | 0.0206 |
196
+ | 0.0166 | 0.7818 | 2880 | 0.0206 |
197
+ | 0.0174 | 0.7872 | 2900 | 0.0207 |
198
+ | 0.0208 | 0.7927 | 2920 | 0.0205 |
199
+ | 0.013 | 0.7981 | 2940 | 0.0204 |
200
+ | 0.0112 | 0.8035 | 2960 | 0.0205 |
201
+ | 0.0174 | 0.8090 | 2980 | 0.0203 |
202
+ | 0.0169 | 0.8144 | 3000 | 0.0204 |
203
+ | 0.0129 | 0.8198 | 3020 | 0.0202 |
204
+ | 0.0191 | 0.8252 | 3040 | 0.0201 |
205
+ | 0.0166 | 0.8307 | 3060 | 0.0202 |
206
+ | 0.0182 | 0.8361 | 3080 | 0.0201 |
207
+ | 0.0208 | 0.8415 | 3100 | 0.0201 |
208
+ | 0.015 | 0.8470 | 3120 | 0.0200 |
209
+ | 0.0246 | 0.8524 | 3140 | 0.0200 |
210
+ | 0.0158 | 0.8578 | 3160 | 0.0201 |
211
+ | 0.0183 | 0.8633 | 3180 | 0.0201 |
212
+ | 0.0171 | 0.8687 | 3200 | 0.0201 |
213
+ | 0.0173 | 0.8741 | 3220 | 0.0199 |
214
+ | 0.017 | 0.8795 | 3240 | 0.0199 |
215
+ | 0.0123 | 0.8850 | 3260 | 0.0199 |
216
+ | 0.0144 | 0.8904 | 3280 | 0.0200 |
217
+ | 0.0149 | 0.8958 | 3300 | 0.0199 |
218
+ | 0.0139 | 0.9013 | 3320 | 0.0199 |
219
+ | 0.0199 | 0.9067 | 3340 | 0.0198 |
220
+ | 0.0164 | 0.9121 | 3360 | 0.0197 |
221
+ | 0.0154 | 0.9175 | 3380 | 0.0198 |
222
+ | 0.0225 | 0.9230 | 3400 | 0.0198 |
223
+ | 0.0232 | 0.9284 | 3420 | 0.0198 |
224
+ | 0.0135 | 0.9338 | 3440 | 0.0197 |
225
+ | 0.016 | 0.9393 | 3460 | 0.0197 |
226
+ | 0.0164 | 0.9447 | 3480 | 0.0197 |
227
+ | 0.0147 | 0.9501 | 3500 | 0.0197 |
228
+ | 0.015 | 0.9555 | 3520 | 0.0197 |
229
+ | 0.0215 | 0.9610 | 3540 | 0.0196 |
230
+ | 0.0174 | 0.9664 | 3560 | 0.0197 |
231
+ | 0.0203 | 0.9718 | 3580 | 0.0197 |
232
+ | 0.0191 | 0.9773 | 3600 | 0.0197 |
233
+ | 0.02 | 0.9827 | 3620 | 0.0196 |
234
+ | 0.0128 | 0.9881 | 3640 | 0.0196 |
235
+ | 0.0145 | 0.9936 | 3660 | 0.0197 |
236
+ | 0.0179 | 0.9990 | 3680 | 0.0196 |
237
+
238
+
239
+ ### Framework versions
240
+
241
+ - PEFT 0.14.0
242
+ - Transformers 4.49.0
243
+ - Pytorch 2.5.1+cu124
244
+ - Datasets 3.3.2
245
+ - Tokenizers 0.21.1
adapter_config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "HuggingFaceTB/SmolVLM-Instruct",
5
+ "bias": "none",
6
+ "eva_config": null,
7
+ "exclude_modules": null,
8
+ "fan_in_fan_out": false,
9
+ "inference_mode": true,
10
+ "init_lora_weights": true,
11
+ "layer_replication": null,
12
+ "layers_pattern": null,
13
+ "layers_to_transform": null,
14
+ "loftq_config": {},
15
+ "lora_alpha": 64,
16
+ "lora_bias": false,
17
+ "lora_dropout": 0.1,
18
+ "megatron_config": null,
19
+ "megatron_core": "megatron.core",
20
+ "modules_to_save": null,
21
+ "peft_type": "LORA",
22
+ "r": 16,
23
+ "rank_pattern": {},
24
+ "revision": null,
25
+ "target_modules": [
26
+ "fc1",
27
+ "gate_proj",
28
+ "down_proj",
29
+ "out_proj",
30
+ "q_proj",
31
+ "proj",
32
+ "fc2",
33
+ "v_proj",
34
+ "k_proj",
35
+ "up_proj",
36
+ "o_proj"
37
+ ],
38
+ "task_type": "CAUSAL_LM",
39
+ "use_dora": false,
40
+ "use_rslora": false
41
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5faac249625039309923e3d11048422314cd482bd75a9cc74ef9cff9112eb03
3
+ size 108025880
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af56170fc1f10893c7099a94453dcf57a287298df33b0a6ae80a3986e471b3f9
3
+ size 5304