Hugging Face's logo

Pisethan
/
sangapac-math

Safetensors
Khmer
t5
Model card Files
xet
 Community
Pisethan commited on
Commit
1d34364
·
verified ·
1 Parent(s): 581f0f9

Fine-tuned model update

Browse files
Files changed (9) hide show
  1. checkpoint-1605/config.json +49 -0
  2. checkpoint-1605/model.safetensors +3 -0
  3. checkpoint-1605/optimizer.pt +3 -0
  4. checkpoint-1605/rng_state.pth +3 -0
  5. checkpoint-1605/scheduler.pt +3 -0
  6. checkpoint-1605/trainer_state.json +281 -0
  7. checkpoint-1605/training_args.bin +3 -0
  8. config.json +49 -45
  9. model.safetensors +2 -2
checkpoint-1605/config.json ADDED
@@ -0,0 +1,49 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "xlm-roberta-base",
3
+ "architectures": [
4
+ "XLMRobertaForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 768,
13
+ "id2label": {
14
+ "0": "arithmetic",
15
+ "1": "multiplication",
16
+ "2": "division",
17
+ "3": "algebra",
18
+ "4": "geometry",
19
+ "5": "exponents",
20
+ "6": "addition",
21
+ "7": "subtraction"
22
+ },
23
+ "initializer_range": 0.02,
24
+ "intermediate_size": 3072,
25
+ "label2id": {
26
+ "addition": 6,
27
+ "algebra": 3,
28
+ "arithmetic": 0,
29
+ "division": 2,
30
+ "exponents": 5,
31
+ "geometry": 4,
32
+ "multiplication": 1,
33
+ "subtraction": 7
34
+ },
35
+ "layer_norm_eps": 1e-05,
36
+ "max_position_embeddings": 514,
37
+ "model_type": "xlm-roberta",
38
+ "num_attention_heads": 12,
39
+ "num_hidden_layers": 12,
40
+ "output_past": true,
41
+ "pad_token_id": 1,
42
+ "position_embedding_type": "absolute",
43
+ "problem_type": "single_label_classification",
44
+ "torch_dtype": "float32",
45
+ "transformers_version": "4.47.0",
46
+ "type_vocab_size": 1,
47
+ "use_cache": true,
48
+ "vocab_size": 250002
49
+ }
checkpoint-1605/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a78c45c29ec4d43b9c9d84054394a0b682489b9ba02524e706f2df280b697526
3
+ size 1112223464
checkpoint-1605/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e5fb5006a7d66c0f5968f9dd26f810ece329a7fd2aca9b05f1dd33648e843d0
3
+ size 2224566458
checkpoint-1605/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0cabe66da15f8ca03db6f6a03d984749c1a0451e535e3b10bab0a1804a5f5647
3
+ size 14244
checkpoint-1605/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:32792d1cac4ccb1b7fc2ad148fbd22dc436dea5816038d7c4ff32c7d64fcb20b
3
+ size 1064
checkpoint-1605/trainer_state.json ADDED
@@ -0,0 +1,281 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "eval_steps": 500,
6
+ "global_step": 1605,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.09345794392523364,
13
+ "grad_norm": 0.008999723941087723,
14
+ "learning_rate": 4.844236760124611e-05,
15
+ "loss": 0.243,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.18691588785046728,
20
+ "grad_norm": 0.009431353770196438,
21
+ "learning_rate": 4.6884735202492216e-05,
22
+ "loss": 0.0069,
23
+ "step": 100
24
+ },
25
+ {
26
+ "epoch": 0.2803738317757009,
27
+ "grad_norm": 0.00998706091195345,
28
+ "learning_rate": 4.532710280373832e-05,
29
+ "loss": 0.0004,
30
+ "step": 150
31
+ },
32
+ {
33
+ "epoch": 0.37383177570093457,
34
+ "grad_norm": 0.007257778197526932,
35
+ "learning_rate": 4.376947040498442e-05,
36
+ "loss": 0.0035,
37
+ "step": 200
38
+ },
39
+ {
40
+ "epoch": 0.4672897196261682,
41
+ "grad_norm": 0.020342009142041206,
42
+ "learning_rate": 4.221183800623053e-05,
43
+ "loss": 0.0067,
44
+ "step": 250
45
+ },
46
+ {
47
+ "epoch": 0.5607476635514018,
48
+ "grad_norm": 0.004194826819002628,
49
+ "learning_rate": 4.0654205607476636e-05,
50
+ "loss": 0.0004,
51
+ "step": 300
52
+ },
53
+ {
54
+ "epoch": 0.6542056074766355,
55
+ "grad_norm": 0.0033228793181478977,
56
+ "learning_rate": 3.909657320872274e-05,
57
+ "loss": 0.0002,
58
+ "step": 350
59
+ },
60
+ {
61
+ "epoch": 0.7476635514018691,
62
+ "grad_norm": 0.002507598605006933,
63
+ "learning_rate": 3.753894080996885e-05,
64
+ "loss": 0.0001,
65
+ "step": 400
66
+ },
67
+ {
68
+ "epoch": 0.8411214953271028,
69
+ "grad_norm": 0.004089244641363621,
70
+ "learning_rate": 3.5981308411214956e-05,
71
+ "loss": 0.0036,
72
+ "step": 450
73
+ },
74
+ {
75
+ "epoch": 0.9345794392523364,
76
+ "grad_norm": 0.004642685875296593,
77
+ "learning_rate": 3.442367601246106e-05,
78
+ "loss": 0.0036,
79
+ "step": 500
80
+ },
81
+ {
82
+ "epoch": 1.0,
83
+ "eval_loss": 8.867182731628418,
84
+ "eval_runtime": 0.7286,
85
+ "eval_samples_per_second": 6.862,
86
+ "eval_steps_per_second": 1.372,
87
+ "step": 535
88
+ },
89
+ {
90
+ "epoch": 1.02803738317757,
91
+ "grad_norm": 0.026588544249534607,
92
+ "learning_rate": 3.286604361370716e-05,
93
+ "loss": 0.0031,
94
+ "step": 550
95
+ },
96
+ {
97
+ "epoch": 1.1214953271028036,
98
+ "grad_norm": 0.0032198007684201,
99
+ "learning_rate": 3.130841121495327e-05,
100
+ "loss": 0.0002,
101
+ "step": 600
102
+ },
103
+ {
104
+ "epoch": 1.2149532710280373,
105
+ "grad_norm": 0.0027408436872065067,
106
+ "learning_rate": 2.9750778816199376e-05,
107
+ "loss": 0.0001,
108
+ "step": 650
109
+ },
110
+ {
111
+ "epoch": 1.308411214953271,
112
+ "grad_norm": 0.001963774673640728,
113
+ "learning_rate": 2.8193146417445482e-05,
114
+ "loss": 0.0001,
115
+ "step": 700
116
+ },
117
+ {
118
+ "epoch": 1.4018691588785046,
119
+ "grad_norm": 0.0030643907375633717,
120
+ "learning_rate": 2.663551401869159e-05,
121
+ "loss": 0.0036,
122
+ "step": 750
123
+ },
124
+ {
125
+ "epoch": 1.4953271028037383,
126
+ "grad_norm": 0.0027519240975379944,
127
+ "learning_rate": 2.5077881619937692e-05,
128
+ "loss": 0.001,
129
+ "step": 800
130
+ },
131
+ {
132
+ "epoch": 1.588785046728972,
133
+ "grad_norm": 0.002502395771443844,
134
+ "learning_rate": 2.3520249221183802e-05,
135
+ "loss": 0.0001,
136
+ "step": 850
137
+ },
138
+ {
139
+ "epoch": 1.6822429906542056,
140
+ "grad_norm": 0.0024411738850176334,
141
+ "learning_rate": 2.196261682242991e-05,
142
+ "loss": 0.0008,
143
+ "step": 900
144
+ },
145
+ {
146
+ "epoch": 1.7757009345794392,
147
+ "grad_norm": 0.003065146040171385,
148
+ "learning_rate": 2.0404984423676012e-05,
149
+ "loss": 0.0034,
150
+ "step": 950
151
+ },
152
+ {
153
+ "epoch": 1.8691588785046729,
154
+ "grad_norm": 0.0014599463902413845,
155
+ "learning_rate": 1.884735202492212e-05,
156
+ "loss": 0.0001,
157
+ "step": 1000
158
+ },
159
+ {
160
+ "epoch": 1.9626168224299065,
161
+ "grad_norm": 0.0023229168727993965,
162
+ "learning_rate": 1.7289719626168225e-05,
163
+ "loss": 0.0038,
164
+ "step": 1050
165
+ },
166
+ {
167
+ "epoch": 2.0,
168
+ "eval_loss": 7.628686428070068,
169
+ "eval_runtime": 0.3776,
170
+ "eval_samples_per_second": 13.241,
171
+ "eval_steps_per_second": 2.648,
172
+ "step": 1070
173
+ },
174
+ {
175
+ "epoch": 2.05607476635514,
176
+ "grad_norm": 0.0019721754360944033,
177
+ "learning_rate": 1.573208722741433e-05,
178
+ "loss": 0.0001,
179
+ "step": 1100
180
+ },
181
+ {
182
+ "epoch": 2.149532710280374,
183
+ "grad_norm": 0.002189771970734,
184
+ "learning_rate": 1.4174454828660435e-05,
185
+ "loss": 0.0034,
186
+ "step": 1150
187
+ },
188
+ {
189
+ "epoch": 2.2429906542056073,
190
+ "grad_norm": 0.0022546553518623114,
191
+ "learning_rate": 1.2616822429906542e-05,
192
+ "loss": 0.0001,
193
+ "step": 1200
194
+ },
195
+ {
196
+ "epoch": 2.336448598130841,
197
+ "grad_norm": 0.0020222938619554043,
198
+ "learning_rate": 1.1059190031152649e-05,
199
+ "loss": 0.0007,
200
+ "step": 1250
201
+ },
202
+ {
203
+ "epoch": 2.4299065420560746,
204
+ "grad_norm": 0.0017899508820846677,
205
+ "learning_rate": 9.501557632398754e-06,
206
+ "loss": 0.0001,
207
+ "step": 1300
208
+ },
209
+ {
210
+ "epoch": 2.5233644859813085,
211
+ "grad_norm": 0.00168142793700099,
212
+ "learning_rate": 7.94392523364486e-06,
213
+ "loss": 0.0001,
214
+ "step": 1350
215
+ },
216
+ {
217
+ "epoch": 2.616822429906542,
218
+ "grad_norm": 0.0016604465199634433,
219
+ "learning_rate": 6.386292834890965e-06,
220
+ "loss": 0.0008,
221
+ "step": 1400
222
+ },
223
+ {
224
+ "epoch": 2.710280373831776,
225
+ "grad_norm": 0.0016119988868013024,
226
+ "learning_rate": 4.828660436137072e-06,
227
+ "loss": 0.0001,
228
+ "step": 1450
229
+ },
230
+ {
231
+ "epoch": 2.803738317757009,
232
+ "grad_norm": 0.005219247657805681,
233
+ "learning_rate": 3.2710280373831774e-06,
234
+ "loss": 0.0043,
235
+ "step": 1500
236
+ },
237
+ {
238
+ "epoch": 2.897196261682243,
239
+ "grad_norm": 0.0015339914243668318,
240
+ "learning_rate": 1.7133956386292835e-06,
241
+ "loss": 0.0007,
242
+ "step": 1550
243
+ },
244
+ {
245
+ "epoch": 2.9906542056074765,
246
+ "grad_norm": 0.001584639772772789,
247
+ "learning_rate": 1.5576323987538942e-07,
248
+ "loss": 0.0008,
249
+ "step": 1600
250
+ },
251
+ {
252
+ "epoch": 3.0,
253
+ "eval_loss": 6.102351665496826,
254
+ "eval_runtime": 0.3512,
255
+ "eval_samples_per_second": 14.238,
256
+ "eval_steps_per_second": 2.848,
257
+ "step": 1605
258
+ }
259
+ ],
260
+ "logging_steps": 50,
261
+ "max_steps": 1605,
262
+ "num_input_tokens_seen": 0,
263
+ "num_train_epochs": 3,
264
+ "save_steps": 500,
265
+ "stateful_callbacks": {
266
+ "TrainerControl": {
267
+ "args": {
268
+ "should_epoch_stop": false,
269
+ "should_evaluate": false,
270
+ "should_log": false,
271
+ "should_save": true,
272
+ "should_training_stop": true
273
+ },
274
+ "attributes": {}
275
+ }
276
+ },
277
+ "total_flos": 6748570106652672.0,
278
+ "train_batch_size": 64,
279
+ "trial_name": null,
280
+ "trial_params": null
281
+ }
checkpoint-1605/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db1e710e71eadddc626160bd791618b41838ca83bf4a24d0234972dcf509bbe6
3
+ size 5240
config.json CHANGED
@@ -1,45 +1,49 @@
1
- {
2
- "_name_or_path": "D:/sangapac-math-model",
3
- "architectures": [
4
- "XLMRobertaForSequenceClassification"
5
- ],
6
- "attention_probs_dropout_prob": 0.1,
7
- "bos_token_id": 0,
8
- "classifier_dropout": null,
9
- "eos_token_id": 2,
10
- "hidden_act": "gelu",
11
- "hidden_dropout_prob": 0.1,
12
- "hidden_size": 768,
13
- "id2label": {
14
- "0": "arithmetic",
15
- "1": "multiplication",
16
- "2": "division",
17
- "3": "algebra",
18
- "4": "geometry",
19
- "5": "exponents"
20
- },
21
- "initializer_range": 0.02,
22
- "intermediate_size": 3072,
23
- "label2id": {
24
- "algebra": 3,
25
- "arithmetic": 0,
26
- "division": 2,
27
- "exponents": 5,
28
- "geometry": 4,
29
- "multiplication": 1
30
- },
31
- "layer_norm_eps": 1e-05,
32
- "max_position_embeddings": 514,
33
- "model_type": "xlm-roberta",
34
- "num_attention_heads": 12,
35
- "num_hidden_layers": 12,
36
- "output_past": true,
37
- "pad_token_id": 1,
38
- "position_embedding_type": "absolute",
39
- "problem_type": "single_label_classification",
40
- "torch_dtype": "float32",
41
- "transformers_version": "4.46.2",
42
- "type_vocab_size": 1,
43
- "use_cache": true,
44
- "vocab_size": 250002
45
- }
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "xlm-roberta-base",
3
+ "architectures": [
4
+ "XLMRobertaForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 768,
13
+ "id2label": {
14
+ "0": "arithmetic",
15
+ "1": "multiplication",
16
+ "2": "division",
17
+ "3": "algebra",
18
+ "4": "geometry",
19
+ "5": "exponents",
20
+ "6": "addition",
21
+ "7": "subtraction"
22
+ },
23
+ "initializer_range": 0.02,
24
+ "intermediate_size": 3072,
25
+ "label2id": {
26
+ "addition": 6,
27
+ "algebra": 3,
28
+ "arithmetic": 0,
29
+ "division": 2,
30
+ "exponents": 5,
31
+ "geometry": 4,
32
+ "multiplication": 1,
33
+ "subtraction": 7
34
+ },
35
+ "layer_norm_eps": 1e-05,
36
+ "max_position_embeddings": 514,
37
+ "model_type": "xlm-roberta",
38
+ "num_attention_heads": 12,
39
+ "num_hidden_layers": 12,
40
+ "output_past": true,
41
+ "pad_token_id": 1,
42
+ "position_embedding_type": "absolute",
43
+ "problem_type": "single_label_classification",
44
+ "torch_dtype": "float32",
45
+ "transformers_version": "4.47.0",
46
+ "type_vocab_size": 1,
47
+ "use_cache": true,
48
+ "vocab_size": 250002
49
+ }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5340d27752a9d66985ccc8cc9e0fcfd88f15be422bd13736f1bb2b6e9263c1ff
3
- size 1112217312
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a78c45c29ec4d43b9c9d84054394a0b682489b9ba02524e706f2df280b697526
3
+ size 1112223464