Hugging Face's logo

dahaludba
/
QSolver_Scratch

Text Classification
Transformers
Safetensors
trained-from-scratch
deberta
sequence-classification
multiple-choice
question-answering
Model card Files
xet
 Community
dahaludba commited on
Commit
d0fb9f7
·
verified ·
1 Parent(s): f13cf86

Upload folder using huggingface_hub

Browse files
Files changed (15) hide show
  1. fold_3/checkpoint-7690/config.json +45 -0
  2. fold_3/checkpoint-7690/model.safetensors +3 -0
  3. fold_3/checkpoint-7690/optimizer.pt +3 -0
  4. fold_3/checkpoint-7690/rng_state.pth +3 -0
  5. fold_3/checkpoint-7690/scaler.pt +3 -0
  6. fold_3/checkpoint-7690/scheduler.pt +3 -0
  7. fold_3/checkpoint-7690/tokenizer.json +0 -0
  8. fold_3/checkpoint-7690/tokenizer_config.json +23 -0
  9. fold_3/checkpoint-7690/trainer_state.json +204 -0
  10. fold_3/checkpoint-7690/training_args.bin +3 -0
  11. fold_3/config.json +45 -0
  12. fold_3/model.safetensors +3 -0
  13. fold_3/tokenizer.json +0 -0
  14. fold_3/tokenizer_config.json +23 -0
  15. fold_3/training_args.bin +3 -0
fold_3/checkpoint-7690/config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "DebertaV2ForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "bos_token_id": 1,
7
+ "dtype": "float32",
8
+ "eos_token_id": 2,
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 256,
12
+ "id2label": {
13
+ "0": "LABEL_0"
14
+ },
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 1024,
17
+ "label2id": {
18
+ "LABEL_0": 0
19
+ },
20
+ "layer_norm_eps": 1e-07,
21
+ "legacy": true,
22
+ "max_position_embeddings": 512,
23
+ "max_relative_positions": -1,
24
+ "model_type": "deberta-v2",
25
+ "norm_rel_ebd": "layer_norm",
26
+ "num_attention_heads": 4,
27
+ "num_hidden_layers": 4,
28
+ "pad_token_id": 0,
29
+ "pooler_dropout": 0,
30
+ "pooler_hidden_act": "gelu",
31
+ "pooler_hidden_size": 256,
32
+ "pos_att_type": [
33
+ "p2c",
34
+ "c2p"
35
+ ],
36
+ "position_biased_input": false,
37
+ "position_buckets": 256,
38
+ "relative_attention": true,
39
+ "share_att_key": true,
40
+ "tie_word_embeddings": true,
41
+ "transformers_version": "5.0.0",
42
+ "type_vocab_size": 0,
43
+ "use_cache": false,
44
+ "vocab_size": 128100
45
+ }
fold_3/checkpoint-7690/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cb3be835eafad2676bcfebc17da3c91b895f05235fe1bd6ac170c2996a6252d3
3
+ size 144611964
fold_3/checkpoint-7690/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2dc6e26f5bb33c5e9d78bebdd2ddb86a663ec609876dfff30f5c3a1d3acbdc58
3
+ size 289269963
fold_3/checkpoint-7690/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba1487549698a15e7ddbe51e184bd4ed8fb685f17efe0353d91b91a1752e0a99
3
+ size 14645
fold_3/checkpoint-7690/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3db751112c374d8a97d504e994f3937da62131f95b52bdea347942783164785b
3
+ size 1383
fold_3/checkpoint-7690/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7746ed0c61d08e439b3da47b74490e61e7d211741c0ec5c01a778df48e882028
3
+ size 1465
fold_3/checkpoint-7690/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
fold_3/checkpoint-7690/tokenizer_config.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "[CLS]",
5
+ "cls_token": "[CLS]",
6
+ "do_lower_case": false,
7
+ "eos_token": "[SEP]",
8
+ "extra_special_tokens": [
9
+ "[PAD]",
10
+ "[CLS]",
11
+ "[SEP]"
12
+ ],
13
+ "is_local": false,
14
+ "mask_token": "[MASK]",
15
+ "model_max_length": 1000000000000000019884624838656,
16
+ "pad_token": "[PAD]",
17
+ "sep_token": "[SEP]",
18
+ "split_by_punct": false,
19
+ "tokenizer_class": "DebertaV2Tokenizer",
20
+ "unk_id": 3,
21
+ "unk_token": "[UNK]",
22
+ "vocab_type": "spm"
23
+ }
fold_3/checkpoint-7690/trainer_state.json ADDED
@@ -0,0 +1,204 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 7690,
3
+ "best_metric": 0.9740967961826857,
4
+ "best_model_checkpoint": "/kaggle/working/QSolver_Checkpoints/scratch_fold_3/checkpoint-7690",
5
+ "epoch": 5.0,
6
+ "eval_steps": 500,
7
+ "global_step": 7690,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.3250975292587776,
14
+ "grad_norm": 0.104410819709301,
15
+ "learning_rate": 9.931967777272626e-05,
16
+ "loss": 0.04502335357666016,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 0.6501950585175552,
21
+ "grad_norm": 0.26247072219848633,
22
+ "learning_rate": 9.657816844825987e-05,
23
+ "loss": 0.014309930801391601,
24
+ "step": 1000
25
+ },
26
+ {
27
+ "epoch": 0.9752925877763329,
28
+ "grad_norm": 0.11667367815971375,
29
+ "learning_rate": 9.184878922543432e-05,
30
+ "loss": 0.009399337768554688,
31
+ "step": 1500
32
+ },
33
+ {
34
+ "epoch": 1.0,
35
+ "eval_accuracy": 0.869120654396728,
36
+ "eval_f1": 0.8649563830398103,
37
+ "eval_loss": 0.11769432574510574,
38
+ "eval_map@3": 0.9219495569188819,
39
+ "eval_precision": 0.8657620476328406,
40
+ "eval_recall": 0.8673524582371904,
41
+ "eval_runtime": 1.2117,
42
+ "eval_samples_per_second": 2017.853,
43
+ "eval_steps_per_second": 16.506,
44
+ "step": 1538
45
+ },
46
+ {
47
+ "epoch": 1.3003901170351106,
48
+ "grad_norm": 0.17101551592350006,
49
+ "learning_rate": 8.533338124374017e-05,
50
+ "loss": 0.007008582592010498,
51
+ "step": 2000
52
+ },
53
+ {
54
+ "epoch": 1.6254876462938883,
55
+ "grad_norm": 0.1663733571767807,
56
+ "learning_rate": 7.731001003975445e-05,
57
+ "loss": 0.0054134888648986815,
58
+ "step": 2500
59
+ },
60
+ {
61
+ "epoch": 1.9505851755526658,
62
+ "grad_norm": 0.18735824525356293,
63
+ "learning_rate": 6.81210982254242e-05,
64
+ "loss": 0.004760495662689209,
65
+ "step": 3000
66
+ },
67
+ {
68
+ "epoch": 2.0,
69
+ "eval_accuracy": 0.934560327198364,
70
+ "eval_f1": 0.9323989455717403,
71
+ "eval_loss": 0.0737488642334938,
72
+ "eval_map@3": 0.9601226993865031,
73
+ "eval_precision": 0.9334717583693026,
74
+ "eval_recall": 0.9325315651655389,
75
+ "eval_runtime": 1.1932,
76
+ "eval_samples_per_second": 2049.102,
77
+ "eval_steps_per_second": 16.762,
78
+ "step": 3076
79
+ },
80
+ {
81
+ "epoch": 2.2756827048114436,
82
+ "grad_norm": 0.32222193479537964,
83
+ "learning_rate": 5.815881152565712e-05,
84
+ "loss": 0.003930224657058716,
85
+ "step": 3500
86
+ },
87
+ {
88
+ "epoch": 2.6007802340702213,
89
+ "grad_norm": 0.057121261954307556,
90
+ "learning_rate": 4.7848321872047936e-05,
91
+ "loss": 0.0035245070457458495,
92
+ "step": 4000
93
+ },
94
+ {
95
+ "epoch": 2.925877763328999,
96
+ "grad_norm": 0.030139535665512085,
97
+ "learning_rate": 3.7629661852879294e-05,
98
+ "loss": 0.0032367331981658937,
99
+ "step": 4500
100
+ },
101
+ {
102
+ "epoch": 3.0,
103
+ "eval_accuracy": 0.9570552147239264,
104
+ "eval_f1": 0.9548006503338626,
105
+ "eval_loss": 0.04081879183650017,
106
+ "eval_map@3": 0.9734151329243353,
107
+ "eval_precision": 0.955626979019141,
108
+ "eval_recall": 0.9551920700353461,
109
+ "eval_runtime": 1.2464,
110
+ "eval_samples_per_second": 1961.593,
111
+ "eval_steps_per_second": 16.046,
112
+ "step": 4614
113
+ },
114
+ {
115
+ "epoch": 3.250975292587776,
116
+ "grad_norm": 0.18474030494689941,
117
+ "learning_rate": 2.7938944937838923e-05,
118
+ "loss": 0.0023533060550689697,
119
+ "step": 5000
120
+ },
121
+ {
122
+ "epoch": 3.576072821846554,
123
+ "grad_norm": 0.013804451562464237,
124
+ "learning_rate": 1.918975296345461e-05,
125
+ "loss": 0.0024521441459655763,
126
+ "step": 5500
127
+ },
128
+ {
129
+ "epoch": 3.9011703511053315,
130
+ "grad_norm": 0.016357487067580223,
131
+ "learning_rate": 1.1755485226870489e-05,
132
+ "loss": 0.0022003064155578615,
133
+ "step": 6000
134
+ },
135
+ {
136
+ "epoch": 4.0,
137
+ "eval_accuracy": 0.9550102249488752,
138
+ "eval_f1": 0.9529694343395565,
139
+ "eval_loss": 0.02794707380235195,
140
+ "eval_map@3": 0.9730743012951601,
141
+ "eval_precision": 0.9546482362488943,
142
+ "eval_recall": 0.9526279674712436,
143
+ "eval_runtime": 1.2213,
144
+ "eval_samples_per_second": 2002.0,
145
+ "eval_steps_per_second": 16.376,
146
+ "step": 6152
147
+ },
148
+ {
149
+ "epoch": 4.22626788036411,
150
+ "grad_norm": 0.044519033282995224,
151
+ "learning_rate": 5.95342249592774e-06,
152
+ "loss": 0.0018794862031936645,
153
+ "step": 6500
154
+ },
155
+ {
156
+ "epoch": 4.551365409622887,
157
+ "grad_norm": 0.11848070472478867,
158
+ "learning_rate": 2.0311860540618234e-06,
159
+ "loss": 0.0014441205263137817,
160
+ "step": 7000
161
+ },
162
+ {
163
+ "epoch": 4.876462938881664,
164
+ "grad_norm": 0.02074478194117546,
165
+ "learning_rate": 1.561696828829895e-07,
166
+ "loss": 0.0018407306671142578,
167
+ "step": 7500
168
+ },
169
+ {
170
+ "epoch": 5.0,
171
+ "eval_accuracy": 0.9570552147239264,
172
+ "eval_f1": 0.9548942768021744,
173
+ "eval_loss": 0.026033323258161545,
174
+ "eval_map@3": 0.9740967961826857,
175
+ "eval_precision": 0.9566771628265378,
176
+ "eval_recall": 0.9542946341379102,
177
+ "eval_runtime": 1.1949,
178
+ "eval_samples_per_second": 2046.129,
179
+ "eval_steps_per_second": 16.737,
180
+ "step": 7690
181
+ }
182
+ ],
183
+ "logging_steps": 500,
184
+ "max_steps": 7690,
185
+ "num_input_tokens_seen": 0,
186
+ "num_train_epochs": 5,
187
+ "save_steps": 500,
188
+ "stateful_callbacks": {
189
+ "TrainerControl": {
190
+ "args": {
191
+ "should_epoch_stop": false,
192
+ "should_evaluate": false,
193
+ "should_log": false,
194
+ "should_save": true,
195
+ "should_training_stop": true
196
+ },
197
+ "attributes": {}
198
+ }
199
+ },
200
+ "total_flos": 1689212478350430.0,
201
+ "train_batch_size": 128,
202
+ "trial_name": null,
203
+ "trial_params": null
204
+ }
fold_3/checkpoint-7690/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b9545f104192a44da00231b10713c16bdded44d326f7747f13ffd7e4a78ca44
3
+ size 5265
fold_3/config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "DebertaV2ForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "bos_token_id": 1,
7
+ "dtype": "float32",
8
+ "eos_token_id": 2,
9
+ "hidden_act": "gelu",
10
+ "hidden_dropout_prob": 0.1,
11
+ "hidden_size": 256,
12
+ "id2label": {
13
+ "0": "LABEL_0"
14
+ },
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 1024,
17
+ "label2id": {
18
+ "LABEL_0": 0
19
+ },
20
+ "layer_norm_eps": 1e-07,
21
+ "legacy": true,
22
+ "max_position_embeddings": 512,
23
+ "max_relative_positions": -1,
24
+ "model_type": "deberta-v2",
25
+ "norm_rel_ebd": "layer_norm",
26
+ "num_attention_heads": 4,
27
+ "num_hidden_layers": 4,
28
+ "pad_token_id": 0,
29
+ "pooler_dropout": 0,
30
+ "pooler_hidden_act": "gelu",
31
+ "pooler_hidden_size": 256,
32
+ "pos_att_type": [
33
+ "p2c",
34
+ "c2p"
35
+ ],
36
+ "position_biased_input": false,
37
+ "position_buckets": 256,
38
+ "relative_attention": true,
39
+ "share_att_key": true,
40
+ "tie_word_embeddings": true,
41
+ "transformers_version": "5.0.0",
42
+ "type_vocab_size": 0,
43
+ "use_cache": false,
44
+ "vocab_size": 128100
45
+ }
fold_3/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cb3be835eafad2676bcfebc17da3c91b895f05235fe1bd6ac170c2996a6252d3
3
+ size 144611964
fold_3/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
fold_3/tokenizer_config.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": true,
3
+ "backend": "tokenizers",
4
+ "bos_token": "[CLS]",
5
+ "cls_token": "[CLS]",
6
+ "do_lower_case": false,
7
+ "eos_token": "[SEP]",
8
+ "extra_special_tokens": [
9
+ "[PAD]",
10
+ "[CLS]",
11
+ "[SEP]"
12
+ ],
13
+ "is_local": false,
14
+ "mask_token": "[MASK]",
15
+ "model_max_length": 1000000000000000019884624838656,
16
+ "pad_token": "[PAD]",
17
+ "sep_token": "[SEP]",
18
+ "split_by_punct": false,
19
+ "tokenizer_class": "DebertaV2Tokenizer",
20
+ "unk_id": 3,
21
+ "unk_token": "[UNK]",
22
+ "vocab_type": "spm"
23
+ }
fold_3/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b9545f104192a44da00231b10713c16bdded44d326f7747f13ffd7e4a78ca44
3
+ size 5265