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
4be8351
·
verified ·
1 Parent(s): d0fb9f7

Upload folder using huggingface_hub

Browse files
Files changed (15) hide show
  1. fold_4/checkpoint-6152/config.json +45 -0
  2. fold_4/checkpoint-6152/model.safetensors +3 -0
  3. fold_4/checkpoint-6152/optimizer.pt +3 -0
  4. fold_4/checkpoint-6152/rng_state.pth +3 -0
  5. fold_4/checkpoint-6152/scaler.pt +3 -0
  6. fold_4/checkpoint-6152/scheduler.pt +3 -0
  7. fold_4/checkpoint-6152/tokenizer.json +0 -0
  8. fold_4/checkpoint-6152/tokenizer_config.json +23 -0
  9. fold_4/checkpoint-6152/trainer_state.json +170 -0
  10. fold_4/checkpoint-6152/training_args.bin +3 -0
  11. fold_4/config.json +45 -0
  12. fold_4/model.safetensors +3 -0
  13. fold_4/tokenizer.json +0 -0
  14. fold_4/tokenizer_config.json +23 -0
  15. fold_4/training_args.bin +3 -0
fold_4/checkpoint-6152/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_4/checkpoint-6152/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f3d2df966646812e713455d5ddf6d2d91b9f2fd5c1800d106edff104d6dadcff
3
+ size 144611964
fold_4/checkpoint-6152/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ecf998fc0149612ece9913f9c3a6418e2f79a7aff5b6ed53c4df3c1475e5b202
3
+ size 289269963
fold_4/checkpoint-6152/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3443a16be77906cc521c2df01b69052a2e5f7299f4ab78852e01fd1be0695790
3
+ size 14645
fold_4/checkpoint-6152/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0dddf28f374d10bde94dde1fa8504246fdef9ed66570600da89db122cc13c53c
3
+ size 1383
fold_4/checkpoint-6152/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eca2a08d2077cc878df7ee971c8ab2c931c1a24ef33fc14844246c7edc1d91bc
3
+ size 1465
fold_4/checkpoint-6152/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
fold_4/checkpoint-6152/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_4/checkpoint-6152/trainer_state.json ADDED
@@ -0,0 +1,170 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 6152,
3
+ "best_metric": 0.9689207650273224,
4
+ "best_model_checkpoint": "/kaggle/working/QSolver_Checkpoints/scratch_fold_4/checkpoint-6152",
5
+ "epoch": 4.0,
6
+ "eval_steps": 500,
7
+ "global_step": 6152,
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": 0.0843004360795021,
15
+ "learning_rate": 9.931967777272626e-05,
16
+ "loss": 0.04558158111572266,
17
+ "step": 500
18
+ },
19
+ {
20
+ "epoch": 0.6501950585175552,
21
+ "grad_norm": 0.2093271017074585,
22
+ "learning_rate": 9.657816844825987e-05,
23
+ "loss": 0.013354792594909667,
24
+ "step": 1000
25
+ },
26
+ {
27
+ "epoch": 0.9752925877763329,
28
+ "grad_norm": 0.11299759149551392,
29
+ "learning_rate": 9.184878922543432e-05,
30
+ "loss": 0.010161470413208008,
31
+ "step": 1500
32
+ },
33
+ {
34
+ "epoch": 1.0,
35
+ "eval_accuracy": 0.8647540983606558,
36
+ "eval_f1": 0.8595073096090298,
37
+ "eval_loss": 0.12974396347999573,
38
+ "eval_map@3": 0.9159836065573768,
39
+ "eval_precision": 0.8623281377663259,
40
+ "eval_recall": 0.8609769258874117,
41
+ "eval_runtime": 1.2468,
42
+ "eval_samples_per_second": 1957.054,
43
+ "eval_steps_per_second": 16.041,
44
+ "step": 1538
45
+ },
46
+ {
47
+ "epoch": 1.3003901170351106,
48
+ "grad_norm": 0.07239209860563278,
49
+ "learning_rate": 8.533338124374017e-05,
50
+ "loss": 0.006697566509246826,
51
+ "step": 2000
52
+ },
53
+ {
54
+ "epoch": 1.6254876462938883,
55
+ "grad_norm": 0.06487059593200684,
56
+ "learning_rate": 7.731001003975445e-05,
57
+ "loss": 0.005165702342987061,
58
+ "step": 2500
59
+ },
60
+ {
61
+ "epoch": 1.9505851755526658,
62
+ "grad_norm": 0.1278851479291916,
63
+ "learning_rate": 6.81210982254242e-05,
64
+ "loss": 0.004400352001190186,
65
+ "step": 3000
66
+ },
67
+ {
68
+ "epoch": 2.0,
69
+ "eval_accuracy": 0.9098360655737705,
70
+ "eval_f1": 0.9061041443884875,
71
+ "eval_loss": 0.06499335169792175,
72
+ "eval_map@3": 0.9508196721311474,
73
+ "eval_precision": 0.9092675846186282,
74
+ "eval_recall": 0.9038964615946714,
75
+ "eval_runtime": 1.2501,
76
+ "eval_samples_per_second": 1951.848,
77
+ "eval_steps_per_second": 15.999,
78
+ "step": 3076
79
+ },
80
+ {
81
+ "epoch": 2.2756827048114436,
82
+ "grad_norm": 0.03591925650835037,
83
+ "learning_rate": 5.815881152565712e-05,
84
+ "loss": 0.0037573509216308593,
85
+ "step": 3500
86
+ },
87
+ {
88
+ "epoch": 2.6007802340702213,
89
+ "grad_norm": 0.05408729612827301,
90
+ "learning_rate": 4.7848321872047936e-05,
91
+ "loss": 0.0034550411701202394,
92
+ "step": 4000
93
+ },
94
+ {
95
+ "epoch": 2.925877763328999,
96
+ "grad_norm": 0.4929114878177643,
97
+ "learning_rate": 3.7629661852879294e-05,
98
+ "loss": 0.0032080700397491453,
99
+ "step": 4500
100
+ },
101
+ {
102
+ "epoch": 3.0,
103
+ "eval_accuracy": 0.9405737704918032,
104
+ "eval_f1": 0.9378624397423584,
105
+ "eval_loss": 0.05050104111433029,
106
+ "eval_map@3": 0.965846994535519,
107
+ "eval_precision": 0.9426319971784787,
108
+ "eval_recall": 0.9355405038269489,
109
+ "eval_runtime": 1.2723,
110
+ "eval_samples_per_second": 1917.811,
111
+ "eval_steps_per_second": 15.72,
112
+ "step": 4614
113
+ },
114
+ {
115
+ "epoch": 3.250975292587776,
116
+ "grad_norm": 0.060536615550518036,
117
+ "learning_rate": 2.7938944937838923e-05,
118
+ "loss": 0.002462742805480957,
119
+ "step": 5000
120
+ },
121
+ {
122
+ "epoch": 3.576072821846554,
123
+ "grad_norm": 0.04568943753838539,
124
+ "learning_rate": 1.918975296345461e-05,
125
+ "loss": 0.0025600173473358154,
126
+ "step": 5500
127
+ },
128
+ {
129
+ "epoch": 3.9011703511053315,
130
+ "grad_norm": 0.24042949080467224,
131
+ "learning_rate": 1.1755485226870489e-05,
132
+ "loss": 0.002174916982650757,
133
+ "step": 6000
134
+ },
135
+ {
136
+ "epoch": 4.0,
137
+ "eval_accuracy": 0.9487704918032787,
138
+ "eval_f1": 0.9471295609826346,
139
+ "eval_loss": 0.031837400048971176,
140
+ "eval_map@3": 0.9689207650273224,
141
+ "eval_precision": 0.9517005633943976,
142
+ "eval_recall": 0.9450643133507584,
143
+ "eval_runtime": 1.2444,
144
+ "eval_samples_per_second": 1960.806,
145
+ "eval_steps_per_second": 16.072,
146
+ "step": 6152
147
+ }
148
+ ],
149
+ "logging_steps": 500,
150
+ "max_steps": 7690,
151
+ "num_input_tokens_seen": 0,
152
+ "num_train_epochs": 5,
153
+ "save_steps": 500,
154
+ "stateful_callbacks": {
155
+ "TrainerControl": {
156
+ "args": {
157
+ "should_epoch_stop": false,
158
+ "should_evaluate": false,
159
+ "should_log": false,
160
+ "should_save": true,
161
+ "should_training_stop": false
162
+ },
163
+ "attributes": {}
164
+ }
165
+ },
166
+ "total_flos": 1342224416885616.0,
167
+ "train_batch_size": 128,
168
+ "trial_name": null,
169
+ "trial_params": null
170
+ }
fold_4/checkpoint-6152/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23d8d42a3470ddef9e1cd458014ab5ce8802b119883f8c1b85723501678cea64
3
+ size 5265
fold_4/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_4/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f3d2df966646812e713455d5ddf6d2d91b9f2fd5c1800d106edff104d6dadcff
3
+ size 144611964
fold_4/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
fold_4/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_4/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23d8d42a3470ddef9e1cd458014ab5ce8802b119883f8c1b85723501678cea64
3
+ size 5265