Hugging Face's logo

ogoshi2000
/
stance-nystromformer

Text Classification
Transformers
PyTorch
nystromformer
Inference Endpoints
Model card Files Community
ogoshi2000 commited on
Commit
e359990
•
1 Parent(s): a2a0833

Upload 10 files

Browse files
Files changed (10) hide show
  1. config.json +45 -0
  2. merges.txt +0 -0
  3. pytorch_model.bin +3 -0
  4. run_config.json +50 -0
  5. special_tokens_map.json +51 -0
  6. tokenizer.json +0 -0
  7. tokenizer_config.json +64 -0
  8. trainer_state.json +195 -0
  9. training_args.bin +3 -0
  10. vocab.json +0 -0
config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "LennartKeller/nystromformer-gottbert-base-8192",
3
+ "architectures": [
4
+ "NystromformerForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "conv_kernel_size": 65,
10
+ "eos_token_id": 2,
11
+ "gradient_checkpointing": false,
12
+ "hidden_act": "gelu",
13
+ "hidden_dropout_prob": 0.1,
14
+ "hidden_size": 768,
15
+ "id2label": {
16
+ "0": "LABEL_0",
17
+ "1": "LABEL_1",
18
+ "2": "LABEL_2",
19
+ "3": "LABEL_3"
20
+ },
21
+ "initializer_range": 0.02,
22
+ "intermediate_size": 3072,
23
+ "inv_coeff_init_option": false,
24
+ "label2id": {
25
+ "LABEL_0": 0,
26
+ "LABEL_1": 1,
27
+ "LABEL_2": 2,
28
+ "LABEL_3": 3
29
+ },
30
+ "layer_norm_eps": 1e-05,
31
+ "max_position_embeddings": 8192,
32
+ "model_type": "nystromformer",
33
+ "num_attention_heads": 12,
34
+ "num_hidden_layers": 12,
35
+ "num_landmarks": 64,
36
+ "pad_token_id": 1,
37
+ "position_embedding_type": "absolute",
38
+ "problem_type": "single_label_classification",
39
+ "segment_means_seq_len": 64,
40
+ "torch_dtype": "float32",
41
+ "transformers_version": "4.23.1",
42
+ "type_vocab_size": 1,
43
+ "use_cache": true,
44
+ "vocab_size": 52009
45
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba04dc85ea3bbbb275d3c3de1199e2ec42ecab655fe79a3fbe8998bedde3800b
3
+ size 527720853
run_config.json ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_CHeeSEArguments": "CHeeSEArguments",
3
+ "model_name_or_path": "LennartKeller/nystromformer-gottbert-base-8192",
4
+ "dataset_path": "./dataset.py",
5
+ "metric_path": "./metric.py",
6
+ "task": "stance_detection",
7
+ "first_sentence_inputs": [
8
+ "question"
9
+ ],
10
+ "second_sentence_inputs": [
11
+ "title",
12
+ "snippet",
13
+ "paragraphs"
14
+ ],
15
+ "labels_to_predict": [
16
+ "stance"
17
+ ],
18
+ "_CHeeSETrainingArguments": "CHeeSETrainingArguments",
19
+ "do_train": true,
20
+ "do_eval": true,
21
+ "do_predict": true,
22
+ "do_cross_validation": false,
23
+ "output_dir": "/netscratch/schnitzler/nystromformer-8",
24
+ "logging_dir": "/netscratch/schnitzler/nystromformer-8",
25
+ "save_total_limit": 3,
26
+ "log_to_file": true,
27
+ "logging_strategy": "steps",
28
+ "logging_steps": 50,
29
+ "cross_validation_folds": 5,
30
+ "save_steps": 1000,
31
+ "evaluation_strategy": "steps",
32
+ "eval_steps": 500,
33
+ "seed": 121,
34
+ "num_train_epochs": 4,
35
+ "per_device_train_batch_size": 8,
36
+ "per_device_eval_batch_size": 8,
37
+ "learning_rate": 3e-05,
38
+ "prediction_csv_kwargs": {
39
+ "index": false
40
+ },
41
+ "prediction_columns_to_include": [
42
+ "title",
43
+ "snippet",
44
+ "paragraphs",
45
+ "question",
46
+ "labels",
47
+ "stance"
48
+ ],
49
+ "overwrite_output_dir": true
50
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "cls_token": {
10
+ "content": "<s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "eos_token": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": true,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "mask_token": {
24
+ "content": "<mask>",
25
+ "lstrip": true,
26
+ "normalized": true,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ },
30
+ "pad_token": {
31
+ "content": "<pad>",
32
+ "lstrip": false,
33
+ "normalized": true,
34
+ "rstrip": false,
35
+ "single_word": false
36
+ },
37
+ "sep_token": {
38
+ "content": "</s>",
39
+ "lstrip": false,
40
+ "normalized": true,
41
+ "rstrip": false,
42
+ "single_word": false
43
+ },
44
+ "unk_token": {
45
+ "content": "<unk>",
46
+ "lstrip": false,
47
+ "normalized": true,
48
+ "rstrip": false,
49
+ "single_word": false
50
+ }
51
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "bos_token": {
4
+ "__type": "AddedToken",
5
+ "content": "<s>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false
10
+ },
11
+ "cls_token": {
12
+ "__type": "AddedToken",
13
+ "content": "<s>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false
18
+ },
19
+ "eos_token": {
20
+ "__type": "AddedToken",
21
+ "content": "</s>",
22
+ "lstrip": false,
23
+ "normalized": true,
24
+ "rstrip": false,
25
+ "single_word": false
26
+ },
27
+ "errors": "replace",
28
+ "mask_token": {
29
+ "__type": "AddedToken",
30
+ "content": "<mask>",
31
+ "lstrip": true,
32
+ "normalized": true,
33
+ "rstrip": false,
34
+ "single_word": false
35
+ },
36
+ "name_or_path": "LennartKeller/nystromformer-gottbert-base-8192",
37
+ "pad_token": {
38
+ "__type": "AddedToken",
39
+ "content": "<pad>",
40
+ "lstrip": false,
41
+ "normalized": true,
42
+ "rstrip": false,
43
+ "single_word": false
44
+ },
45
+ "sep_token": {
46
+ "__type": "AddedToken",
47
+ "content": "</s>",
48
+ "lstrip": false,
49
+ "normalized": true,
50
+ "rstrip": false,
51
+ "single_word": false
52
+ },
53
+ "special_tokens_map_file": null,
54
+ "tokenizer_class": "RobertaTokenizer",
55
+ "trim_offsets": true,
56
+ "unk_token": {
57
+ "__type": "AddedToken",
58
+ "content": "<unk>",
59
+ "lstrip": false,
60
+ "normalized": true,
61
+ "rstrip": false,
62
+ "single_word": false
63
+ }
64
+ }
trainer_state.json ADDED
@@ -0,0 +1,195 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 4.0,
5
+ "global_step": 1196,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.17,
12
+ "learning_rate": 2.874581939799331e-05,
13
+ "loss": 1.3059,
14
+ "step": 50
15
+ },
16
+ {
17
+ "epoch": 0.33,
18
+ "learning_rate": 2.7491638795986622e-05,
19
+ "loss": 1.2769,
20
+ "step": 100
21
+ },
22
+ {
23
+ "epoch": 0.5,
24
+ "learning_rate": 2.6237458193979933e-05,
25
+ "loss": 1.2016,
26
+ "step": 150
27
+ },
28
+ {
29
+ "epoch": 0.67,
30
+ "learning_rate": 2.4983277591973243e-05,
31
+ "loss": 1.2823,
32
+ "step": 200
33
+ },
34
+ {
35
+ "epoch": 0.84,
36
+ "learning_rate": 2.3729096989966557e-05,
37
+ "loss": 1.2613,
38
+ "step": 250
39
+ },
40
+ {
41
+ "epoch": 1.0,
42
+ "learning_rate": 2.2474916387959865e-05,
43
+ "loss": 1.2375,
44
+ "step": 300
45
+ },
46
+ {
47
+ "epoch": 1.17,
48
+ "learning_rate": 2.122073578595318e-05,
49
+ "loss": 1.2233,
50
+ "step": 350
51
+ },
52
+ {
53
+ "epoch": 1.34,
54
+ "learning_rate": 1.996655518394649e-05,
55
+ "loss": 1.2468,
56
+ "step": 400
57
+ },
58
+ {
59
+ "epoch": 1.51,
60
+ "learning_rate": 1.87123745819398e-05,
61
+ "loss": 1.1199,
62
+ "step": 450
63
+ },
64
+ {
65
+ "epoch": 1.67,
66
+ "learning_rate": 1.745819397993311e-05,
67
+ "loss": 1.0252,
68
+ "step": 500
69
+ },
70
+ {
71
+ "epoch": 1.67,
72
+ "eval_accuracy": 0.55625,
73
+ "eval_f1_macro": 0.31421845574387947,
74
+ "eval_f1_micro": 0.55625,
75
+ "eval_f1_unweighted_diskutierend": 0.0,
76
+ "eval_f1_unweighted_ja-daf\u00fcr": 0.5111111111111111,
77
+ "eval_f1_unweighted_kein-bezug": 0.7457627118644068,
78
+ "eval_f1_unweighted_nein-dagegen": 0.0,
79
+ "eval_f1_weighted": 0.4473987758945386,
80
+ "eval_loss": 1.1650867462158203,
81
+ "eval_runtime": 23.2791,
82
+ "eval_samples_per_second": 6.873,
83
+ "eval_steps_per_second": 0.859,
84
+ "step": 500
85
+ },
86
+ {
87
+ "epoch": 1.84,
88
+ "learning_rate": 1.620401337792642e-05,
89
+ "loss": 1.0867,
90
+ "step": 550
91
+ },
92
+ {
93
+ "epoch": 2.01,
94
+ "learning_rate": 1.4949832775919733e-05,
95
+ "loss": 1.0217,
96
+ "step": 600
97
+ },
98
+ {
99
+ "epoch": 2.17,
100
+ "learning_rate": 1.3695652173913042e-05,
101
+ "loss": 0.8828,
102
+ "step": 650
103
+ },
104
+ {
105
+ "epoch": 2.34,
106
+ "learning_rate": 1.2441471571906355e-05,
107
+ "loss": 0.9718,
108
+ "step": 700
109
+ },
110
+ {
111
+ "epoch": 2.51,
112
+ "learning_rate": 1.1187290969899665e-05,
113
+ "loss": 0.9971,
114
+ "step": 750
115
+ },
116
+ {
117
+ "epoch": 2.68,
118
+ "learning_rate": 9.933110367892978e-06,
119
+ "loss": 0.8862,
120
+ "step": 800
121
+ },
122
+ {
123
+ "epoch": 2.84,
124
+ "learning_rate": 8.678929765886286e-06,
125
+ "loss": 0.9743,
126
+ "step": 850
127
+ },
128
+ {
129
+ "epoch": 3.01,
130
+ "learning_rate": 7.424749163879599e-06,
131
+ "loss": 0.9939,
132
+ "step": 900
133
+ },
134
+ {
135
+ "epoch": 3.18,
136
+ "learning_rate": 6.17056856187291e-06,
137
+ "loss": 0.7587,
138
+ "step": 950
139
+ },
140
+ {
141
+ "epoch": 3.34,
142
+ "learning_rate": 4.916387959866221e-06,
143
+ "loss": 0.8624,
144
+ "step": 1000
145
+ },
146
+ {
147
+ "epoch": 3.34,
148
+ "eval_accuracy": 0.6625,
149
+ "eval_f1_macro": 0.4798663717403574,
150
+ "eval_f1_micro": 0.6625,
151
+ "eval_f1_unweighted_diskutierend": 0.5569620253164557,
152
+ "eval_f1_unweighted_ja-daf\u00fcr": 0.5217391304347826,
153
+ "eval_f1_unweighted_kein-bezug": 0.8407643312101911,
154
+ "eval_f1_unweighted_nein-dagegen": 0.0,
155
+ "eval_f1_weighted": 0.6282338774980808,
156
+ "eval_loss": 0.8862520456314087,
157
+ "eval_runtime": 23.317,
158
+ "eval_samples_per_second": 6.862,
159
+ "eval_steps_per_second": 0.858,
160
+ "step": 1000
161
+ },
162
+ {
163
+ "epoch": 3.51,
164
+ "learning_rate": 3.662207357859532e-06,
165
+ "loss": 0.8056,
166
+ "step": 1050
167
+ },
168
+ {
169
+ "epoch": 3.68,
170
+ "learning_rate": 2.408026755852843e-06,
171
+ "loss": 0.8175,
172
+ "step": 1100
173
+ },
174
+ {
175
+ "epoch": 3.85,
176
+ "learning_rate": 1.153846153846154e-06,
177
+ "loss": 0.7386,
178
+ "step": 1150
179
+ },
180
+ {
181
+ "epoch": 4.0,
182
+ "step": 1196,
183
+ "total_flos": 2.0142135211720704e+16,
184
+ "train_loss": 1.0348390400609044,
185
+ "train_runtime": 5404.336,
186
+ "train_samples_per_second": 1.77,
187
+ "train_steps_per_second": 0.221
188
+ }
189
+ ],
190
+ "max_steps": 1196,
191
+ "num_train_epochs": 4,
192
+ "total_flos": 2.0142135211720704e+16,
193
+ "trial_name": null,
194
+ "trial_params": null
195
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3161bba896202214abd8449fb048896500bd0ca0d5de801b2a5121718a245c6
3
+ size 3631
vocab.json ADDED
The diff for this file is too large to render. See raw diff