Hugging Face's logo

mehrshadk
/
pet-DEBERTL-seed791-lr2e-05

Safetensors
deberta
Model card Files Community
mehrshadk commited on
Commit
f8ce86e
·
verified ·
1 Parent(s): 192b003

Training in progress, epoch 3, checkpoint

Browse files
Files changed (4) hide show
  1. checkpoint-300/config.json +33 -0
  2. checkpoint-300/model.safetensors +3 -0
  3. checkpoint-300/trainer_state.json +171 -0
  4. checkpoint-300/training_args.bin +3 -0
checkpoint-300/config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/deberta-large",
3
+ "architectures": [
4
+ "DebertaForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "hidden_act": "gelu",
8
+ "hidden_dropout_prob": 0.1,
9
+ "hidden_size": 1024,
10
+ "initializer_range": 0.02,
11
+ "intermediate_size": 4096,
12
+ "layer_norm_eps": 1e-07,
13
+ "legacy": true,
14
+ "max_position_embeddings": 512,
15
+ "max_relative_positions": -1,
16
+ "model_type": "deberta",
17
+ "num_attention_heads": 16,
18
+ "num_hidden_layers": 24,
19
+ "pad_token_id": 0,
20
+ "pooler_dropout": 0,
21
+ "pooler_hidden_act": "gelu",
22
+ "pooler_hidden_size": 1024,
23
+ "pos_att_type": [
24
+ "c2p",
25
+ "p2c"
26
+ ],
27
+ "position_biased_input": false,
28
+ "relative_attention": true,
29
+ "torch_dtype": "float32",
30
+ "transformers_version": "4.48.3",
31
+ "type_vocab_size": 0,
32
+ "vocab_size": 50265
33
+ }
checkpoint-300/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e99be98a6a9c7ba9ca7d73050c71fb07a9bf2983d97c2e0ad8040cd1f97bced4
3
+ size 1624907048
checkpoint-300/trainer_state.json ADDED
@@ -0,0 +1,171 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "eval_steps": 50,
6
+ "global_step": 300,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.5,
13
+ "grad_norm": 6.080199718475342,
14
+ "learning_rate": 1.8333333333333333e-05,
15
+ "loss": 0.5304,
16
+ "step": 50
17
+ },
18
+ {
19
+ "epoch": 0.5,
20
+ "eval_pet_AUC": 0.8846723646723647,
21
+ "eval_pet_accuracy": 0.825,
22
+ "eval_pet_f1": 0.7569444444444444,
23
+ "eval_pet_loss": 0.4055028557777405,
24
+ "eval_pet_macro_f1": 0.8101128472222222,
25
+ "eval_pet_micro_f1": 0.825,
26
+ "eval_pet_precision": 0.689873417721519,
27
+ "eval_pet_recall": 0.8384615384615385,
28
+ "eval_pet_runtime": 0.9669,
29
+ "eval_pet_samples_per_second": 413.704,
30
+ "eval_pet_specificity": 0.8185185185185185,
31
+ "eval_pet_steps_per_second": 1.034,
32
+ "step": 50
33
+ },
34
+ {
35
+ "epoch": 1.0,
36
+ "grad_norm": 7.429731369018555,
37
+ "learning_rate": 1.6666666666666667e-05,
38
+ "loss": 0.4825,
39
+ "step": 100
40
+ },
41
+ {
42
+ "epoch": 1.0,
43
+ "eval_pet_AUC": 0.9121652421652421,
44
+ "eval_pet_accuracy": 0.8325,
45
+ "eval_pet_f1": 0.7598566308243727,
46
+ "eval_pet_loss": 0.35728055238723755,
47
+ "eval_pet_macro_f1": 0.8156288912279253,
48
+ "eval_pet_micro_f1": 0.8325,
49
+ "eval_pet_precision": 0.7114093959731543,
50
+ "eval_pet_recall": 0.8153846153846154,
51
+ "eval_pet_runtime": 0.9571,
52
+ "eval_pet_samples_per_second": 417.916,
53
+ "eval_pet_specificity": 0.8407407407407408,
54
+ "eval_pet_steps_per_second": 1.045,
55
+ "step": 100
56
+ },
57
+ {
58
+ "epoch": 1.5,
59
+ "grad_norm": 7.952208042144775,
60
+ "learning_rate": 1.5000000000000002e-05,
61
+ "loss": 0.3115,
62
+ "step": 150
63
+ },
64
+ {
65
+ "epoch": 1.5,
66
+ "eval_pet_AUC": 0.9031339031339031,
67
+ "eval_pet_accuracy": 0.85,
68
+ "eval_pet_f1": 0.7744360902255639,
69
+ "eval_pet_loss": 0.5131158828735352,
70
+ "eval_pet_macro_f1": 0.831038269831883,
71
+ "eval_pet_micro_f1": 0.85,
72
+ "eval_pet_precision": 0.7573529411764706,
73
+ "eval_pet_recall": 0.7923076923076923,
74
+ "eval_pet_runtime": 0.9601,
75
+ "eval_pet_samples_per_second": 416.631,
76
+ "eval_pet_specificity": 0.8777777777777778,
77
+ "eval_pet_steps_per_second": 1.042,
78
+ "step": 150
79
+ },
80
+ {
81
+ "epoch": 2.0,
82
+ "grad_norm": 3.0394058227539062,
83
+ "learning_rate": 1.3333333333333333e-05,
84
+ "loss": 0.2963,
85
+ "step": 200
86
+ },
87
+ {
88
+ "epoch": 2.0,
89
+ "eval_pet_AUC": 0.9202564102564103,
90
+ "eval_pet_accuracy": 0.8425,
91
+ "eval_pet_f1": 0.7622641509433963,
92
+ "eval_pet_loss": 0.3516935706138611,
93
+ "eval_pet_macro_f1": 0.822253570798801,
94
+ "eval_pet_micro_f1": 0.8425,
95
+ "eval_pet_precision": 0.7481481481481481,
96
+ "eval_pet_recall": 0.7769230769230769,
97
+ "eval_pet_runtime": 0.9543,
98
+ "eval_pet_samples_per_second": 419.175,
99
+ "eval_pet_specificity": 0.8740740740740741,
100
+ "eval_pet_steps_per_second": 1.048,
101
+ "step": 200
102
+ },
103
+ {
104
+ "epoch": 2.5,
105
+ "grad_norm": 13.06711483001709,
106
+ "learning_rate": 1.1666666666666668e-05,
107
+ "loss": 0.2196,
108
+ "step": 250
109
+ },
110
+ {
111
+ "epoch": 2.5,
112
+ "eval_pet_AUC": 0.9203703703703704,
113
+ "eval_pet_accuracy": 0.85,
114
+ "eval_pet_f1": 0.7902097902097902,
115
+ "eval_pet_loss": 0.4240196645259857,
116
+ "eval_pet_macro_f1": 0.8367391363500313,
117
+ "eval_pet_micro_f1": 0.85,
118
+ "eval_pet_precision": 0.7243589743589743,
119
+ "eval_pet_recall": 0.8692307692307693,
120
+ "eval_pet_runtime": 0.9601,
121
+ "eval_pet_samples_per_second": 416.603,
122
+ "eval_pet_specificity": 0.8407407407407408,
123
+ "eval_pet_steps_per_second": 1.042,
124
+ "step": 250
125
+ },
126
+ {
127
+ "epoch": 3.0,
128
+ "grad_norm": 18.414968490600586,
129
+ "learning_rate": 1e-05,
130
+ "loss": 0.1567,
131
+ "step": 300
132
+ },
133
+ {
134
+ "epoch": 3.0,
135
+ "eval_pet_AUC": 0.9211965811965812,
136
+ "eval_pet_accuracy": 0.8475,
137
+ "eval_pet_f1": 0.769811320754717,
138
+ "eval_pet_loss": 0.492168128490448,
139
+ "eval_pet_macro_f1": 0.827896314582966,
140
+ "eval_pet_micro_f1": 0.8475,
141
+ "eval_pet_precision": 0.7555555555555555,
142
+ "eval_pet_recall": 0.7846153846153846,
143
+ "eval_pet_runtime": 0.9572,
144
+ "eval_pet_samples_per_second": 417.89,
145
+ "eval_pet_specificity": 0.8777777777777778,
146
+ "eval_pet_steps_per_second": 1.045,
147
+ "step": 300
148
+ }
149
+ ],
150
+ "logging_steps": 50,
151
+ "max_steps": 600,
152
+ "num_input_tokens_seen": 0,
153
+ "num_train_epochs": 6,
154
+ "save_steps": 500,
155
+ "stateful_callbacks": {
156
+ "TrainerControl": {
157
+ "args": {
158
+ "should_epoch_stop": false,
159
+ "should_evaluate": false,
160
+ "should_log": false,
161
+ "should_save": true,
162
+ "should_training_stop": false
163
+ },
164
+ "attributes": {}
165
+ }
166
+ },
167
+ "total_flos": 407456319744000.0,
168
+ "train_batch_size": 16,
169
+ "trial_name": null,
170
+ "trial_params": null
171
+ }
checkpoint-300/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0764263474b97260fc37ee4a274e139015dd761304086c91f05087cd69a8986e
3
+ size 5432