xshubhamx commited on
Commit
38c4c9c
1 Parent(s): 3eb4668

Upload folder using huggingface_hub

Browse files
training_checkpoints/checkpoint-19290/config.json ADDED
@@ -0,0 +1,100 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/bart-base",
3
+ "activation_dropout": 0.1,
4
+ "activation_function": "gelu",
5
+ "add_bias_logits": false,
6
+ "add_final_layer_norm": false,
7
+ "architectures": [
8
+ "BartForSequenceClassification"
9
+ ],
10
+ "attention_dropout": 0.1,
11
+ "bos_token_id": 0,
12
+ "classif_dropout": 0.1,
13
+ "classifier_dropout": 0.0,
14
+ "d_model": 768,
15
+ "decoder_attention_heads": 12,
16
+ "decoder_ffn_dim": 3072,
17
+ "decoder_layerdrop": 0.0,
18
+ "decoder_layers": 6,
19
+ "decoder_start_token_id": 2,
20
+ "dropout": 0.1,
21
+ "early_stopping": true,
22
+ "encoder_attention_heads": 12,
23
+ "encoder_ffn_dim": 3072,
24
+ "encoder_layerdrop": 0.0,
25
+ "encoder_layers": 6,
26
+ "eos_token_id": 2,
27
+ "forced_bos_token_id": 0,
28
+ "forced_eos_token_id": 2,
29
+ "gradient_checkpointing": false,
30
+ "id2label": {
31
+ "0": "Issue",
32
+ "1": "Court Discourse",
33
+ "2": "Conclusion",
34
+ "3": "Precedent Analysis",
35
+ "4": "Section Analysis",
36
+ "5": "Argument by Petitioner",
37
+ "6": "Fact",
38
+ "7": "Argument by Respondent",
39
+ "8": "Ratio",
40
+ "9": "Appellant",
41
+ "10": "Respondent",
42
+ "11": "Argument by Appellant",
43
+ "12": "Petitioner",
44
+ "13": "Judge",
45
+ "14": "Argument by Defendant"
46
+ },
47
+ "init_std": 0.02,
48
+ "is_encoder_decoder": true,
49
+ "label2id": {
50
+ "Appellant": 9,
51
+ "Argument by Appellant": 11,
52
+ "Argument by Defendant": 14,
53
+ "Argument by Petitioner": 5,
54
+ "Argument by Respondent": 7,
55
+ "Conclusion": 2,
56
+ "Court Discourse": 1,
57
+ "Fact": 6,
58
+ "Issue": 0,
59
+ "Judge": 13,
60
+ "Petitioner": 12,
61
+ "Precedent Analysis": 3,
62
+ "Ratio": 8,
63
+ "Respondent": 10,
64
+ "Section Analysis": 4
65
+ },
66
+ "max_position_embeddings": 1024,
67
+ "model_type": "bart",
68
+ "no_repeat_ngram_size": 3,
69
+ "normalize_before": false,
70
+ "normalize_embedding": true,
71
+ "num_beams": 4,
72
+ "num_hidden_layers": 6,
73
+ "pad_token_id": 1,
74
+ "problem_type": "single_label_classification",
75
+ "scale_embedding": false,
76
+ "task_specific_params": {
77
+ "summarization": {
78
+ "length_penalty": 1.0,
79
+ "max_length": 128,
80
+ "min_length": 12,
81
+ "num_beams": 4
82
+ },
83
+ "summarization_cnn": {
84
+ "length_penalty": 2.0,
85
+ "max_length": 142,
86
+ "min_length": 56,
87
+ "num_beams": 4
88
+ },
89
+ "summarization_xsum": {
90
+ "length_penalty": 1.0,
91
+ "max_length": 62,
92
+ "min_length": 11,
93
+ "num_beams": 6
94
+ }
95
+ },
96
+ "torch_dtype": "float32",
97
+ "transformers_version": "4.35.2",
98
+ "use_cache": true,
99
+ "vocab_size": 50265
100
+ }
training_checkpoints/checkpoint-19290/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
training_checkpoints/checkpoint-19290/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e0334354144e16b97cf35c09bb5cfa789c41aca2c828dfef43d90ba82d1942a
3
+ size 560120380
training_checkpoints/checkpoint-19290/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:75398cb19be5b20eeb5296011f35214d0d0657101bedf9b712a7aff380737ee0
3
+ size 1120400122
training_checkpoints/checkpoint-19290/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d10a3b1bf64a3666e3f1703a3ea91dae514e5ac036bdbe5112b58ad98d5b4025
3
+ size 14244
training_checkpoints/checkpoint-19290/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d36f27ed9e7e0d74e671c3920e9ed835210da9888c14bcc42f5a1195c99458f1
3
+ size 1064
training_checkpoints/checkpoint-19290/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": true,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
training_checkpoints/checkpoint-19290/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
training_checkpoints/checkpoint-19290/tokenizer_config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "0": {
5
+ "content": "<s>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "1": {
13
+ "content": "<pad>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "2": {
21
+ "content": "</s>",
22
+ "lstrip": false,
23
+ "normalized": true,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "3": {
29
+ "content": "<unk>",
30
+ "lstrip": false,
31
+ "normalized": true,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "50264": {
37
+ "content": "<mask>",
38
+ "lstrip": true,
39
+ "normalized": true,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": true
43
+ }
44
+ },
45
+ "bos_token": "<s>",
46
+ "clean_up_tokenization_spaces": true,
47
+ "cls_token": "<s>",
48
+ "eos_token": "</s>",
49
+ "errors": "replace",
50
+ "mask_token": "<mask>",
51
+ "model_max_length": 1024,
52
+ "pad_token": "<pad>",
53
+ "sep_token": "</s>",
54
+ "tokenizer_class": "BartTokenizer",
55
+ "trim_offsets": true,
56
+ "unk_token": "<unk>"
57
+ }
training_checkpoints/checkpoint-19290/trainer_state.json ADDED
@@ -0,0 +1,907 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7780054113531673,
3
+ "best_model_checkpoint": "bart-base/checkpoint-10288",
4
+ "epoch": 30.0,
5
+ "eval_steps": 500,
6
+ "global_step": 19290,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.78,
13
+ "learning_rate": 4.871176775531364e-05,
14
+ "loss": 1.2028,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 1.0,
19
+ "eval_accuracy": 0.7598760650658405,
20
+ "eval_f1_macro": 0.6063644681337579,
21
+ "eval_f1_micro": 0.7598760650658405,
22
+ "eval_f1_weighted": 0.7489266744747745,
23
+ "eval_loss": 0.8429557681083679,
24
+ "eval_macro_fpr": 0.023226157303183535,
25
+ "eval_macro_sensitivity": 0.6366904609623596,
26
+ "eval_macro_specificity": 0.981694425945952,
27
+ "eval_precision": 0.7600880594489849,
28
+ "eval_precision_macro": 0.6003620126022547,
29
+ "eval_recall": 0.7598760650658405,
30
+ "eval_recall_macro": 0.6366904609623596,
31
+ "eval_runtime": 43.5934,
32
+ "eval_samples_per_second": 29.615,
33
+ "eval_steps_per_second": 3.716,
34
+ "eval_weighted_fpr": 0.02207348333808032,
35
+ "eval_weighted_sensitivity": 0.7598760650658405,
36
+ "eval_weighted_specificity": 0.9655403241234398,
37
+ "step": 643
38
+ },
39
+ {
40
+ "epoch": 1.56,
41
+ "learning_rate": 4.741575946086055e-05,
42
+ "loss": 0.715,
43
+ "step": 1000
44
+ },
45
+ {
46
+ "epoch": 2.0,
47
+ "eval_accuracy": 0.7931835786212239,
48
+ "eval_f1_macro": 0.7213629796675549,
49
+ "eval_f1_micro": 0.793183578621224,
50
+ "eval_f1_weighted": 0.7853477061503886,
51
+ "eval_loss": 0.7331996560096741,
52
+ "eval_macro_fpr": 0.018982515099395962,
53
+ "eval_macro_sensitivity": 0.7320748785168393,
54
+ "eval_macro_specificity": 0.9845140063777027,
55
+ "eval_precision": 0.8019650926098448,
56
+ "eval_precision_macro": 0.7386000214414028,
57
+ "eval_recall": 0.7931835786212239,
58
+ "eval_recall_macro": 0.7320748785168393,
59
+ "eval_runtime": 77.2393,
60
+ "eval_samples_per_second": 16.714,
61
+ "eval_steps_per_second": 2.097,
62
+ "eval_weighted_fpr": 0.018283914264192288,
63
+ "eval_weighted_sensitivity": 0.7931835786212239,
64
+ "eval_weighted_specificity": 0.9745265170443169,
65
+ "step": 1286
66
+ },
67
+ {
68
+ "epoch": 2.33,
69
+ "learning_rate": 4.611975116640747e-05,
70
+ "loss": 0.578,
71
+ "step": 1500
72
+ },
73
+ {
74
+ "epoch": 3.0,
75
+ "eval_accuracy": 0.7939581719597212,
76
+ "eval_f1_macro": 0.6998313050068915,
77
+ "eval_f1_micro": 0.7939581719597211,
78
+ "eval_f1_weighted": 0.79006368267302,
79
+ "eval_loss": 0.8045273423194885,
80
+ "eval_macro_fpr": 0.018545878882531782,
81
+ "eval_macro_sensitivity": 0.7068834234702027,
82
+ "eval_macro_specificity": 0.9847671550097226,
83
+ "eval_precision": 0.8075005963568469,
84
+ "eval_precision_macro": 0.7230632440844518,
85
+ "eval_recall": 0.7939581719597212,
86
+ "eval_recall_macro": 0.7068834234702027,
87
+ "eval_runtime": 86.4099,
88
+ "eval_samples_per_second": 14.94,
89
+ "eval_steps_per_second": 1.875,
90
+ "eval_weighted_fpr": 0.018199233716475097,
91
+ "eval_weighted_sensitivity": 0.7939581719597212,
92
+ "eval_weighted_specificity": 0.977549153186118,
93
+ "step": 1929
94
+ },
95
+ {
96
+ "epoch": 3.11,
97
+ "learning_rate": 4.482633488854329e-05,
98
+ "loss": 0.481,
99
+ "step": 2000
100
+ },
101
+ {
102
+ "epoch": 3.89,
103
+ "learning_rate": 4.3530326594090203e-05,
104
+ "loss": 0.3938,
105
+ "step": 2500
106
+ },
107
+ {
108
+ "epoch": 4.0,
109
+ "eval_accuracy": 0.8156467854376452,
110
+ "eval_f1_macro": 0.7369315728111729,
111
+ "eval_f1_micro": 0.8156467854376452,
112
+ "eval_f1_weighted": 0.8105168035589256,
113
+ "eval_loss": 0.8291025161743164,
114
+ "eval_macro_fpr": 0.016938074954137138,
115
+ "eval_macro_sensitivity": 0.7217605983180461,
116
+ "eval_macro_specificity": 0.9857842035419169,
117
+ "eval_precision": 0.8171220006778526,
118
+ "eval_precision_macro": 0.7937034642937142,
119
+ "eval_recall": 0.8156467854376452,
120
+ "eval_recall_macro": 0.7217605983180461,
121
+ "eval_runtime": 84.4366,
122
+ "eval_samples_per_second": 15.29,
123
+ "eval_steps_per_second": 1.919,
124
+ "eval_weighted_fpr": 0.01588785046728972,
125
+ "eval_weighted_sensitivity": 0.8156467854376452,
126
+ "eval_weighted_specificity": 0.9711162676911109,
127
+ "step": 2572
128
+ },
129
+ {
130
+ "epoch": 4.67,
131
+ "learning_rate": 4.2236910316226025e-05,
132
+ "loss": 0.3238,
133
+ "step": 3000
134
+ },
135
+ {
136
+ "epoch": 5.0,
137
+ "eval_accuracy": 0.7939581719597212,
138
+ "eval_f1_macro": 0.7360579650857649,
139
+ "eval_f1_micro": 0.7939581719597211,
140
+ "eval_f1_weighted": 0.7995200606905476,
141
+ "eval_loss": 0.8888681530952454,
142
+ "eval_macro_fpr": 0.018767728188278533,
143
+ "eval_macro_sensitivity": 0.7515467286591376,
144
+ "eval_macro_specificity": 0.9846803211687324,
145
+ "eval_precision": 0.8146370555192632,
146
+ "eval_precision_macro": 0.7464262358644205,
147
+ "eval_recall": 0.7939581719597212,
148
+ "eval_recall_macro": 0.7515467286591376,
149
+ "eval_runtime": 82.179,
150
+ "eval_samples_per_second": 15.71,
151
+ "eval_steps_per_second": 1.971,
152
+ "eval_weighted_fpr": 0.018199233716475097,
153
+ "eval_weighted_sensitivity": 0.7939581719597212,
154
+ "eval_weighted_specificity": 0.9762466455712672,
155
+ "step": 3215
156
+ },
157
+ {
158
+ "epoch": 5.44,
159
+ "learning_rate": 4.094090202177294e-05,
160
+ "loss": 0.246,
161
+ "step": 3500
162
+ },
163
+ {
164
+ "epoch": 6.0,
165
+ "eval_accuracy": 0.7955073586367157,
166
+ "eval_f1_macro": 0.7362210043644567,
167
+ "eval_f1_micro": 0.7955073586367158,
168
+ "eval_f1_weighted": 0.7945977979999089,
169
+ "eval_loss": 1.1628626585006714,
170
+ "eval_macro_fpr": 0.01863187379355594,
171
+ "eval_macro_sensitivity": 0.7599986341032186,
172
+ "eval_macro_specificity": 0.9846932142915057,
173
+ "eval_precision": 0.8066758597712774,
174
+ "eval_precision_macro": 0.7483163388808308,
175
+ "eval_recall": 0.7955073586367157,
176
+ "eval_recall_macro": 0.7599986341032186,
177
+ "eval_runtime": 82.1621,
178
+ "eval_samples_per_second": 15.713,
179
+ "eval_steps_per_second": 1.972,
180
+ "eval_weighted_fpr": 0.018030323726266905,
181
+ "eval_weighted_sensitivity": 0.7955073586367157,
182
+ "eval_weighted_specificity": 0.9748908557358684,
183
+ "step": 3858
184
+ },
185
+ {
186
+ "epoch": 6.22,
187
+ "learning_rate": 3.964489372731986e-05,
188
+ "loss": 0.228,
189
+ "step": 4000
190
+ },
191
+ {
192
+ "epoch": 7.0,
193
+ "learning_rate": 3.834888543286677e-05,
194
+ "loss": 0.1791,
195
+ "step": 4500
196
+ },
197
+ {
198
+ "epoch": 7.0,
199
+ "eval_accuracy": 0.8179705654531371,
200
+ "eval_f1_macro": 0.7736489492761546,
201
+ "eval_f1_micro": 0.817970565453137,
202
+ "eval_f1_weighted": 0.8146592022169725,
203
+ "eval_loss": 1.1354283094406128,
204
+ "eval_macro_fpr": 0.016454731140779742,
205
+ "eval_macro_sensitivity": 0.7696557438486226,
206
+ "eval_macro_specificity": 0.9861772746118799,
207
+ "eval_precision": 0.8150792023791334,
208
+ "eval_precision_macro": 0.7832166669456337,
209
+ "eval_recall": 0.8179705654531371,
210
+ "eval_recall_macro": 0.7696557438486226,
211
+ "eval_runtime": 85.4548,
212
+ "eval_samples_per_second": 15.107,
213
+ "eval_steps_per_second": 1.896,
214
+ "eval_weighted_fpr": 0.01564684732671949,
215
+ "eval_weighted_sensitivity": 0.8179705654531371,
216
+ "eval_weighted_specificity": 0.9746885537250626,
217
+ "step": 4501
218
+ },
219
+ {
220
+ "epoch": 7.78,
221
+ "learning_rate": 3.705546915500259e-05,
222
+ "loss": 0.1305,
223
+ "step": 5000
224
+ },
225
+ {
226
+ "epoch": 8.0,
227
+ "eval_accuracy": 0.8109992254066615,
228
+ "eval_f1_macro": 0.7368975448573482,
229
+ "eval_f1_micro": 0.8109992254066615,
230
+ "eval_f1_weighted": 0.808822910649843,
231
+ "eval_loss": 1.282464861869812,
232
+ "eval_macro_fpr": 0.016943107063520774,
233
+ "eval_macro_sensitivity": 0.7488577570944955,
234
+ "eval_macro_specificity": 0.9858309939135591,
235
+ "eval_precision": 0.8147632227505013,
236
+ "eval_precision_macro": 0.7422132710401315,
237
+ "eval_recall": 0.8109992254066615,
238
+ "eval_recall_macro": 0.7488577570944955,
239
+ "eval_runtime": 87.6049,
240
+ "eval_samples_per_second": 14.737,
241
+ "eval_steps_per_second": 1.849,
242
+ "eval_weighted_fpr": 0.016373641121997046,
243
+ "eval_weighted_sensitivity": 0.8109992254066615,
244
+ "eval_weighted_specificity": 0.9764656832967246,
245
+ "step": 5144
246
+ },
247
+ {
248
+ "epoch": 8.55,
249
+ "learning_rate": 3.575946086054951e-05,
250
+ "loss": 0.0924,
251
+ "step": 5500
252
+ },
253
+ {
254
+ "epoch": 9.0,
255
+ "eval_accuracy": 0.8040278853601859,
256
+ "eval_f1_macro": 0.7559721746710123,
257
+ "eval_f1_micro": 0.8040278853601859,
258
+ "eval_f1_weighted": 0.80145679310201,
259
+ "eval_loss": 1.421708583831787,
260
+ "eval_macro_fpr": 0.017802568540250287,
261
+ "eval_macro_sensitivity": 0.7809236600744539,
262
+ "eval_macro_specificity": 0.9853492039682624,
263
+ "eval_precision": 0.8113877598541575,
264
+ "eval_precision_macro": 0.7464644897074477,
265
+ "eval_recall": 0.8040278853601859,
266
+ "eval_recall_macro": 0.7809236600744539,
267
+ "eval_runtime": 89.5999,
268
+ "eval_samples_per_second": 14.408,
269
+ "eval_steps_per_second": 1.808,
270
+ "eval_weighted_fpr": 0.017111937774771727,
271
+ "eval_weighted_sensitivity": 0.8040278853601859,
272
+ "eval_weighted_specificity": 0.9762101741637506,
273
+ "step": 5787
274
+ },
275
+ {
276
+ "epoch": 9.33,
277
+ "learning_rate": 3.446345256609643e-05,
278
+ "loss": 0.0953,
279
+ "step": 6000
280
+ },
281
+ {
282
+ "epoch": 10.0,
283
+ "eval_accuracy": 0.8024786986831913,
284
+ "eval_f1_macro": 0.7661426626929498,
285
+ "eval_f1_micro": 0.8024786986831914,
286
+ "eval_f1_weighted": 0.8001228885030943,
287
+ "eval_loss": 1.5552372932434082,
288
+ "eval_macro_fpr": 0.018250858387380567,
289
+ "eval_macro_sensitivity": 0.7822186507003115,
290
+ "eval_macro_specificity": 0.9849087494473102,
291
+ "eval_precision": 0.8055922813865947,
292
+ "eval_precision_macro": 0.7702182213697708,
293
+ "eval_recall": 0.8024786986831913,
294
+ "eval_recall_macro": 0.7822186507003115,
295
+ "eval_runtime": 85.2534,
296
+ "eval_samples_per_second": 15.143,
297
+ "eval_steps_per_second": 1.9,
298
+ "eval_weighted_fpr": 0.017277593332881633,
299
+ "eval_weighted_sensitivity": 0.8024786986831913,
300
+ "eval_weighted_specificity": 0.9711525430264618,
301
+ "step": 6430
302
+ },
303
+ {
304
+ "epoch": 10.11,
305
+ "learning_rate": 3.316744427164334e-05,
306
+ "loss": 0.0668,
307
+ "step": 6500
308
+ },
309
+ {
310
+ "epoch": 10.89,
311
+ "learning_rate": 3.1871435977190256e-05,
312
+ "loss": 0.0617,
313
+ "step": 7000
314
+ },
315
+ {
316
+ "epoch": 11.0,
317
+ "eval_accuracy": 0.8040278853601859,
318
+ "eval_f1_macro": 0.7422420279485343,
319
+ "eval_f1_micro": 0.8040278853601859,
320
+ "eval_f1_weighted": 0.8069943234338898,
321
+ "eval_loss": 1.5223894119262695,
322
+ "eval_macro_fpr": 0.017602619829831197,
323
+ "eval_macro_sensitivity": 0.7512050029477755,
324
+ "eval_macro_specificity": 0.9853474796083734,
325
+ "eval_precision": 0.8144253833246468,
326
+ "eval_precision_macro": 0.745702567690635,
327
+ "eval_recall": 0.8040278853601859,
328
+ "eval_recall_macro": 0.7512050029477755,
329
+ "eval_runtime": 85.7004,
330
+ "eval_samples_per_second": 15.064,
331
+ "eval_steps_per_second": 1.89,
332
+ "eval_weighted_fpr": 0.017111937774771727,
333
+ "eval_weighted_sensitivity": 0.8040278853601859,
334
+ "eval_weighted_specificity": 0.9761843087654142,
335
+ "step": 7073
336
+ },
337
+ {
338
+ "epoch": 11.66,
339
+ "learning_rate": 3.057542768273717e-05,
340
+ "loss": 0.0582,
341
+ "step": 7500
342
+ },
343
+ {
344
+ "epoch": 12.0,
345
+ "eval_accuracy": 0.7970565453137103,
346
+ "eval_f1_macro": 0.7465072775190384,
347
+ "eval_f1_micro": 0.7970565453137104,
348
+ "eval_f1_weighted": 0.8021748784206515,
349
+ "eval_loss": 1.6427809000015259,
350
+ "eval_macro_fpr": 0.01832354936918133,
351
+ "eval_macro_sensitivity": 0.7654834054509102,
352
+ "eval_macro_specificity": 0.9849466228972559,
353
+ "eval_precision": 0.8148454054398186,
354
+ "eval_precision_macro": 0.7470312881962948,
355
+ "eval_recall": 0.7970565453137103,
356
+ "eval_recall_macro": 0.7654834054509102,
357
+ "eval_runtime": 95.4565,
358
+ "eval_samples_per_second": 13.524,
359
+ "eval_steps_per_second": 1.697,
360
+ "eval_weighted_fpr": 0.017862012544314153,
361
+ "eval_weighted_sensitivity": 0.7970565453137103,
362
+ "eval_weighted_specificity": 0.9771427981451267,
363
+ "step": 7716
364
+ },
365
+ {
366
+ "epoch": 12.44,
367
+ "learning_rate": 2.9279419388284086e-05,
368
+ "loss": 0.0511,
369
+ "step": 8000
370
+ },
371
+ {
372
+ "epoch": 13.0,
373
+ "eval_accuracy": 0.8195197521301317,
374
+ "eval_f1_macro": 0.7586168211948496,
375
+ "eval_f1_micro": 0.8195197521301317,
376
+ "eval_f1_weighted": 0.8186828900418365,
377
+ "eval_loss": 1.4951962232589722,
378
+ "eval_macro_fpr": 0.016182326988526896,
379
+ "eval_macro_sensitivity": 0.7580273333502953,
380
+ "eval_macro_specificity": 0.9863597979839632,
381
+ "eval_precision": 0.8208277188287908,
382
+ "eval_precision_macro": 0.7645206618886077,
383
+ "eval_recall": 0.8195197521301317,
384
+ "eval_recall_macro": 0.7580273333502953,
385
+ "eval_runtime": 85.8279,
386
+ "eval_samples_per_second": 15.042,
387
+ "eval_steps_per_second": 1.887,
388
+ "eval_weighted_fpr": 0.01548687271518777,
389
+ "eval_weighted_sensitivity": 0.8195197521301317,
390
+ "eval_weighted_specificity": 0.9758772176293167,
391
+ "step": 8359
392
+ },
393
+ {
394
+ "epoch": 13.22,
395
+ "learning_rate": 2.7983411093831004e-05,
396
+ "loss": 0.0403,
397
+ "step": 8500
398
+ },
399
+ {
400
+ "epoch": 14.0,
401
+ "learning_rate": 2.668740279937792e-05,
402
+ "loss": 0.0476,
403
+ "step": 9000
404
+ },
405
+ {
406
+ "epoch": 14.0,
407
+ "eval_accuracy": 0.7970565453137103,
408
+ "eval_f1_macro": 0.7416650578915932,
409
+ "eval_f1_micro": 0.7970565453137104,
410
+ "eval_f1_weighted": 0.7913296065268454,
411
+ "eval_loss": 1.7132102251052856,
412
+ "eval_macro_fpr": 0.018853564171086353,
413
+ "eval_macro_sensitivity": 0.7328360987022614,
414
+ "eval_macro_specificity": 0.9845211676049321,
415
+ "eval_precision": 0.795801987953062,
416
+ "eval_precision_macro": 0.763738385548772,
417
+ "eval_recall": 0.7970565453137103,
418
+ "eval_recall_macro": 0.7328360987022614,
419
+ "eval_runtime": 90.1914,
420
+ "eval_samples_per_second": 14.314,
421
+ "eval_steps_per_second": 1.796,
422
+ "eval_weighted_fpr": 0.017862012544314153,
423
+ "eval_weighted_sensitivity": 0.7970565453137103,
424
+ "eval_weighted_specificity": 0.9707609687602743,
425
+ "step": 9002
426
+ },
427
+ {
428
+ "epoch": 14.77,
429
+ "learning_rate": 2.5391394504924833e-05,
430
+ "loss": 0.0375,
431
+ "step": 9500
432
+ },
433
+ {
434
+ "epoch": 15.0,
435
+ "eval_accuracy": 0.8001549186676995,
436
+ "eval_f1_macro": 0.7437118125777374,
437
+ "eval_f1_micro": 0.8001549186676995,
438
+ "eval_f1_weighted": 0.8017305518379161,
439
+ "eval_loss": 1.7058144807815552,
440
+ "eval_macro_fpr": 0.018188016275518495,
441
+ "eval_macro_sensitivity": 0.7696355635707867,
442
+ "eval_macro_specificity": 0.9850538317438372,
443
+ "eval_precision": 0.8109768674187026,
444
+ "eval_precision_macro": 0.7369403978788467,
445
+ "eval_recall": 0.8001549186676995,
446
+ "eval_recall_macro": 0.7696355635707867,
447
+ "eval_runtime": 89.6171,
448
+ "eval_samples_per_second": 14.406,
449
+ "eval_steps_per_second": 1.808,
450
+ "eval_weighted_fpr": 0.017527173913043478,
451
+ "eval_weighted_sensitivity": 0.8001549186676995,
452
+ "eval_weighted_specificity": 0.9756525574898595,
453
+ "step": 9645
454
+ },
455
+ {
456
+ "epoch": 15.55,
457
+ "learning_rate": 2.4095386210471748e-05,
458
+ "loss": 0.0241,
459
+ "step": 10000
460
+ },
461
+ {
462
+ "epoch": 16.0,
463
+ "eval_accuracy": 0.8179705654531371,
464
+ "eval_f1_macro": 0.7780054113531673,
465
+ "eval_f1_micro": 0.817970565453137,
466
+ "eval_f1_weighted": 0.8174074193030834,
467
+ "eval_loss": 1.7145990133285522,
468
+ "eval_macro_fpr": 0.01624192113353871,
469
+ "eval_macro_sensitivity": 0.7786912615456476,
470
+ "eval_macro_specificity": 0.9862689663514356,
471
+ "eval_precision": 0.8189471437742178,
472
+ "eval_precision_macro": 0.7852139494262297,
473
+ "eval_recall": 0.8179705654531371,
474
+ "eval_recall_macro": 0.7786912615456476,
475
+ "eval_runtime": 95.9969,
476
+ "eval_samples_per_second": 13.448,
477
+ "eval_steps_per_second": 1.688,
478
+ "eval_weighted_fpr": 0.01564684732671949,
479
+ "eval_weighted_sensitivity": 0.8179705654531371,
480
+ "eval_weighted_specificity": 0.9760639298183975,
481
+ "step": 10288
482
+ },
483
+ {
484
+ "epoch": 16.33,
485
+ "learning_rate": 2.280196993260757e-05,
486
+ "loss": 0.0226,
487
+ "step": 10500
488
+ },
489
+ {
490
+ "epoch": 17.0,
491
+ "eval_accuracy": 0.820294345468629,
492
+ "eval_f1_macro": 0.7713653883357572,
493
+ "eval_f1_micro": 0.8202943454686291,
494
+ "eval_f1_weighted": 0.8205584776004088,
495
+ "eval_loss": 1.703484296798706,
496
+ "eval_macro_fpr": 0.015955392684873622,
497
+ "eval_macro_sensitivity": 0.7780791588964692,
498
+ "eval_macro_specificity": 0.9865149979627027,
499
+ "eval_precision": 0.8238197570339643,
500
+ "eval_precision_macro": 0.7731970694747214,
501
+ "eval_recall": 0.820294345468629,
502
+ "eval_recall_macro": 0.7780791588964692,
503
+ "eval_runtime": 92.4633,
504
+ "eval_samples_per_second": 13.962,
505
+ "eval_steps_per_second": 1.752,
506
+ "eval_weighted_fpr": 0.015407092575375215,
507
+ "eval_weighted_sensitivity": 0.820294345468629,
508
+ "eval_weighted_specificity": 0.9774306239719132,
509
+ "step": 10931
510
+ },
511
+ {
512
+ "epoch": 17.11,
513
+ "learning_rate": 2.1505961638154484e-05,
514
+ "loss": 0.017,
515
+ "step": 11000
516
+ },
517
+ {
518
+ "epoch": 17.88,
519
+ "learning_rate": 2.0209953343701402e-05,
520
+ "loss": 0.0189,
521
+ "step": 11500
522
+ },
523
+ {
524
+ "epoch": 18.0,
525
+ "eval_accuracy": 0.8164213787761425,
526
+ "eval_f1_macro": 0.7577737009484876,
527
+ "eval_f1_micro": 0.8164213787761425,
528
+ "eval_f1_weighted": 0.814862576730404,
529
+ "eval_loss": 1.8079380989074707,
530
+ "eval_macro_fpr": 0.016561980735895838,
531
+ "eval_macro_sensitivity": 0.7677111310031827,
532
+ "eval_macro_specificity": 0.9860906809860878,
533
+ "eval_precision": 0.8160431006408752,
534
+ "eval_precision_macro": 0.7583264160007077,
535
+ "eval_recall": 0.8164213787761425,
536
+ "eval_recall_macro": 0.7677111310031827,
537
+ "eval_runtime": 93.2101,
538
+ "eval_samples_per_second": 13.85,
539
+ "eval_steps_per_second": 1.738,
540
+ "eval_weighted_fpr": 0.01580737677582872,
541
+ "eval_weighted_sensitivity": 0.8164213787761425,
542
+ "eval_weighted_specificity": 0.9749388360151747,
543
+ "step": 11574
544
+ },
545
+ {
546
+ "epoch": 18.66,
547
+ "learning_rate": 1.8913945049248317e-05,
548
+ "loss": 0.026,
549
+ "step": 12000
550
+ },
551
+ {
552
+ "epoch": 19.0,
553
+ "eval_accuracy": 0.8125484120836561,
554
+ "eval_f1_macro": 0.7718943136435594,
555
+ "eval_f1_micro": 0.8125484120836561,
556
+ "eval_f1_weighted": 0.8138445092504973,
557
+ "eval_loss": 1.8187031745910645,
558
+ "eval_macro_fpr": 0.016881633170148187,
559
+ "eval_macro_sensitivity": 0.7832768313344278,
560
+ "eval_macro_specificity": 0.9858228106032249,
561
+ "eval_precision": 0.8169875491486152,
562
+ "eval_precision_macro": 0.7674779420581895,
563
+ "eval_recall": 0.8125484120836561,
564
+ "eval_recall_macro": 0.7832768313344278,
565
+ "eval_runtime": 90.0764,
566
+ "eval_samples_per_second": 14.332,
567
+ "eval_steps_per_second": 1.798,
568
+ "eval_weighted_fpr": 0.016211146838156484,
569
+ "eval_weighted_sensitivity": 0.8125484120836561,
570
+ "eval_weighted_specificity": 0.9747937469647154,
571
+ "step": 12217
572
+ },
573
+ {
574
+ "epoch": 19.44,
575
+ "learning_rate": 1.761793675479523e-05,
576
+ "loss": 0.0101,
577
+ "step": 12500
578
+ },
579
+ {
580
+ "epoch": 20.0,
581
+ "eval_accuracy": 0.8187451587916343,
582
+ "eval_f1_macro": 0.7709574718137391,
583
+ "eval_f1_micro": 0.8187451587916342,
584
+ "eval_f1_weighted": 0.8179619842347855,
585
+ "eval_loss": 1.8353675603866577,
586
+ "eval_macro_fpr": 0.01611361621724302,
587
+ "eval_macro_sensitivity": 0.7817701569841005,
588
+ "eval_macro_specificity": 0.9863995310478259,
589
+ "eval_precision": 0.8219549803396689,
590
+ "eval_precision_macro": 0.7748222861859084,
591
+ "eval_recall": 0.8187451587916343,
592
+ "eval_recall_macro": 0.7817701569841005,
593
+ "eval_runtime": 86.3867,
594
+ "eval_samples_per_second": 14.944,
595
+ "eval_steps_per_second": 1.875,
596
+ "eval_weighted_fpr": 0.015566790846194785,
597
+ "eval_weighted_sensitivity": 0.8187451587916343,
598
+ "eval_weighted_specificity": 0.9772478069257525,
599
+ "step": 12860
600
+ },
601
+ {
602
+ "epoch": 20.22,
603
+ "learning_rate": 1.632192846034215e-05,
604
+ "loss": 0.0097,
605
+ "step": 13000
606
+ },
607
+ {
608
+ "epoch": 21.0,
609
+ "learning_rate": 1.502851218247797e-05,
610
+ "loss": 0.0216,
611
+ "step": 13500
612
+ },
613
+ {
614
+ "epoch": 21.0,
615
+ "eval_accuracy": 0.8156467854376452,
616
+ "eval_f1_macro": 0.7618079234508338,
617
+ "eval_f1_micro": 0.8156467854376452,
618
+ "eval_f1_weighted": 0.8164347717885334,
619
+ "eval_loss": 1.8372349739074707,
620
+ "eval_macro_fpr": 0.016334927583449652,
621
+ "eval_macro_sensitivity": 0.7857843546777831,
622
+ "eval_macro_specificity": 0.9863030854921722,
623
+ "eval_precision": 0.8219293193926849,
624
+ "eval_precision_macro": 0.7502422859641604,
625
+ "eval_recall": 0.8156467854376452,
626
+ "eval_recall_macro": 0.7857843546777831,
627
+ "eval_runtime": 87.8538,
628
+ "eval_samples_per_second": 14.695,
629
+ "eval_steps_per_second": 1.844,
630
+ "eval_weighted_fpr": 0.01588785046728972,
631
+ "eval_weighted_sensitivity": 0.8156467854376452,
632
+ "eval_weighted_specificity": 0.9788994969449356,
633
+ "step": 13503
634
+ },
635
+ {
636
+ "epoch": 21.77,
637
+ "learning_rate": 1.3732503888024884e-05,
638
+ "loss": 0.0138,
639
+ "step": 14000
640
+ },
641
+ {
642
+ "epoch": 22.0,
643
+ "eval_accuracy": 0.820294345468629,
644
+ "eval_f1_macro": 0.7661945549993608,
645
+ "eval_f1_micro": 0.8202943454686291,
646
+ "eval_f1_weighted": 0.8222065987286614,
647
+ "eval_loss": 1.8472038507461548,
648
+ "eval_macro_fpr": 0.015872843311496198,
649
+ "eval_macro_sensitivity": 0.7796378584356709,
650
+ "eval_macro_specificity": 0.9865941123818214,
651
+ "eval_precision": 0.8262928188362502,
652
+ "eval_precision_macro": 0.7613141480374475,
653
+ "eval_recall": 0.820294345468629,
654
+ "eval_recall_macro": 0.7796378584356709,
655
+ "eval_runtime": 85.3935,
656
+ "eval_samples_per_second": 15.118,
657
+ "eval_steps_per_second": 1.897,
658
+ "eval_weighted_fpr": 0.015407092575375215,
659
+ "eval_weighted_sensitivity": 0.820294345468629,
660
+ "eval_weighted_specificity": 0.9786173402586936,
661
+ "step": 14146
662
+ },
663
+ {
664
+ "epoch": 22.55,
665
+ "learning_rate": 1.2436495593571799e-05,
666
+ "loss": 0.0169,
667
+ "step": 14500
668
+ },
669
+ {
670
+ "epoch": 23.0,
671
+ "eval_accuracy": 0.8218435321456236,
672
+ "eval_f1_macro": 0.760658140409385,
673
+ "eval_f1_micro": 0.8218435321456236,
674
+ "eval_f1_weighted": 0.8209229699908378,
675
+ "eval_loss": 1.8103588819503784,
676
+ "eval_macro_fpr": 0.016045052484127836,
677
+ "eval_macro_sensitivity": 0.7595040114200765,
678
+ "eval_macro_specificity": 0.986451381533847,
679
+ "eval_precision": 0.8251783607661838,
680
+ "eval_precision_macro": 0.7718974635406236,
681
+ "eval_recall": 0.8218435321456236,
682
+ "eval_recall_macro": 0.7595040114200765,
683
+ "eval_runtime": 92.1092,
684
+ "eval_samples_per_second": 14.016,
685
+ "eval_steps_per_second": 1.759,
686
+ "eval_weighted_fpr": 0.015247944842216919,
687
+ "eval_weighted_sensitivity": 0.8218435321456236,
688
+ "eval_weighted_specificity": 0.9749271908620799,
689
+ "step": 14789
690
+ },
691
+ {
692
+ "epoch": 23.33,
693
+ "learning_rate": 1.1140487299118715e-05,
694
+ "loss": 0.0079,
695
+ "step": 15000
696
+ },
697
+ {
698
+ "epoch": 24.0,
699
+ "eval_accuracy": 0.8109992254066615,
700
+ "eval_f1_macro": 0.7524207866195778,
701
+ "eval_f1_micro": 0.8109992254066615,
702
+ "eval_f1_weighted": 0.8122562112865754,
703
+ "eval_loss": 1.9252820014953613,
704
+ "eval_macro_fpr": 0.017055628961045247,
705
+ "eval_macro_sensitivity": 0.7576415197380852,
706
+ "eval_macro_specificity": 0.985793052330349,
707
+ "eval_precision": 0.8202456922159745,
708
+ "eval_precision_macro": 0.7621718512762216,
709
+ "eval_recall": 0.8109992254066615,
710
+ "eval_recall_macro": 0.7576415197380852,
711
+ "eval_runtime": 93.885,
712
+ "eval_samples_per_second": 13.751,
713
+ "eval_steps_per_second": 1.726,
714
+ "eval_weighted_fpr": 0.016373641121997046,
715
+ "eval_weighted_sensitivity": 0.8109992254066615,
716
+ "eval_weighted_specificity": 0.9758965595485739,
717
+ "step": 15432
718
+ },
719
+ {
720
+ "epoch": 24.11,
721
+ "learning_rate": 9.84447900466563e-06,
722
+ "loss": 0.0094,
723
+ "step": 15500
724
+ },
725
+ {
726
+ "epoch": 24.88,
727
+ "learning_rate": 8.551062726801453e-06,
728
+ "loss": 0.0017,
729
+ "step": 16000
730
+ },
731
+ {
732
+ "epoch": 25.0,
733
+ "eval_accuracy": 0.8156467854376452,
734
+ "eval_f1_macro": 0.7594027956673404,
735
+ "eval_f1_micro": 0.8156467854376452,
736
+ "eval_f1_weighted": 0.8154952009918864,
737
+ "eval_loss": 1.9110798835754395,
738
+ "eval_macro_fpr": 0.016444882591068778,
739
+ "eval_macro_sensitivity": 0.7742299386060195,
740
+ "eval_macro_specificity": 0.9862078449547576,
741
+ "eval_precision": 0.8193166881646313,
742
+ "eval_precision_macro": 0.7554028042588868,
743
+ "eval_recall": 0.8156467854376452,
744
+ "eval_recall_macro": 0.7742299386060195,
745
+ "eval_runtime": 92.0946,
746
+ "eval_samples_per_second": 14.018,
747
+ "eval_steps_per_second": 1.759,
748
+ "eval_weighted_fpr": 0.01588785046728972,
749
+ "eval_weighted_sensitivity": 0.8156467854376452,
750
+ "eval_weighted_specificity": 0.97747088888372,
751
+ "step": 16075
752
+ },
753
+ {
754
+ "epoch": 25.66,
755
+ "learning_rate": 7.255054432348367e-06,
756
+ "loss": 0.0071,
757
+ "step": 16500
758
+ },
759
+ {
760
+ "epoch": 26.0,
761
+ "eval_accuracy": 0.8187451587916343,
762
+ "eval_f1_macro": 0.7601055992728972,
763
+ "eval_f1_micro": 0.8187451587916342,
764
+ "eval_f1_weighted": 0.8199059390372757,
765
+ "eval_loss": 1.880871057510376,
766
+ "eval_macro_fpr": 0.01605804835750131,
767
+ "eval_macro_sensitivity": 0.7748936539821518,
768
+ "eval_macro_specificity": 0.9864696047286632,
769
+ "eval_precision": 0.824436161286693,
770
+ "eval_precision_macro": 0.7594879689938191,
771
+ "eval_recall": 0.8187451587916343,
772
+ "eval_recall_macro": 0.7748936539821518,
773
+ "eval_runtime": 88.2427,
774
+ "eval_samples_per_second": 14.63,
775
+ "eval_steps_per_second": 1.836,
776
+ "eval_weighted_fpr": 0.015566790846194785,
777
+ "eval_weighted_sensitivity": 0.8187451587916343,
778
+ "eval_weighted_specificity": 0.9782989121383131,
779
+ "step": 16718
780
+ },
781
+ {
782
+ "epoch": 26.44,
783
+ "learning_rate": 5.9590461378952825e-06,
784
+ "loss": 0.0032,
785
+ "step": 17000
786
+ },
787
+ {
788
+ "epoch": 27.0,
789
+ "eval_accuracy": 0.8272656855151046,
790
+ "eval_f1_macro": 0.7718154001656008,
791
+ "eval_f1_micro": 0.8272656855151046,
792
+ "eval_f1_weighted": 0.8288829948767273,
793
+ "eval_loss": 1.8246089220046997,
794
+ "eval_macro_fpr": 0.015208141271413749,
795
+ "eval_macro_sensitivity": 0.7806866379450543,
796
+ "eval_macro_specificity": 0.9870589996943918,
797
+ "eval_precision": 0.8333256234957093,
798
+ "eval_precision_macro": 0.772651547029787,
799
+ "eval_recall": 0.8272656855151046,
800
+ "eval_recall_macro": 0.7806866379450543,
801
+ "eval_runtime": 86.7033,
802
+ "eval_samples_per_second": 14.89,
803
+ "eval_steps_per_second": 1.868,
804
+ "eval_weighted_fpr": 0.014695222405271829,
805
+ "eval_weighted_sensitivity": 0.8272656855151046,
806
+ "eval_weighted_specificity": 0.9786193099007718,
807
+ "step": 17361
808
+ },
809
+ {
810
+ "epoch": 27.22,
811
+ "learning_rate": 4.663037843442198e-06,
812
+ "loss": 0.006,
813
+ "step": 17500
814
+ },
815
+ {
816
+ "epoch": 27.99,
817
+ "learning_rate": 3.3696215655780196e-06,
818
+ "loss": 0.0014,
819
+ "step": 18000
820
+ },
821
+ {
822
+ "epoch": 28.0,
823
+ "eval_accuracy": 0.8264910921766073,
824
+ "eval_f1_macro": 0.764831239702779,
825
+ "eval_f1_micro": 0.8264910921766073,
826
+ "eval_f1_weighted": 0.8282063495789155,
827
+ "eval_loss": 1.8353731632232666,
828
+ "eval_macro_fpr": 0.015371416416681213,
829
+ "eval_macro_sensitivity": 0.7806016856623184,
830
+ "eval_macro_specificity": 0.9869917438299329,
831
+ "eval_precision": 0.8337362569761955,
832
+ "eval_precision_macro": 0.7623740799673432,
833
+ "eval_recall": 0.8264910921766073,
834
+ "eval_recall_macro": 0.7806016856623184,
835
+ "eval_runtime": 89.3121,
836
+ "eval_samples_per_second": 14.455,
837
+ "eval_steps_per_second": 1.814,
838
+ "eval_weighted_fpr": 0.014773776546629732,
839
+ "eval_weighted_sensitivity": 0.8264910921766073,
840
+ "eval_weighted_specificity": 0.9783850652723848,
841
+ "step": 18004
842
+ },
843
+ {
844
+ "epoch": 28.77,
845
+ "learning_rate": 2.073613271124935e-06,
846
+ "loss": 0.0004,
847
+ "step": 18500
848
+ },
849
+ {
850
+ "epoch": 29.0,
851
+ "eval_accuracy": 0.8233927188226181,
852
+ "eval_f1_macro": 0.7646400059289129,
853
+ "eval_f1_micro": 0.8233927188226181,
854
+ "eval_f1_weighted": 0.8233976779729975,
855
+ "eval_loss": 1.8558473587036133,
856
+ "eval_macro_fpr": 0.01570985992861236,
857
+ "eval_macro_sensitivity": 0.7801423076748631,
858
+ "eval_macro_specificity": 0.9867461963920389,
859
+ "eval_precision": 0.8277172724014995,
860
+ "eval_precision_macro": 0.7615653842067026,
861
+ "eval_recall": 0.8233927188226181,
862
+ "eval_recall_macro": 0.7801423076748631,
863
+ "eval_runtime": 93.8609,
864
+ "eval_samples_per_second": 13.754,
865
+ "eval_steps_per_second": 1.726,
866
+ "eval_weighted_fpr": 0.015089344804765056,
867
+ "eval_weighted_sensitivity": 0.8233927188226181,
868
+ "eval_weighted_specificity": 0.9778002270579647,
869
+ "step": 18647
870
+ },
871
+ {
872
+ "epoch": 29.55,
873
+ "learning_rate": 7.776049766718508e-07,
874
+ "loss": 0.0012,
875
+ "step": 19000
876
+ },
877
+ {
878
+ "epoch": 30.0,
879
+ "eval_accuracy": 0.8233927188226181,
880
+ "eval_f1_macro": 0.7641703955582007,
881
+ "eval_f1_micro": 0.8233927188226181,
882
+ "eval_f1_weighted": 0.8237235126724278,
883
+ "eval_loss": 1.8391920328140259,
884
+ "eval_macro_fpr": 0.015672757930611226,
885
+ "eval_macro_sensitivity": 0.7799056980188922,
886
+ "eval_macro_specificity": 0.986769188821419,
887
+ "eval_precision": 0.8281388668138719,
888
+ "eval_precision_macro": 0.7608494934948501,
889
+ "eval_recall": 0.8233927188226181,
890
+ "eval_recall_macro": 0.7799056980188922,
891
+ "eval_runtime": 95.7111,
892
+ "eval_samples_per_second": 13.489,
893
+ "eval_steps_per_second": 1.693,
894
+ "eval_weighted_fpr": 0.015089344804765056,
895
+ "eval_weighted_sensitivity": 0.8233927188226181,
896
+ "eval_weighted_specificity": 0.9781451134986645,
897
+ "step": 19290
898
+ }
899
+ ],
900
+ "logging_steps": 500,
901
+ "max_steps": 19290,
902
+ "num_train_epochs": 30,
903
+ "save_steps": 500,
904
+ "total_flos": 4.727750685751296e+16,
905
+ "trial_name": null,
906
+ "trial_params": null
907
+ }
training_checkpoints/checkpoint-19290/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cd0cdec3791602f3a208801abdf1b96f9559896ff555c412e7903eee9d6a2b88
3
+ size 4600
training_checkpoints/checkpoint-19290/vocab.json ADDED
The diff for this file is too large to render. See raw diff