Upload folder using huggingface_hub
Browse files- training_checkpoints/checkpoint-4501/config.json +63 -0
- training_checkpoints/checkpoint-4501/model.safetensors +3 -0
- training_checkpoints/checkpoint-4501/optimizer.pt +3 -0
- training_checkpoints/checkpoint-4501/rng_state.pth +3 -0
- training_checkpoints/checkpoint-4501/scheduler.pt +3 -0
- training_checkpoints/checkpoint-4501/trainer_state.json +238 -0
- training_checkpoints/checkpoint-4501/training_args.bin +3 -0
training_checkpoints/checkpoint-4501/config.json
ADDED
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "nlpaueb/legal-bert-base-uncased",
|
3 |
+
"architectures": [
|
4 |
+
"BertForSequenceClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"bos_token_id": 0,
|
8 |
+
"classifier_dropout": null,
|
9 |
+
"eos_token_ids": 0,
|
10 |
+
"hidden_act": "gelu",
|
11 |
+
"hidden_dropout_prob": 0.1,
|
12 |
+
"hidden_size": 768,
|
13 |
+
"id2label": {
|
14 |
+
"0": "Issue",
|
15 |
+
"1": "Court Discourse",
|
16 |
+
"2": "Conclusion",
|
17 |
+
"3": "Precedent Analysis",
|
18 |
+
"4": "Section Analysis",
|
19 |
+
"5": "Argument by Petitioner",
|
20 |
+
"6": "Fact",
|
21 |
+
"7": "Argument by Respondent",
|
22 |
+
"8": "Ratio",
|
23 |
+
"9": "Appellant",
|
24 |
+
"10": "Respondent",
|
25 |
+
"11": "Argument by Appellant",
|
26 |
+
"12": "Petitioner",
|
27 |
+
"13": "Judge",
|
28 |
+
"14": "Argument by Defendant"
|
29 |
+
},
|
30 |
+
"initializer_range": 0.02,
|
31 |
+
"intermediate_size": 3072,
|
32 |
+
"label2id": {
|
33 |
+
"Appellant": 9,
|
34 |
+
"Argument by Appellant": 11,
|
35 |
+
"Argument by Defendant": 14,
|
36 |
+
"Argument by Petitioner": 5,
|
37 |
+
"Argument by Respondent": 7,
|
38 |
+
"Conclusion": 2,
|
39 |
+
"Court Discourse": 1,
|
40 |
+
"Fact": 6,
|
41 |
+
"Issue": 0,
|
42 |
+
"Judge": 13,
|
43 |
+
"Petitioner": 12,
|
44 |
+
"Precedent Analysis": 3,
|
45 |
+
"Ratio": 8,
|
46 |
+
"Respondent": 10,
|
47 |
+
"Section Analysis": 4
|
48 |
+
},
|
49 |
+
"layer_norm_eps": 1e-12,
|
50 |
+
"max_position_embeddings": 512,
|
51 |
+
"model_type": "bert",
|
52 |
+
"num_attention_heads": 12,
|
53 |
+
"num_hidden_layers": 12,
|
54 |
+
"output_past": true,
|
55 |
+
"pad_token_id": 0,
|
56 |
+
"position_embedding_type": "absolute",
|
57 |
+
"problem_type": "single_label_classification",
|
58 |
+
"torch_dtype": "float32",
|
59 |
+
"transformers_version": "4.38.2",
|
60 |
+
"type_vocab_size": 2,
|
61 |
+
"use_cache": true,
|
62 |
+
"vocab_size": 30522
|
63 |
+
}
|
training_checkpoints/checkpoint-4501/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:765f4983f68dae271f03961bc29d1bab1cadf80e9b5fd716b11cda74d2ad9979
|
3 |
+
size 437998636
|
training_checkpoints/checkpoint-4501/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:af08777dde8c2b1dec0ca6c7a3d3664ee207fae2559e8d95f6d5e3b50966459d
|
3 |
+
size 876118266
|
training_checkpoints/checkpoint-4501/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1430a669b2089ff30c842627f0e95cbbaa558bb927d1920e488b2a0cf1a6b002
|
3 |
+
size 14244
|
training_checkpoints/checkpoint-4501/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9d3a1e78d3bd3700591cddf160cbdce35f5f037b933cdf3b831de5621311dab5
|
3 |
+
size 1064
|
training_checkpoints/checkpoint-4501/trainer_state.json
ADDED
@@ -0,0 +1,238 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.7839844036299938,
|
3 |
+
"best_model_checkpoint": "legal-bert-base-uncased/checkpoint-3858",
|
4 |
+
"epoch": 7.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 4501,
|
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 |
+
"grad_norm": 6.735908031463623,
|
14 |
+
"learning_rate": 4.742353551062727e-05,
|
15 |
+
"loss": 1.1096,
|
16 |
+
"step": 500
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 1.0,
|
20 |
+
"eval_accuracy": 0.7978311386522076,
|
21 |
+
"eval_f1_macro": 0.613418966848975,
|
22 |
+
"eval_f1_micro": 0.7978311386522076,
|
23 |
+
"eval_f1_weighted": 0.7839880886224406,
|
24 |
+
"eval_loss": 0.6747979521751404,
|
25 |
+
"eval_macro_fpr": 0.018830502815301575,
|
26 |
+
"eval_macro_sensitivity": 0.6339578261148517,
|
27 |
+
"eval_macro_specificity": 0.9845345883924012,
|
28 |
+
"eval_precision": 0.7854860044786733,
|
29 |
+
"eval_precision_macro": 0.6239440349375213,
|
30 |
+
"eval_recall": 0.7978311386522076,
|
31 |
+
"eval_recall_macro": 0.6339578261148517,
|
32 |
+
"eval_runtime": 30.9921,
|
33 |
+
"eval_samples_per_second": 41.656,
|
34 |
+
"eval_steps_per_second": 5.227,
|
35 |
+
"eval_weighted_fpr": 0.017778080512226686,
|
36 |
+
"eval_weighted_sensitivity": 0.7978311386522076,
|
37 |
+
"eval_weighted_specificity": 0.9701876872338092,
|
38 |
+
"step": 643
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 1.56,
|
42 |
+
"grad_norm": 6.824150562286377,
|
43 |
+
"learning_rate": 4.48315189217211e-05,
|
44 |
+
"loss": 0.6187,
|
45 |
+
"step": 1000
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 2.0,
|
49 |
+
"eval_accuracy": 0.8109992254066615,
|
50 |
+
"eval_f1_macro": 0.7268228216676939,
|
51 |
+
"eval_f1_micro": 0.8109992254066615,
|
52 |
+
"eval_f1_weighted": 0.8089875323563979,
|
53 |
+
"eval_loss": 0.6449354887008667,
|
54 |
+
"eval_macro_fpr": 0.016938802987367406,
|
55 |
+
"eval_macro_sensitivity": 0.7327307314578203,
|
56 |
+
"eval_macro_specificity": 0.9857686909763926,
|
57 |
+
"eval_precision": 0.8196084470842065,
|
58 |
+
"eval_precision_macro": 0.7805936449528967,
|
59 |
+
"eval_recall": 0.8109992254066615,
|
60 |
+
"eval_recall_macro": 0.7327307314578203,
|
61 |
+
"eval_runtime": 58.3071,
|
62 |
+
"eval_samples_per_second": 22.141,
|
63 |
+
"eval_steps_per_second": 2.778,
|
64 |
+
"eval_weighted_fpr": 0.016373641121997046,
|
65 |
+
"eval_weighted_sensitivity": 0.8109992254066615,
|
66 |
+
"eval_weighted_specificity": 0.9755311392392266,
|
67 |
+
"step": 1286
|
68 |
+
},
|
69 |
+
{
|
70 |
+
"epoch": 2.33,
|
71 |
+
"grad_norm": 20.75279998779297,
|
72 |
+
"learning_rate": 4.223950233281493e-05,
|
73 |
+
"loss": 0.4747,
|
74 |
+
"step": 1500
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 3.0,
|
78 |
+
"eval_accuracy": 0.814872192099148,
|
79 |
+
"eval_f1_macro": 0.7369957890904685,
|
80 |
+
"eval_f1_micro": 0.814872192099148,
|
81 |
+
"eval_f1_weighted": 0.8124596447054482,
|
82 |
+
"eval_loss": 0.8151384592056274,
|
83 |
+
"eval_macro_fpr": 0.016571707879599946,
|
84 |
+
"eval_macro_sensitivity": 0.739039840019417,
|
85 |
+
"eval_macro_specificity": 0.9860629698102752,
|
86 |
+
"eval_precision": 0.8192276160546627,
|
87 |
+
"eval_precision_macro": 0.7659183928188651,
|
88 |
+
"eval_recall": 0.814872192099148,
|
89 |
+
"eval_recall_macro": 0.739039840019417,
|
90 |
+
"eval_runtime": 55.9094,
|
91 |
+
"eval_samples_per_second": 23.091,
|
92 |
+
"eval_steps_per_second": 2.898,
|
93 |
+
"eval_weighted_fpr": 0.015968463954032203,
|
94 |
+
"eval_weighted_sensitivity": 0.814872192099148,
|
95 |
+
"eval_weighted_specificity": 0.9760723550549795,
|
96 |
+
"step": 1929
|
97 |
+
},
|
98 |
+
{
|
99 |
+
"epoch": 3.11,
|
100 |
+
"grad_norm": 10.737870216369629,
|
101 |
+
"learning_rate": 3.9652669777086575e-05,
|
102 |
+
"loss": 0.3866,
|
103 |
+
"step": 2000
|
104 |
+
},
|
105 |
+
{
|
106 |
+
"epoch": 3.89,
|
107 |
+
"grad_norm": 13.75886344909668,
|
108 |
+
"learning_rate": 3.7060653188180404e-05,
|
109 |
+
"loss": 0.2645,
|
110 |
+
"step": 2500
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 4.0,
|
114 |
+
"eval_accuracy": 0.8218435321456236,
|
115 |
+
"eval_f1_macro": 0.738478964900184,
|
116 |
+
"eval_f1_micro": 0.8218435321456236,
|
117 |
+
"eval_f1_weighted": 0.8189030195253059,
|
118 |
+
"eval_loss": 0.9344585537910461,
|
119 |
+
"eval_macro_fpr": 0.015763618193954002,
|
120 |
+
"eval_macro_sensitivity": 0.7412892486668318,
|
121 |
+
"eval_macro_specificity": 0.986618111597957,
|
122 |
+
"eval_precision": 0.8198244063808544,
|
123 |
+
"eval_precision_macro": 0.7446229748697999,
|
124 |
+
"eval_recall": 0.8218435321456236,
|
125 |
+
"eval_recall_macro": 0.7412892486668318,
|
126 |
+
"eval_runtime": 75.578,
|
127 |
+
"eval_samples_per_second": 17.082,
|
128 |
+
"eval_steps_per_second": 2.143,
|
129 |
+
"eval_weighted_fpr": 0.015247944842216919,
|
130 |
+
"eval_weighted_sensitivity": 0.8218435321456236,
|
131 |
+
"eval_weighted_specificity": 0.9774281418237326,
|
132 |
+
"step": 2572
|
133 |
+
},
|
134 |
+
{
|
135 |
+
"epoch": 4.67,
|
136 |
+
"grad_norm": 8.32991886138916,
|
137 |
+
"learning_rate": 3.4468636599274234e-05,
|
138 |
+
"loss": 0.1901,
|
139 |
+
"step": 3000
|
140 |
+
},
|
141 |
+
{
|
142 |
+
"epoch": 5.0,
|
143 |
+
"eval_accuracy": 0.8195197521301317,
|
144 |
+
"eval_f1_macro": 0.7594855831156184,
|
145 |
+
"eval_f1_micro": 0.8195197521301317,
|
146 |
+
"eval_f1_weighted": 0.81658937168797,
|
147 |
+
"eval_loss": 1.0929232835769653,
|
148 |
+
"eval_macro_fpr": 0.01614879224826684,
|
149 |
+
"eval_macro_sensitivity": 0.743232314276478,
|
150 |
+
"eval_macro_specificity": 0.9863035723232563,
|
151 |
+
"eval_precision": 0.8241924882578779,
|
152 |
+
"eval_precision_macro": 0.8264084869739744,
|
153 |
+
"eval_recall": 0.8195197521301317,
|
154 |
+
"eval_recall_macro": 0.743232314276478,
|
155 |
+
"eval_runtime": 65.7326,
|
156 |
+
"eval_samples_per_second": 19.64,
|
157 |
+
"eval_steps_per_second": 2.465,
|
158 |
+
"eval_weighted_fpr": 0.01548687271518777,
|
159 |
+
"eval_weighted_sensitivity": 0.8195197521301317,
|
160 |
+
"eval_weighted_specificity": 0.9750338327187125,
|
161 |
+
"step": 3215
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 5.44,
|
165 |
+
"grad_norm": 50.74209976196289,
|
166 |
+
"learning_rate": 3.187662001036807e-05,
|
167 |
+
"loss": 0.1131,
|
168 |
+
"step": 3500
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 6.0,
|
172 |
+
"eval_accuracy": 0.820294345468629,
|
173 |
+
"eval_f1_macro": 0.7839844036299938,
|
174 |
+
"eval_f1_micro": 0.8202943454686291,
|
175 |
+
"eval_f1_weighted": 0.8197212343639901,
|
176 |
+
"eval_loss": 1.1536158323287964,
|
177 |
+
"eval_macro_fpr": 0.01593810671400651,
|
178 |
+
"eval_macro_sensitivity": 0.7786487516932519,
|
179 |
+
"eval_macro_specificity": 0.9864617261648754,
|
180 |
+
"eval_precision": 0.821246410485655,
|
181 |
+
"eval_precision_macro": 0.7968460348821496,
|
182 |
+
"eval_recall": 0.820294345468629,
|
183 |
+
"eval_recall_macro": 0.7786487516932519,
|
184 |
+
"eval_runtime": 54.9389,
|
185 |
+
"eval_samples_per_second": 23.499,
|
186 |
+
"eval_steps_per_second": 2.949,
|
187 |
+
"eval_weighted_fpr": 0.015407092575375215,
|
188 |
+
"eval_weighted_sensitivity": 0.820294345468629,
|
189 |
+
"eval_weighted_specificity": 0.9766315470045015,
|
190 |
+
"step": 3858
|
191 |
+
},
|
192 |
+
{
|
193 |
+
"epoch": 6.22,
|
194 |
+
"grad_norm": 0.02055634744465351,
|
195 |
+
"learning_rate": 2.928978745463971e-05,
|
196 |
+
"loss": 0.0975,
|
197 |
+
"step": 4000
|
198 |
+
},
|
199 |
+
{
|
200 |
+
"epoch": 7.0,
|
201 |
+
"grad_norm": 32.151573181152344,
|
202 |
+
"learning_rate": 2.669777086573354e-05,
|
203 |
+
"loss": 0.063,
|
204 |
+
"step": 4500
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 7.0,
|
208 |
+
"eval_accuracy": 0.8117738187451587,
|
209 |
+
"eval_f1_macro": 0.7495479890901372,
|
210 |
+
"eval_f1_micro": 0.8117738187451587,
|
211 |
+
"eval_f1_weighted": 0.8135965211162124,
|
212 |
+
"eval_loss": 1.3217942714691162,
|
213 |
+
"eval_macro_fpr": 0.01664162427116977,
|
214 |
+
"eval_macro_sensitivity": 0.7526435450791772,
|
215 |
+
"eval_macro_specificity": 0.9859379584321952,
|
216 |
+
"eval_precision": 0.818420676521013,
|
217 |
+
"eval_precision_macro": 0.7517839435012339,
|
218 |
+
"eval_recall": 0.8117738187451587,
|
219 |
+
"eval_recall_macro": 0.7526435450791772,
|
220 |
+
"eval_runtime": 59.9636,
|
221 |
+
"eval_samples_per_second": 21.53,
|
222 |
+
"eval_steps_per_second": 2.702,
|
223 |
+
"eval_weighted_fpr": 0.016292323164599398,
|
224 |
+
"eval_weighted_sensitivity": 0.8117738187451587,
|
225 |
+
"eval_weighted_specificity": 0.9772955577377688,
|
226 |
+
"step": 4501
|
227 |
+
}
|
228 |
+
],
|
229 |
+
"logging_steps": 500,
|
230 |
+
"max_steps": 9645,
|
231 |
+
"num_input_tokens_seen": 0,
|
232 |
+
"num_train_epochs": 15,
|
233 |
+
"save_steps": 500,
|
234 |
+
"total_flos": 9464156761135104.0,
|
235 |
+
"train_batch_size": 8,
|
236 |
+
"trial_name": null,
|
237 |
+
"trial_params": null
|
238 |
+
}
|
training_checkpoints/checkpoint-4501/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:21938ff48cdf92422f693d69544854f8291bee6a4c7091c49d48bd0e59616530
|
3 |
+
size 4920
|