josecannete
commited on
Commit
•
60c5917
1
Parent(s):
c0d12cb
adding model finetuned on MLDoc
Browse files- all_results.json +14 -0
- config.json +46 -0
- eval_results.json +9 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- train_results.json +8 -0
- trainer_state.json +256 -0
- training_args.bin +3 -0
all_results.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"eval_accuracy": 0.9639999866485596,
|
4 |
+
"eval_loss": 0.15715740621089935,
|
5 |
+
"eval_runtime": 4.3422,
|
6 |
+
"eval_samples": 1000,
|
7 |
+
"eval_samples_per_second": 230.296,
|
8 |
+
"eval_steps_per_second": 14.509,
|
9 |
+
"train_loss": 0.07163469533662538,
|
10 |
+
"train_runtime": 1355.547,
|
11 |
+
"train_samples": 9458,
|
12 |
+
"train_samples_per_second": 27.909,
|
13 |
+
"train_steps_per_second": 1.747
|
14 |
+
}
|
config.json
ADDED
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "CenIA/albert_base_spanish",
|
3 |
+
"architectures": [
|
4 |
+
"AlbertForSequenceClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0,
|
7 |
+
"bos_token_id": 2,
|
8 |
+
"classifier_dropout_prob": 0.1,
|
9 |
+
"down_scale_factor": 1,
|
10 |
+
"embedding_size": 128,
|
11 |
+
"eos_token_id": 3,
|
12 |
+
"gap_size": 0,
|
13 |
+
"hidden_act": "gelu",
|
14 |
+
"hidden_dropout_prob": 0,
|
15 |
+
"hidden_size": 768,
|
16 |
+
"id2label": {
|
17 |
+
"0": "LABEL_0",
|
18 |
+
"1": "LABEL_1",
|
19 |
+
"2": "LABEL_2",
|
20 |
+
"3": "LABEL_3"
|
21 |
+
},
|
22 |
+
"initializer_range": 0.02,
|
23 |
+
"inner_group_num": 1,
|
24 |
+
"intermediate_size": 3072,
|
25 |
+
"label2id": {
|
26 |
+
"LABEL_0": 0,
|
27 |
+
"LABEL_1": 1,
|
28 |
+
"LABEL_2": 2,
|
29 |
+
"LABEL_3": 3
|
30 |
+
},
|
31 |
+
"layer_norm_eps": 1e-12,
|
32 |
+
"max_position_embeddings": 512,
|
33 |
+
"model_type": "albert",
|
34 |
+
"net_structure_type": 0,
|
35 |
+
"num_attention_heads": 12,
|
36 |
+
"num_hidden_groups": 1,
|
37 |
+
"num_hidden_layers": 12,
|
38 |
+
"num_memory_blocks": 0,
|
39 |
+
"pad_token_id": 0,
|
40 |
+
"position_embedding_type": "absolute",
|
41 |
+
"problem_type": "single_label_classification",
|
42 |
+
"torch_dtype": "float32",
|
43 |
+
"transformers_version": "4.12.5",
|
44 |
+
"type_vocab_size": 2,
|
45 |
+
"vocab_size": 31000
|
46 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"eval_accuracy": 0.9639999866485596,
|
4 |
+
"eval_loss": 0.15715740621089935,
|
5 |
+
"eval_runtime": 4.3422,
|
6 |
+
"eval_samples": 1000,
|
7 |
+
"eval_samples_per_second": 230.296,
|
8 |
+
"eval_steps_per_second": 14.509
|
9 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:812cb1e206989fb55fd538abe04b066ffe408ae26ac87a862eec74bebf8e218a
|
3 |
+
size 47273681
|
special_tokens_map.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"bos_token": "[CLS]", "eos_token": "[SEP]", "unk_token": "<unk>", "sep_token": "[SEP]", "pad_token": "<pad>", "cls_token": "[CLS]", "mask_token": "[MASK]"}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"do_lower_case": true, "remove_space": true, "keep_accents": true, "bos_token": "[CLS]", "eos_token": "[SEP]", "unk_token": "<unk>", "sep_token": "[SEP]", "pad_token": "<pad>", "cls_token": "[CLS]", "mask_token": {"content": "[MASK]", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "CenIA/albert_base_spanish", "tokenizer_class": "AlbertTokenizer"}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 4.0,
|
3 |
+
"train_loss": 0.07163469533662538,
|
4 |
+
"train_runtime": 1355.547,
|
5 |
+
"train_samples": 9458,
|
6 |
+
"train_samples_per_second": 27.909,
|
7 |
+
"train_steps_per_second": 1.747
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,256 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.15715740621089935,
|
3 |
+
"best_model_checkpoint": "/data/jcanete/all_results/mldoc/albeto_base/epochs_4_bs_16_lr_5e-5/checkpoint-1400",
|
4 |
+
"epoch": 4.0,
|
5 |
+
"global_step": 2368,
|
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 |
+
"eval_accuracy": 0.7229999899864197,
|
13 |
+
"eval_loss": 1.0613000392913818,
|
14 |
+
"eval_runtime": 4.4363,
|
15 |
+
"eval_samples_per_second": 225.412,
|
16 |
+
"eval_steps_per_second": 14.201,
|
17 |
+
"step": 100
|
18 |
+
},
|
19 |
+
{
|
20 |
+
"epoch": 0.34,
|
21 |
+
"eval_accuracy": 0.7990000247955322,
|
22 |
+
"eval_loss": 0.672461748123169,
|
23 |
+
"eval_runtime": 4.3549,
|
24 |
+
"eval_samples_per_second": 229.626,
|
25 |
+
"eval_steps_per_second": 14.466,
|
26 |
+
"step": 200
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.51,
|
30 |
+
"eval_accuracy": 0.8479999899864197,
|
31 |
+
"eval_loss": 0.523521900177002,
|
32 |
+
"eval_runtime": 4.3517,
|
33 |
+
"eval_samples_per_second": 229.796,
|
34 |
+
"eval_steps_per_second": 14.477,
|
35 |
+
"step": 300
|
36 |
+
},
|
37 |
+
{
|
38 |
+
"epoch": 0.68,
|
39 |
+
"eval_accuracy": 0.8930000066757202,
|
40 |
+
"eval_loss": 0.4281160831451416,
|
41 |
+
"eval_runtime": 4.3489,
|
42 |
+
"eval_samples_per_second": 229.946,
|
43 |
+
"eval_steps_per_second": 14.487,
|
44 |
+
"step": 400
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.84,
|
48 |
+
"learning_rate": 3.94847972972973e-05,
|
49 |
+
"loss": 0.1597,
|
50 |
+
"step": 500
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.84,
|
54 |
+
"eval_accuracy": 0.9039999842643738,
|
55 |
+
"eval_loss": 0.4125765860080719,
|
56 |
+
"eval_runtime": 4.3451,
|
57 |
+
"eval_samples_per_second": 230.146,
|
58 |
+
"eval_steps_per_second": 14.499,
|
59 |
+
"step": 500
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 1.01,
|
63 |
+
"eval_accuracy": 0.9020000100135803,
|
64 |
+
"eval_loss": 0.40425220131874084,
|
65 |
+
"eval_runtime": 4.3435,
|
66 |
+
"eval_samples_per_second": 230.229,
|
67 |
+
"eval_steps_per_second": 14.504,
|
68 |
+
"step": 600
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 1.18,
|
72 |
+
"eval_accuracy": 0.9340000152587891,
|
73 |
+
"eval_loss": 0.3148828148841858,
|
74 |
+
"eval_runtime": 4.3516,
|
75 |
+
"eval_samples_per_second": 229.803,
|
76 |
+
"eval_steps_per_second": 14.478,
|
77 |
+
"step": 700
|
78 |
+
},
|
79 |
+
{
|
80 |
+
"epoch": 1.35,
|
81 |
+
"eval_accuracy": 0.9399999976158142,
|
82 |
+
"eval_loss": 0.30127567052841187,
|
83 |
+
"eval_runtime": 4.3657,
|
84 |
+
"eval_samples_per_second": 229.057,
|
85 |
+
"eval_steps_per_second": 14.431,
|
86 |
+
"step": 800
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 1.52,
|
90 |
+
"eval_accuracy": 0.949999988079071,
|
91 |
+
"eval_loss": 0.18032802641391754,
|
92 |
+
"eval_runtime": 4.3502,
|
93 |
+
"eval_samples_per_second": 229.873,
|
94 |
+
"eval_steps_per_second": 14.482,
|
95 |
+
"step": 900
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 1.69,
|
99 |
+
"learning_rate": 2.8927364864864863e-05,
|
100 |
+
"loss": 0.0779,
|
101 |
+
"step": 1000
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 1.69,
|
105 |
+
"eval_accuracy": 0.906000018119812,
|
106 |
+
"eval_loss": 0.4576403498649597,
|
107 |
+
"eval_runtime": 4.3481,
|
108 |
+
"eval_samples_per_second": 229.983,
|
109 |
+
"eval_steps_per_second": 14.489,
|
110 |
+
"step": 1000
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 1.86,
|
114 |
+
"eval_accuracy": 0.9229999780654907,
|
115 |
+
"eval_loss": 0.30020907521247864,
|
116 |
+
"eval_runtime": 4.3386,
|
117 |
+
"eval_samples_per_second": 230.489,
|
118 |
+
"eval_steps_per_second": 14.521,
|
119 |
+
"step": 1100
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 2.03,
|
123 |
+
"eval_accuracy": 0.9449999928474426,
|
124 |
+
"eval_loss": 0.23847746849060059,
|
125 |
+
"eval_runtime": 4.3357,
|
126 |
+
"eval_samples_per_second": 230.642,
|
127 |
+
"eval_steps_per_second": 14.53,
|
128 |
+
"step": 1200
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 2.2,
|
132 |
+
"eval_accuracy": 0.9580000042915344,
|
133 |
+
"eval_loss": 0.2000477910041809,
|
134 |
+
"eval_runtime": 4.3536,
|
135 |
+
"eval_samples_per_second": 229.694,
|
136 |
+
"eval_steps_per_second": 14.471,
|
137 |
+
"step": 1300
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 2.36,
|
141 |
+
"eval_accuracy": 0.9639999866485596,
|
142 |
+
"eval_loss": 0.15715740621089935,
|
143 |
+
"eval_runtime": 4.3678,
|
144 |
+
"eval_samples_per_second": 228.95,
|
145 |
+
"eval_steps_per_second": 14.424,
|
146 |
+
"step": 1400
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 2.53,
|
150 |
+
"learning_rate": 1.8369932432432433e-05,
|
151 |
+
"loss": 0.0475,
|
152 |
+
"step": 1500
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 2.53,
|
156 |
+
"eval_accuracy": 0.9620000123977661,
|
157 |
+
"eval_loss": 0.19045613706111908,
|
158 |
+
"eval_runtime": 4.3587,
|
159 |
+
"eval_samples_per_second": 229.426,
|
160 |
+
"eval_steps_per_second": 14.454,
|
161 |
+
"step": 1500
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 2.7,
|
165 |
+
"eval_accuracy": 0.9639999866485596,
|
166 |
+
"eval_loss": 0.20174138247966766,
|
167 |
+
"eval_runtime": 4.348,
|
168 |
+
"eval_samples_per_second": 229.989,
|
169 |
+
"eval_steps_per_second": 14.489,
|
170 |
+
"step": 1600
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 2.87,
|
174 |
+
"eval_accuracy": 0.9520000219345093,
|
175 |
+
"eval_loss": 0.23974017798900604,
|
176 |
+
"eval_runtime": 4.3452,
|
177 |
+
"eval_samples_per_second": 230.141,
|
178 |
+
"eval_steps_per_second": 14.499,
|
179 |
+
"step": 1700
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 3.04,
|
183 |
+
"eval_accuracy": 0.9549999833106995,
|
184 |
+
"eval_loss": 0.1878175288438797,
|
185 |
+
"eval_runtime": 4.3435,
|
186 |
+
"eval_samples_per_second": 230.227,
|
187 |
+
"eval_steps_per_second": 14.504,
|
188 |
+
"step": 1800
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 3.21,
|
192 |
+
"eval_accuracy": 0.9520000219345093,
|
193 |
+
"eval_loss": 0.2655267119407654,
|
194 |
+
"eval_runtime": 4.3468,
|
195 |
+
"eval_samples_per_second": 230.053,
|
196 |
+
"eval_steps_per_second": 14.493,
|
197 |
+
"step": 1900
|
198 |
+
},
|
199 |
+
{
|
200 |
+
"epoch": 3.38,
|
201 |
+
"learning_rate": 7.8125e-06,
|
202 |
+
"loss": 0.0362,
|
203 |
+
"step": 2000
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 3.38,
|
207 |
+
"eval_accuracy": 0.9629999995231628,
|
208 |
+
"eval_loss": 0.1756611317396164,
|
209 |
+
"eval_runtime": 4.3509,
|
210 |
+
"eval_samples_per_second": 229.836,
|
211 |
+
"eval_steps_per_second": 14.48,
|
212 |
+
"step": 2000
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 3.55,
|
216 |
+
"eval_accuracy": 0.9620000123977661,
|
217 |
+
"eval_loss": 0.19551004469394684,
|
218 |
+
"eval_runtime": 4.3485,
|
219 |
+
"eval_samples_per_second": 229.966,
|
220 |
+
"eval_steps_per_second": 14.488,
|
221 |
+
"step": 2100
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 3.72,
|
225 |
+
"eval_accuracy": 0.9559999704360962,
|
226 |
+
"eval_loss": 0.21292370557785034,
|
227 |
+
"eval_runtime": 4.357,
|
228 |
+
"eval_samples_per_second": 229.514,
|
229 |
+
"eval_steps_per_second": 14.459,
|
230 |
+
"step": 2200
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 3.89,
|
234 |
+
"eval_accuracy": 0.9559999704360962,
|
235 |
+
"eval_loss": 0.22236208617687225,
|
236 |
+
"eval_runtime": 4.3612,
|
237 |
+
"eval_samples_per_second": 229.297,
|
238 |
+
"eval_steps_per_second": 14.446,
|
239 |
+
"step": 2300
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 4.0,
|
243 |
+
"step": 2368,
|
244 |
+
"total_flos": 739476514563456.0,
|
245 |
+
"train_loss": 0.07163469533662538,
|
246 |
+
"train_runtime": 1355.547,
|
247 |
+
"train_samples_per_second": 27.909,
|
248 |
+
"train_steps_per_second": 1.747
|
249 |
+
}
|
250 |
+
],
|
251 |
+
"max_steps": 2368,
|
252 |
+
"num_train_epochs": 4,
|
253 |
+
"total_flos": 739476514563456.0,
|
254 |
+
"trial_name": null,
|
255 |
+
"trial_params": null
|
256 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:48ba4bc41792dc921e6b818b1569afd0bc40189c659f9aeb1570ac39136e0ced
|
3 |
+
size 2863
|