Jason Wu
commited on
Commit
•
fe02190
1
Parent(s):
e451506
init
Browse files- config.json +63 -0
- special_tokens_map.json +1 -0
- tokenizer_config.json +1 -0
- trainer_state.json +257 -0
- training_args.bin +0 -0
config.json
ADDED
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "allenai/unifiedqa-t5-base",
|
3 |
+
"architectures": [
|
4 |
+
"T5ForConditionalGeneration"
|
5 |
+
],
|
6 |
+
"d_ff": 3072,
|
7 |
+
"d_kv": 64,
|
8 |
+
"d_model": 768,
|
9 |
+
"decoder_start_token_id": 0,
|
10 |
+
"dropout_rate": 0.1,
|
11 |
+
"early_stopping": true,
|
12 |
+
"eos_token_id": 1,
|
13 |
+
"feed_forward_proj": "relu",
|
14 |
+
"initializer_factor": 1.0,
|
15 |
+
"is_encoder_decoder": true,
|
16 |
+
"layer_norm_epsilon": 1e-06,
|
17 |
+
"length_penalty": 2.0,
|
18 |
+
"max_length": 200,
|
19 |
+
"min_length": 30,
|
20 |
+
"model_type": "t5",
|
21 |
+
"n_positions": 512,
|
22 |
+
"no_repeat_ngram_size": 3,
|
23 |
+
"num_beams": 4,
|
24 |
+
"num_decoder_layers": 12,
|
25 |
+
"num_heads": 12,
|
26 |
+
"num_layers": 12,
|
27 |
+
"output_past": true,
|
28 |
+
"pad_token_id": 0,
|
29 |
+
"prefix": "summarize: ",
|
30 |
+
"relative_attention_num_buckets": 32,
|
31 |
+
"task_specific_params": {
|
32 |
+
"summarization": {
|
33 |
+
"early_stopping": true,
|
34 |
+
"length_penalty": 2.0,
|
35 |
+
"max_length": 200,
|
36 |
+
"min_length": 30,
|
37 |
+
"no_repeat_ngram_size": 3,
|
38 |
+
"num_beams": 4,
|
39 |
+
"prefix": "summarize: "
|
40 |
+
},
|
41 |
+
"translation_en_to_de": {
|
42 |
+
"early_stopping": true,
|
43 |
+
"max_length": 300,
|
44 |
+
"num_beams": 4,
|
45 |
+
"prefix": "translate English to German: "
|
46 |
+
},
|
47 |
+
"translation_en_to_fr": {
|
48 |
+
"early_stopping": true,
|
49 |
+
"max_length": 300,
|
50 |
+
"num_beams": 4,
|
51 |
+
"prefix": "translate English to French: "
|
52 |
+
},
|
53 |
+
"translation_en_to_ro": {
|
54 |
+
"early_stopping": true,
|
55 |
+
"max_length": 300,
|
56 |
+
"num_beams": 4,
|
57 |
+
"prefix": "translate English to Romanian: "
|
58 |
+
}
|
59 |
+
},
|
60 |
+
"transformers_version": "4.5.1",
|
61 |
+
"use_cache": true,
|
62 |
+
"vocab_size": 32128
|
63 |
+
}
|
special_tokens_map.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "additional_special_tokens": ["<extra_id_0>", "<extra_id_1>", "<extra_id_2>", "<extra_id_3>", "<extra_id_4>", "<extra_id_5>", "<extra_id_6>", "<extra_id_7>", "<extra_id_8>", "<extra_id_9>", "<extra_id_10>", "<extra_id_11>", "<extra_id_12>", "<extra_id_13>", "<extra_id_14>", "<extra_id_15>", "<extra_id_16>", "<extra_id_17>", "<extra_id_18>", "<extra_id_19>", "<extra_id_20>", "<extra_id_21>", "<extra_id_22>", "<extra_id_23>", "<extra_id_24>", "<extra_id_25>", "<extra_id_26>", "<extra_id_27>", "<extra_id_28>", "<extra_id_29>", "<extra_id_30>", "<extra_id_31>", "<extra_id_32>", "<extra_id_33>", "<extra_id_34>", "<extra_id_35>", "<extra_id_36>", "<extra_id_37>", "<extra_id_38>", "<extra_id_39>", "<extra_id_40>", "<extra_id_41>", "<extra_id_42>", "<extra_id_43>", "<extra_id_44>", "<extra_id_45>", "<extra_id_46>", "<extra_id_47>", "<extra_id_48>", "<extra_id_49>", "<extra_id_50>", "<extra_id_51>", "<extra_id_52>", "<extra_id_53>", "<extra_id_54>", "<extra_id_55>", "<extra_id_56>", "<extra_id_57>", "<extra_id_58>", "<extra_id_59>", "<extra_id_60>", "<extra_id_61>", "<extra_id_62>", "<extra_id_63>", "<extra_id_64>", "<extra_id_65>", "<extra_id_66>", "<extra_id_67>", "<extra_id_68>", "<extra_id_69>", "<extra_id_70>", "<extra_id_71>", "<extra_id_72>", "<extra_id_73>", "<extra_id_74>", "<extra_id_75>", "<extra_id_76>", "<extra_id_77>", "<extra_id_78>", "<extra_id_79>", "<extra_id_80>", "<extra_id_81>", "<extra_id_82>", "<extra_id_83>", "<extra_id_84>", "<extra_id_85>", "<extra_id_86>", "<extra_id_87>", "<extra_id_88>", "<extra_id_89>", "<extra_id_90>", "<extra_id_91>", "<extra_id_92>", "<extra_id_93>", "<extra_id_94>", "<extra_id_95>", "<extra_id_96>", "<extra_id_97>", "<extra_id_98>", "<extra_id_99>"]}
|
tokenizer_config.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "extra_ids": 100, "additional_special_tokens": ["<extra_id_0>", "<extra_id_1>", "<extra_id_2>", "<extra_id_3>", "<extra_id_4>", "<extra_id_5>", "<extra_id_6>", "<extra_id_7>", "<extra_id_8>", "<extra_id_9>", "<extra_id_10>", "<extra_id_11>", "<extra_id_12>", "<extra_id_13>", "<extra_id_14>", "<extra_id_15>", "<extra_id_16>", "<extra_id_17>", "<extra_id_18>", "<extra_id_19>", "<extra_id_20>", "<extra_id_21>", "<extra_id_22>", "<extra_id_23>", "<extra_id_24>", "<extra_id_25>", "<extra_id_26>", "<extra_id_27>", "<extra_id_28>", "<extra_id_29>", "<extra_id_30>", "<extra_id_31>", "<extra_id_32>", "<extra_id_33>", "<extra_id_34>", "<extra_id_35>", "<extra_id_36>", "<extra_id_37>", "<extra_id_38>", "<extra_id_39>", "<extra_id_40>", "<extra_id_41>", "<extra_id_42>", "<extra_id_43>", "<extra_id_44>", "<extra_id_45>", "<extra_id_46>", "<extra_id_47>", "<extra_id_48>", "<extra_id_49>", "<extra_id_50>", "<extra_id_51>", "<extra_id_52>", "<extra_id_53>", "<extra_id_54>", "<extra_id_55>", "<extra_id_56>", "<extra_id_57>", "<extra_id_58>", "<extra_id_59>", "<extra_id_60>", "<extra_id_61>", "<extra_id_62>", "<extra_id_63>", "<extra_id_64>", "<extra_id_65>", "<extra_id_66>", "<extra_id_67>", "<extra_id_68>", "<extra_id_69>", "<extra_id_70>", "<extra_id_71>", "<extra_id_72>", "<extra_id_73>", "<extra_id_74>", "<extra_id_75>", "<extra_id_76>", "<extra_id_77>", "<extra_id_78>", "<extra_id_79>", "<extra_id_80>", "<extra_id_81>", "<extra_id_82>", "<extra_id_83>", "<extra_id_84>", "<extra_id_85>", "<extra_id_86>", "<extra_id_87>", "<extra_id_88>", "<extra_id_89>", "<extra_id_90>", "<extra_id_91>", "<extra_id_92>", "<extra_id_93>", "<extra_id_94>", "<extra_id_95>", "<extra_id_96>", "<extra_id_97>", "<extra_id_98>", "<extra_id_99>"], "model_max_length": 512, "special_tokens_map_file": "/root/.cache/huggingface/transformers/6de9d1094f721d3aa84e0b73e86256cd631e809641d8e864fd1ec11a261ba2de.c94798918c92ded6aeef2d2f0e666d2cc4145eca1aa6e1336fde07f2e13e2f46", "name_or_path": "allenai/unifiedqa-t5-base"}
|
trainer_state.json
ADDED
@@ -0,0 +1,257 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 10.0,
|
5 |
+
"global_step": 68220,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.73,
|
12 |
+
"learning_rate": 2.780123131046614e-05,
|
13 |
+
"loss": 0.7441,
|
14 |
+
"step": 5000
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.73,
|
18 |
+
"eval_gen_len": 33.7,
|
19 |
+
"eval_loss": 0.633007287979126,
|
20 |
+
"eval_rouge1": 19.0253,
|
21 |
+
"eval_rouge2": 11.6724,
|
22 |
+
"eval_rougeL": 18.7632,
|
23 |
+
"eval_rougeLsum": 18.8697,
|
24 |
+
"eval_runtime": 1962.245,
|
25 |
+
"eval_samples_per_second": 1.739,
|
26 |
+
"step": 5000
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 1.47,
|
30 |
+
"learning_rate": 2.5602462620932278e-05,
|
31 |
+
"loss": 0.5879,
|
32 |
+
"step": 10000
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 1.47,
|
36 |
+
"eval_gen_len": 33.8,
|
37 |
+
"eval_loss": 0.6442069411277771,
|
38 |
+
"eval_rouge1": 18.9287,
|
39 |
+
"eval_rouge2": 11.7109,
|
40 |
+
"eval_rougeL": 18.6885,
|
41 |
+
"eval_rougeLsum": 18.7824,
|
42 |
+
"eval_runtime": 1627.9481,
|
43 |
+
"eval_samples_per_second": 2.097,
|
44 |
+
"step": 10000
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 2.2,
|
48 |
+
"learning_rate": 2.3403693931398417e-05,
|
49 |
+
"loss": 0.5011,
|
50 |
+
"step": 15000
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 2.2,
|
54 |
+
"eval_gen_len": 33.7,
|
55 |
+
"eval_loss": 0.6601932048797607,
|
56 |
+
"eval_rouge1": 19.157,
|
57 |
+
"eval_rouge2": 11.7904,
|
58 |
+
"eval_rougeL": 18.8919,
|
59 |
+
"eval_rougeLsum": 18.9663,
|
60 |
+
"eval_runtime": 1218.763,
|
61 |
+
"eval_samples_per_second": 2.8,
|
62 |
+
"step": 15000
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 2.93,
|
66 |
+
"learning_rate": 2.120492524186456e-05,
|
67 |
+
"loss": 0.4365,
|
68 |
+
"step": 20000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 2.93,
|
72 |
+
"eval_gen_len": 33.8,
|
73 |
+
"eval_loss": 0.6567361950874329,
|
74 |
+
"eval_rouge1": 19.402,
|
75 |
+
"eval_rouge2": 11.846,
|
76 |
+
"eval_rougeL": 19.1085,
|
77 |
+
"eval_rougeLsum": 19.1695,
|
78 |
+
"eval_runtime": 1214.5476,
|
79 |
+
"eval_samples_per_second": 2.81,
|
80 |
+
"step": 20000
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 3.66,
|
84 |
+
"learning_rate": 1.9006156552330694e-05,
|
85 |
+
"loss": 0.3731,
|
86 |
+
"step": 25000
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 3.66,
|
90 |
+
"eval_gen_len": 33.5,
|
91 |
+
"eval_loss": 0.7168177366256714,
|
92 |
+
"eval_rouge1": 20.1185,
|
93 |
+
"eval_rouge2": 12.3512,
|
94 |
+
"eval_rougeL": 19.8009,
|
95 |
+
"eval_rougeLsum": 19.9205,
|
96 |
+
"eval_runtime": 1198.2455,
|
97 |
+
"eval_samples_per_second": 2.848,
|
98 |
+
"step": 25000
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 4.4,
|
102 |
+
"learning_rate": 1.6807387862796836e-05,
|
103 |
+
"loss": 0.3377,
|
104 |
+
"step": 30000
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 4.4,
|
108 |
+
"eval_gen_len": 33.4,
|
109 |
+
"eval_loss": 0.7644200325012207,
|
110 |
+
"eval_rouge1": 20.026,
|
111 |
+
"eval_rouge2": 12.1819,
|
112 |
+
"eval_rougeL": 19.7149,
|
113 |
+
"eval_rougeLsum": 19.7857,
|
114 |
+
"eval_runtime": 1196.6336,
|
115 |
+
"eval_samples_per_second": 2.852,
|
116 |
+
"step": 30000
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 5.13,
|
120 |
+
"learning_rate": 1.4608619173262973e-05,
|
121 |
+
"loss": 0.3003,
|
122 |
+
"step": 35000
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 5.13,
|
126 |
+
"eval_gen_len": 33.5,
|
127 |
+
"eval_loss": 0.8117037415504456,
|
128 |
+
"eval_rouge1": 20.0409,
|
129 |
+
"eval_rouge2": 12.1197,
|
130 |
+
"eval_rougeL": 19.737,
|
131 |
+
"eval_rougeLsum": 19.8529,
|
132 |
+
"eval_runtime": 1196.6657,
|
133 |
+
"eval_samples_per_second": 2.852,
|
134 |
+
"step": 35000
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 5.86,
|
138 |
+
"learning_rate": 1.2409850483729111e-05,
|
139 |
+
"loss": 0.272,
|
140 |
+
"step": 40000
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 5.86,
|
144 |
+
"eval_gen_len": 33.4,
|
145 |
+
"eval_loss": 0.8414672613143921,
|
146 |
+
"eval_rouge1": 20.3612,
|
147 |
+
"eval_rouge2": 12.4433,
|
148 |
+
"eval_rougeL": 20.0321,
|
149 |
+
"eval_rougeLsum": 20.1456,
|
150 |
+
"eval_runtime": 1188.9907,
|
151 |
+
"eval_samples_per_second": 2.871,
|
152 |
+
"step": 40000
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 6.6,
|
156 |
+
"learning_rate": 1.0211081794195251e-05,
|
157 |
+
"loss": 0.2467,
|
158 |
+
"step": 45000
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 6.6,
|
162 |
+
"eval_gen_len": 33.6,
|
163 |
+
"eval_loss": 0.9067189693450928,
|
164 |
+
"eval_rouge1": 19.9276,
|
165 |
+
"eval_rouge2": 12.162,
|
166 |
+
"eval_rougeL": 19.6218,
|
167 |
+
"eval_rougeLsum": 19.7163,
|
168 |
+
"eval_runtime": 1207.9436,
|
169 |
+
"eval_samples_per_second": 2.825,
|
170 |
+
"step": 45000
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 7.33,
|
174 |
+
"learning_rate": 8.01231310466139e-06,
|
175 |
+
"loss": 0.2253,
|
176 |
+
"step": 50000
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 7.33,
|
180 |
+
"eval_gen_len": 33.2,
|
181 |
+
"eval_loss": 0.9487330317497253,
|
182 |
+
"eval_rouge1": 20.7483,
|
183 |
+
"eval_rouge2": 12.591,
|
184 |
+
"eval_rougeL": 20.4197,
|
185 |
+
"eval_rougeLsum": 20.5083,
|
186 |
+
"eval_runtime": 1187.4731,
|
187 |
+
"eval_samples_per_second": 2.874,
|
188 |
+
"step": 50000
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 8.06,
|
192 |
+
"learning_rate": 5.813544415127529e-06,
|
193 |
+
"loss": 0.2141,
|
194 |
+
"step": 55000
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 8.06,
|
198 |
+
"eval_gen_len": 33.3,
|
199 |
+
"eval_loss": 0.9719064235687256,
|
200 |
+
"eval_rouge1": 20.2944,
|
201 |
+
"eval_rouge2": 12.35,
|
202 |
+
"eval_rougeL": 19.9563,
|
203 |
+
"eval_rougeLsum": 20.0559,
|
204 |
+
"eval_runtime": 1191.6755,
|
205 |
+
"eval_samples_per_second": 2.864,
|
206 |
+
"step": 55000
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 8.8,
|
210 |
+
"learning_rate": 3.6147757255936676e-06,
|
211 |
+
"loss": 0.1951,
|
212 |
+
"step": 60000
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 8.8,
|
216 |
+
"eval_gen_len": 33.5,
|
217 |
+
"eval_loss": 0.9840695261955261,
|
218 |
+
"eval_rouge1": 20.0284,
|
219 |
+
"eval_rouge2": 12.0984,
|
220 |
+
"eval_rougeL": 19.6653,
|
221 |
+
"eval_rougeLsum": 19.7619,
|
222 |
+
"eval_runtime": 1228.4982,
|
223 |
+
"eval_samples_per_second": 2.778,
|
224 |
+
"step": 60000
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 9.53,
|
228 |
+
"learning_rate": 1.4160070360598064e-06,
|
229 |
+
"loss": 0.185,
|
230 |
+
"step": 65000
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 9.53,
|
234 |
+
"eval_gen_len": 33.4,
|
235 |
+
"eval_loss": 1.021308422088623,
|
236 |
+
"eval_rouge1": 20.5124,
|
237 |
+
"eval_rouge2": 12.4505,
|
238 |
+
"eval_rougeL": 20.1615,
|
239 |
+
"eval_rougeLsum": 20.2679,
|
240 |
+
"eval_runtime": 1216.5148,
|
241 |
+
"eval_samples_per_second": 2.806,
|
242 |
+
"step": 65000
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 10.0,
|
246 |
+
"step": 68220,
|
247 |
+
"total_flos": 1.7131748370415616e+17,
|
248 |
+
"train_runtime": 36592.0856,
|
249 |
+
"train_samples_per_second": 1.864
|
250 |
+
}
|
251 |
+
],
|
252 |
+
"max_steps": 68220,
|
253 |
+
"num_train_epochs": 10,
|
254 |
+
"total_flos": 1.7131748370415616e+17,
|
255 |
+
"trial_name": null,
|
256 |
+
"trial_params": null
|
257 |
+
}
|
training_args.bin
ADDED
Binary file (2.61 kB). View file
|
|