tiagoblima
commited on
Commit
•
14699d6
1
Parent(s):
1461b0e
End of training
Browse files- README.md +3 -1
- all_results.json +9 -9
- eval_results.json +5 -5
- train_results.json +5 -5
- trainer_state.json +172 -14
README.md
CHANGED
@@ -3,6 +3,8 @@ license: mit
|
|
3 |
base_model: unicamp-dl/ptt5-large-t5-vocab
|
4 |
tags:
|
5 |
- generated_from_trainer
|
|
|
|
|
6 |
model-index:
|
7 |
- name: t5_large-qg-aap
|
8 |
results: []
|
@@ -13,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
13 |
|
14 |
# t5_large-qg-aap
|
15 |
|
16 |
-
This model is a fine-tuned version of [unicamp-dl/ptt5-large-t5-vocab](https://huggingface.co/unicamp-dl/ptt5-large-t5-vocab) on
|
17 |
It achieves the following results on the evaluation set:
|
18 |
- Loss: 7.4208
|
19 |
|
|
|
3 |
base_model: unicamp-dl/ptt5-large-t5-vocab
|
4 |
tags:
|
5 |
- generated_from_trainer
|
6 |
+
datasets:
|
7 |
+
- tiagoblima/qg_squad_v1_pt
|
8 |
model-index:
|
9 |
- name: t5_large-qg-aap
|
10 |
results: []
|
|
|
15 |
|
16 |
# t5_large-qg-aap
|
17 |
|
18 |
+
This model is a fine-tuned version of [unicamp-dl/ptt5-large-t5-vocab](https://huggingface.co/unicamp-dl/ptt5-large-t5-vocab) on the tiagoblima/qg_squad_v1_pt dataset.
|
19 |
It achieves the following results on the evaluation set:
|
20 |
- Loss: 7.4208
|
21 |
|
all_results.json
CHANGED
@@ -1,13 +1,13 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"eval_loss":
|
4 |
-
"eval_runtime":
|
5 |
"eval_samples": 8869,
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second":
|
8 |
-
"train_loss":
|
9 |
-
"train_runtime":
|
10 |
"train_samples": 51704,
|
11 |
-
"train_samples_per_second":
|
12 |
-
"train_steps_per_second": 0.
|
13 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"eval_loss": 7.420807361602783,
|
4 |
+
"eval_runtime": 587.2121,
|
5 |
"eval_samples": 8869,
|
6 |
+
"eval_samples_per_second": 15.104,
|
7 |
+
"eval_steps_per_second": 7.553,
|
8 |
+
"train_loss": 6.77048750395822,
|
9 |
+
"train_runtime": 16676.4775,
|
10 |
"train_samples": 51704,
|
11 |
+
"train_samples_per_second": 15.502,
|
12 |
+
"train_steps_per_second": 0.121
|
13 |
}
|
eval_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"eval_loss":
|
4 |
-
"eval_runtime":
|
5 |
"eval_samples": 8869,
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second":
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"eval_loss": 7.420807361602783,
|
4 |
+
"eval_runtime": 587.2121,
|
5 |
"eval_samples": 8869,
|
6 |
+
"eval_samples_per_second": 15.104,
|
7 |
+
"eval_steps_per_second": 7.553
|
8 |
}
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"train_loss":
|
4 |
-
"train_runtime":
|
5 |
"train_samples": 51704,
|
6 |
-
"train_samples_per_second":
|
7 |
-
"train_steps_per_second": 0.
|
8 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"train_loss": 6.77048750395822,
|
4 |
+
"train_runtime": 16676.4775,
|
5 |
"train_samples": 51704,
|
6 |
+
"train_samples_per_second": 15.502,
|
7 |
+
"train_steps_per_second": 0.121
|
8 |
}
|
trainer_state.json
CHANGED
@@ -1,30 +1,188 @@
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
-
"epoch":
|
5 |
"eval_steps": 500,
|
6 |
-
"global_step":
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
{
|
12 |
"epoch": 1.0,
|
13 |
-
"
|
14 |
-
"
|
15 |
-
"
|
16 |
-
"
|
17 |
-
"
|
18 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
19 |
}
|
20 |
],
|
21 |
-
"logging_steps":
|
22 |
-
"max_steps":
|
23 |
-
"
|
24 |
-
"num_train_epochs": 1,
|
25 |
"save_steps": 500,
|
26 |
-
"total_flos":
|
27 |
-
"train_batch_size": 128,
|
28 |
"trial_name": null,
|
29 |
"trial_params": null
|
30 |
}
|
|
|
1 |
{
|
2 |
"best_metric": null,
|
3 |
"best_model_checkpoint": null,
|
4 |
+
"epoch": 5.0,
|
5 |
"eval_steps": 500,
|
6 |
+
"global_step": 2020,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.25,
|
13 |
+
"learning_rate": 0.0028514851485148514,
|
14 |
+
"loss": 8.3611,
|
15 |
+
"step": 100
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.5,
|
19 |
+
"learning_rate": 0.002702970297029703,
|
20 |
+
"loss": 8.0607,
|
21 |
+
"step": 200
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.74,
|
25 |
+
"learning_rate": 0.0025544554455445546,
|
26 |
+
"loss": 7.7784,
|
27 |
+
"step": 300
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.99,
|
31 |
+
"learning_rate": 0.0024059405940594063,
|
32 |
+
"loss": 7.5413,
|
33 |
+
"step": 400
|
34 |
+
},
|
35 |
{
|
36 |
"epoch": 1.0,
|
37 |
+
"eval_loss": 8.850217819213867,
|
38 |
+
"eval_runtime": 584.1219,
|
39 |
+
"eval_samples_per_second": 15.183,
|
40 |
+
"eval_steps_per_second": 7.593,
|
41 |
+
"step": 404
|
42 |
+
},
|
43 |
+
{
|
44 |
+
"epoch": 1.24,
|
45 |
+
"learning_rate": 0.0022574257425742577,
|
46 |
+
"loss": 7.3312,
|
47 |
+
"step": 500
|
48 |
+
},
|
49 |
+
{
|
50 |
+
"epoch": 1.49,
|
51 |
+
"learning_rate": 0.0021089108910891086,
|
52 |
+
"loss": 7.1317,
|
53 |
+
"step": 600
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 1.73,
|
57 |
+
"learning_rate": 0.0019603960396039604,
|
58 |
+
"loss": 6.9568,
|
59 |
+
"step": 700
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 1.98,
|
63 |
+
"learning_rate": 0.0018118811881188118,
|
64 |
+
"loss": 6.7965,
|
65 |
+
"step": 800
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 2.0,
|
69 |
+
"eval_loss": 8.118351936340332,
|
70 |
+
"eval_runtime": 586.5561,
|
71 |
+
"eval_samples_per_second": 15.12,
|
72 |
+
"eval_steps_per_second": 7.561,
|
73 |
+
"step": 808
|
74 |
+
},
|
75 |
+
{
|
76 |
+
"epoch": 2.23,
|
77 |
+
"learning_rate": 0.0016633663366336635,
|
78 |
+
"loss": 6.6621,
|
79 |
+
"step": 900
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 2.48,
|
83 |
+
"learning_rate": 0.001514851485148515,
|
84 |
+
"loss": 6.5571,
|
85 |
+
"step": 1000
|
86 |
+
},
|
87 |
+
{
|
88 |
+
"epoch": 2.72,
|
89 |
+
"learning_rate": 0.0013663366336633665,
|
90 |
+
"loss": 6.4736,
|
91 |
+
"step": 1100
|
92 |
+
},
|
93 |
+
{
|
94 |
+
"epoch": 2.97,
|
95 |
+
"learning_rate": 0.001217821782178218,
|
96 |
+
"loss": 6.3963,
|
97 |
+
"step": 1200
|
98 |
+
},
|
99 |
+
{
|
100 |
+
"epoch": 3.0,
|
101 |
+
"eval_loss": 7.695041179656982,
|
102 |
+
"eval_runtime": 586.9195,
|
103 |
+
"eval_samples_per_second": 15.111,
|
104 |
+
"eval_steps_per_second": 7.556,
|
105 |
+
"step": 1212
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 3.22,
|
109 |
+
"learning_rate": 0.0010693069306930692,
|
110 |
+
"loss": 6.3235,
|
111 |
+
"step": 1300
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 3.47,
|
115 |
+
"learning_rate": 0.0009207920792079207,
|
116 |
+
"loss": 6.283,
|
117 |
+
"step": 1400
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"epoch": 3.71,
|
121 |
+
"learning_rate": 0.0007722772277227723,
|
122 |
+
"loss": 6.2224,
|
123 |
+
"step": 1500
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 3.96,
|
127 |
+
"learning_rate": 0.0006237623762376238,
|
128 |
+
"loss": 6.1664,
|
129 |
+
"step": 1600
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 4.0,
|
133 |
+
"eval_loss": 7.485514163970947,
|
134 |
+
"eval_runtime": 587.006,
|
135 |
+
"eval_samples_per_second": 15.109,
|
136 |
+
"eval_steps_per_second": 7.555,
|
137 |
+
"step": 1616
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 4.21,
|
141 |
+
"learning_rate": 0.0004752475247524753,
|
142 |
+
"loss": 6.1536,
|
143 |
+
"step": 1700
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 4.46,
|
147 |
+
"learning_rate": 0.00032673267326732675,
|
148 |
+
"loss": 6.1402,
|
149 |
+
"step": 1800
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 4.7,
|
153 |
+
"learning_rate": 0.0001782178217821782,
|
154 |
+
"loss": 6.0973,
|
155 |
+
"step": 1900
|
156 |
+
},
|
157 |
+
{
|
158 |
+
"epoch": 4.95,
|
159 |
+
"learning_rate": 2.9702970297029706e-05,
|
160 |
+
"loss": 6.1028,
|
161 |
+
"step": 2000
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 5.0,
|
165 |
+
"eval_loss": 7.420807361602783,
|
166 |
+
"eval_runtime": 587.2049,
|
167 |
+
"eval_samples_per_second": 15.104,
|
168 |
+
"eval_steps_per_second": 7.553,
|
169 |
+
"step": 2020
|
170 |
+
},
|
171 |
+
{
|
172 |
+
"epoch": 5.0,
|
173 |
+
"step": 2020,
|
174 |
+
"total_flos": 4.4078196477394944e+17,
|
175 |
+
"train_loss": 6.77048750395822,
|
176 |
+
"train_runtime": 16676.4775,
|
177 |
+
"train_samples_per_second": 15.502,
|
178 |
+
"train_steps_per_second": 0.121
|
179 |
}
|
180 |
],
|
181 |
+
"logging_steps": 100,
|
182 |
+
"max_steps": 2020,
|
183 |
+
"num_train_epochs": 5,
|
|
|
184 |
"save_steps": 500,
|
185 |
+
"total_flos": 4.4078196477394944e+17,
|
|
|
186 |
"trial_name": null,
|
187 |
"trial_params": null
|
188 |
}
|