End of training
Browse files- README.md +21 -8
- all_results.json +18 -0
- eval_results.json +13 -0
- train_results.json +8 -0
- trainer_state.json +303 -0
README.md
CHANGED
@@ -1,11 +1,24 @@
|
|
1 |
---
|
|
|
2 |
tags:
|
3 |
- generated_from_trainer
|
|
|
|
|
4 |
metrics:
|
5 |
- rouge
|
6 |
model-index:
|
7 |
- name: longt5_xl_summ_screen_bp_only_30
|
8 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
---
|
10 |
|
11 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -13,14 +26,14 @@ should probably proofread and complete it, then remove this comment. -->
|
|
13 |
|
14 |
# longt5_xl_summ_screen_bp_only_30
|
15 |
|
16 |
-
This model
|
17 |
It achieves the following results on the evaluation set:
|
18 |
-
- Loss: 2.
|
19 |
-
- Rouge1:
|
20 |
-
- Rouge2:
|
21 |
-
- Rougel:
|
22 |
-
- Rougelsum:
|
23 |
-
- Gen Len:
|
24 |
|
25 |
## Model description
|
26 |
|
|
|
1 |
---
|
2 |
+
base_model: /exports/eddie/scratch/s1970716/models/summarization/longt5_xl_summ_screen_bp_only/checkpoint-210
|
3 |
tags:
|
4 |
- generated_from_trainer
|
5 |
+
datasets:
|
6 |
+
- learn3r/summ_screen_fd_bp
|
7 |
metrics:
|
8 |
- rouge
|
9 |
model-index:
|
10 |
- name: longt5_xl_summ_screen_bp_only_30
|
11 |
+
results:
|
12 |
+
- task:
|
13 |
+
name: Summarization
|
14 |
+
type: summarization
|
15 |
+
dataset:
|
16 |
+
name: learn3r/summ_screen_fd_bp
|
17 |
+
type: learn3r/summ_screen_fd_bp
|
18 |
+
metrics:
|
19 |
+
- name: Rouge1
|
20 |
+
type: rouge
|
21 |
+
value: 40.4943
|
22 |
---
|
23 |
|
24 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
26 |
|
27 |
# longt5_xl_summ_screen_bp_only_30
|
28 |
|
29 |
+
This model is a fine-tuned version of [/exports/eddie/scratch/s1970716/models/summarization/longt5_xl_summ_screen_bp_only/checkpoint-210](https://huggingface.co//exports/eddie/scratch/s1970716/models/summarization/longt5_xl_summ_screen_bp_only/checkpoint-210) on the learn3r/summ_screen_fd_bp dataset.
|
30 |
It achieves the following results on the evaluation set:
|
31 |
+
- Loss: 2.2397
|
32 |
+
- Rouge1: 40.4943
|
33 |
+
- Rouge2: 16.4695
|
34 |
+
- Rougel: 28.0964
|
35 |
+
- Rougelsum: 38.3693
|
36 |
+
- Gen Len: 246.3491
|
37 |
|
38 |
## Model description
|
39 |
|
all_results.json
ADDED
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 13.18,
|
3 |
+
"eval_gen_len": 246.3491124260355,
|
4 |
+
"eval_loss": 2.2397100925445557,
|
5 |
+
"eval_rouge1": 40.4943,
|
6 |
+
"eval_rouge2": 16.4695,
|
7 |
+
"eval_rougeL": 28.0964,
|
8 |
+
"eval_rougeLsum": 38.3693,
|
9 |
+
"eval_runtime": 467.1505,
|
10 |
+
"eval_samples": 338,
|
11 |
+
"eval_samples_per_second": 0.724,
|
12 |
+
"eval_steps_per_second": 0.024,
|
13 |
+
"train_loss": 0.16495737830797833,
|
14 |
+
"train_runtime": 12728.0808,
|
15 |
+
"train_samples": 3673,
|
16 |
+
"train_samples_per_second": 4.329,
|
17 |
+
"train_steps_per_second": 0.004
|
18 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 13.18,
|
3 |
+
"eval_gen_len": 246.3491124260355,
|
4 |
+
"eval_loss": 2.2397100925445557,
|
5 |
+
"eval_rouge1": 40.4943,
|
6 |
+
"eval_rouge2": 16.4695,
|
7 |
+
"eval_rougeL": 28.0964,
|
8 |
+
"eval_rougeLsum": 38.3693,
|
9 |
+
"eval_runtime": 467.1505,
|
10 |
+
"eval_samples": 338,
|
11 |
+
"eval_samples_per_second": 0.724,
|
12 |
+
"eval_steps_per_second": 0.024
|
13 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 13.18,
|
3 |
+
"train_loss": 0.16495737830797833,
|
4 |
+
"train_runtime": 12728.0808,
|
5 |
+
"train_samples": 3673,
|
6 |
+
"train_samples_per_second": 4.329,
|
7 |
+
"train_steps_per_second": 0.004
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,303 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 2.237640380859375,
|
3 |
+
"best_model_checkpoint": "/exports/eddie/scratch/s1970716/models/summarization/longt5_xl_summ_screen_bp_only_30/checkpoint-14",
|
4 |
+
"epoch": 13.182608695652174,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 45,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.14,
|
13 |
+
"learning_rate": 0.0005,
|
14 |
+
"loss": 0.3076,
|
15 |
+
"step": 2
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.28,
|
19 |
+
"learning_rate": 0.0005,
|
20 |
+
"loss": 0.3217,
|
21 |
+
"step": 4
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.42,
|
25 |
+
"learning_rate": 0.0005,
|
26 |
+
"loss": 0.3519,
|
27 |
+
"step": 6
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.56,
|
31 |
+
"learning_rate": 0.0005,
|
32 |
+
"loss": 0.3449,
|
33 |
+
"step": 8
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.7,
|
37 |
+
"learning_rate": 0.0005,
|
38 |
+
"loss": 0.3197,
|
39 |
+
"step": 10
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.83,
|
43 |
+
"learning_rate": 0.0005,
|
44 |
+
"loss": 0.3342,
|
45 |
+
"step": 12
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.97,
|
49 |
+
"learning_rate": 0.0005,
|
50 |
+
"loss": 0.324,
|
51 |
+
"step": 14
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.97,
|
55 |
+
"eval_gen_len": 246.7396449704142,
|
56 |
+
"eval_loss": 2.237640380859375,
|
57 |
+
"eval_rouge1": 40.4388,
|
58 |
+
"eval_rouge2": 16.4662,
|
59 |
+
"eval_rougeL": 28.0771,
|
60 |
+
"eval_rougeLsum": 38.3405,
|
61 |
+
"eval_runtime": 1741.7742,
|
62 |
+
"eval_samples_per_second": 0.194,
|
63 |
+
"eval_steps_per_second": 0.025,
|
64 |
+
"step": 14
|
65 |
+
},
|
66 |
+
{
|
67 |
+
"epoch": 4.83,
|
68 |
+
"eval_gen_len": 503.56508875739644,
|
69 |
+
"eval_loss": 2.572709798812866,
|
70 |
+
"eval_rouge1": 30.0123,
|
71 |
+
"eval_rouge2": 12.3701,
|
72 |
+
"eval_rougeL": 21.2834,
|
73 |
+
"eval_rougeLsum": 28.891,
|
74 |
+
"eval_runtime": 472.1582,
|
75 |
+
"eval_samples_per_second": 0.716,
|
76 |
+
"eval_steps_per_second": 0.023,
|
77 |
+
"step": 15
|
78 |
+
},
|
79 |
+
{
|
80 |
+
"epoch": 5.11,
|
81 |
+
"learning_rate": 0.0005,
|
82 |
+
"loss": 0.278,
|
83 |
+
"step": 16
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 5.67,
|
87 |
+
"learning_rate": 0.0005,
|
88 |
+
"loss": 0.3036,
|
89 |
+
"step": 18
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 5.95,
|
93 |
+
"eval_gen_len": 506.9526627218935,
|
94 |
+
"eval_loss": 2.2658941745758057,
|
95 |
+
"eval_rouge1": 27.8421,
|
96 |
+
"eval_rouge2": 11.1942,
|
97 |
+
"eval_rougeL": 20.4713,
|
98 |
+
"eval_rougeLsum": 26.6097,
|
99 |
+
"eval_runtime": 472.9998,
|
100 |
+
"eval_samples_per_second": 0.715,
|
101 |
+
"eval_steps_per_second": 0.023,
|
102 |
+
"step": 19
|
103 |
+
},
|
104 |
+
{
|
105 |
+
"epoch": 6.23,
|
106 |
+
"learning_rate": 0.0005,
|
107 |
+
"loss": 0.2937,
|
108 |
+
"step": 20
|
109 |
+
},
|
110 |
+
{
|
111 |
+
"epoch": 6.78,
|
112 |
+
"learning_rate": 0.0005,
|
113 |
+
"loss": 0.2941,
|
114 |
+
"step": 22
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 6.78,
|
118 |
+
"eval_gen_len": 284.2307692307692,
|
119 |
+
"eval_loss": 2.2636122703552246,
|
120 |
+
"eval_rouge1": 40.8304,
|
121 |
+
"eval_rouge2": 17.3615,
|
122 |
+
"eval_rougeL": 28.0971,
|
123 |
+
"eval_rougeLsum": 39.0943,
|
124 |
+
"eval_runtime": 468.1667,
|
125 |
+
"eval_samples_per_second": 0.722,
|
126 |
+
"eval_steps_per_second": 0.023,
|
127 |
+
"step": 22
|
128 |
+
},
|
129 |
+
{
|
130 |
+
"epoch": 7.34,
|
131 |
+
"learning_rate": 0.0005,
|
132 |
+
"loss": 0.2508,
|
133 |
+
"step": 24
|
134 |
+
},
|
135 |
+
{
|
136 |
+
"epoch": 7.9,
|
137 |
+
"learning_rate": 0.0005,
|
138 |
+
"loss": 0.2642,
|
139 |
+
"step": 26
|
140 |
+
},
|
141 |
+
{
|
142 |
+
"epoch": 7.9,
|
143 |
+
"eval_gen_len": 341.25147928994085,
|
144 |
+
"eval_loss": 2.286357879638672,
|
145 |
+
"eval_rouge1": 38.3377,
|
146 |
+
"eval_rouge2": 15.8119,
|
147 |
+
"eval_rougeL": 26.4838,
|
148 |
+
"eval_rougeLsum": 36.5174,
|
149 |
+
"eval_runtime": 469.1516,
|
150 |
+
"eval_samples_per_second": 0.72,
|
151 |
+
"eval_steps_per_second": 0.023,
|
152 |
+
"step": 26
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 8.45,
|
156 |
+
"learning_rate": 0.0005,
|
157 |
+
"loss": 0.2604,
|
158 |
+
"step": 28
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 8.73,
|
162 |
+
"eval_gen_len": 435.26331360946745,
|
163 |
+
"eval_loss": 2.455064535140991,
|
164 |
+
"eval_rouge1": 33.2021,
|
165 |
+
"eval_rouge2": 13.6577,
|
166 |
+
"eval_rougeL": 23.3288,
|
167 |
+
"eval_rougeLsum": 31.8326,
|
168 |
+
"eval_runtime": 471.959,
|
169 |
+
"eval_samples_per_second": 0.716,
|
170 |
+
"eval_steps_per_second": 0.023,
|
171 |
+
"step": 29
|
172 |
+
},
|
173 |
+
{
|
174 |
+
"epoch": 9.01,
|
175 |
+
"learning_rate": 0.0005,
|
176 |
+
"loss": 0.2422,
|
177 |
+
"step": 30
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 9.57,
|
181 |
+
"learning_rate": 0.0005,
|
182 |
+
"loss": 0.2237,
|
183 |
+
"step": 32
|
184 |
+
},
|
185 |
+
{
|
186 |
+
"epoch": 9.84,
|
187 |
+
"eval_gen_len": 234.61242603550295,
|
188 |
+
"eval_loss": 2.615253210067749,
|
189 |
+
"eval_rouge1": 40.3297,
|
190 |
+
"eval_rouge2": 15.3786,
|
191 |
+
"eval_rougeL": 28.1208,
|
192 |
+
"eval_rougeLsum": 38.2426,
|
193 |
+
"eval_runtime": 470.1581,
|
194 |
+
"eval_samples_per_second": 0.719,
|
195 |
+
"eval_steps_per_second": 0.023,
|
196 |
+
"step": 33
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 10.12,
|
200 |
+
"learning_rate": 0.0005,
|
201 |
+
"loss": 0.2184,
|
202 |
+
"step": 34
|
203 |
+
},
|
204 |
+
{
|
205 |
+
"epoch": 10.68,
|
206 |
+
"learning_rate": 0.0005,
|
207 |
+
"loss": 0.1904,
|
208 |
+
"step": 36
|
209 |
+
},
|
210 |
+
{
|
211 |
+
"epoch": 10.96,
|
212 |
+
"eval_gen_len": 174.57396449704143,
|
213 |
+
"eval_loss": 2.666517734527588,
|
214 |
+
"eval_rouge1": 39.6006,
|
215 |
+
"eval_rouge2": 14.9586,
|
216 |
+
"eval_rougeL": 27.2453,
|
217 |
+
"eval_rougeLsum": 37.6744,
|
218 |
+
"eval_runtime": 450.9467,
|
219 |
+
"eval_samples_per_second": 0.75,
|
220 |
+
"eval_steps_per_second": 0.024,
|
221 |
+
"step": 37
|
222 |
+
},
|
223 |
+
{
|
224 |
+
"epoch": 11.23,
|
225 |
+
"learning_rate": 0.0005,
|
226 |
+
"loss": 0.2076,
|
227 |
+
"step": 38
|
228 |
+
},
|
229 |
+
{
|
230 |
+
"epoch": 11.79,
|
231 |
+
"learning_rate": 0.0005,
|
232 |
+
"loss": 0.2247,
|
233 |
+
"step": 40
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 11.79,
|
237 |
+
"eval_gen_len": 500.58284023668637,
|
238 |
+
"eval_loss": 2.722362518310547,
|
239 |
+
"eval_rouge1": 30.5957,
|
240 |
+
"eval_rouge2": 13.3496,
|
241 |
+
"eval_rougeL": 21.9712,
|
242 |
+
"eval_rougeLsum": 29.22,
|
243 |
+
"eval_runtime": 477.6586,
|
244 |
+
"eval_samples_per_second": 0.708,
|
245 |
+
"eval_steps_per_second": 0.023,
|
246 |
+
"step": 40
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 12.35,
|
250 |
+
"learning_rate": 0.0005,
|
251 |
+
"loss": 0.1866,
|
252 |
+
"step": 42
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"epoch": 12.9,
|
256 |
+
"learning_rate": 0.0005,
|
257 |
+
"loss": 0.182,
|
258 |
+
"step": 44
|
259 |
+
},
|
260 |
+
{
|
261 |
+
"epoch": 12.9,
|
262 |
+
"eval_gen_len": 259.6568047337278,
|
263 |
+
"eval_loss": 3.271503448486328,
|
264 |
+
"eval_rouge1": 41.6828,
|
265 |
+
"eval_rouge2": 17.0818,
|
266 |
+
"eval_rougeL": 28.087,
|
267 |
+
"eval_rougeLsum": 39.5947,
|
268 |
+
"eval_runtime": 468.2637,
|
269 |
+
"eval_samples_per_second": 0.722,
|
270 |
+
"eval_steps_per_second": 0.023,
|
271 |
+
"step": 44
|
272 |
+
},
|
273 |
+
{
|
274 |
+
"epoch": 13.18,
|
275 |
+
"eval_gen_len": 484.396449704142,
|
276 |
+
"eval_loss": 2.397312641143799,
|
277 |
+
"eval_rouge1": 31.9833,
|
278 |
+
"eval_rouge2": 14.0141,
|
279 |
+
"eval_rougeL": 22.6823,
|
280 |
+
"eval_rougeLsum": 30.6424,
|
281 |
+
"eval_runtime": 471.2439,
|
282 |
+
"eval_samples_per_second": 0.717,
|
283 |
+
"eval_steps_per_second": 0.023,
|
284 |
+
"step": 45
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 13.18,
|
288 |
+
"step": 45,
|
289 |
+
"total_flos": 2.4232217682745754e+18,
|
290 |
+
"train_loss": 0.16495737830797833,
|
291 |
+
"train_runtime": 12728.0808,
|
292 |
+
"train_samples_per_second": 4.329,
|
293 |
+
"train_steps_per_second": 0.004
|
294 |
+
}
|
295 |
+
],
|
296 |
+
"logging_steps": 2,
|
297 |
+
"max_steps": 45,
|
298 |
+
"num_train_epochs": 15,
|
299 |
+
"save_steps": 500,
|
300 |
+
"total_flos": 2.4232217682745754e+18,
|
301 |
+
"trial_name": null,
|
302 |
+
"trial_params": null
|
303 |
+
}
|