Commit
•
a01a840
1
Parent(s):
6015cfb
Training in progress, step 200
Browse files- all_results.json +13 -13
- eval_results.json +5 -5
- pytorch_model.bin +2 -2
- runs/Feb01_00-35-21_node1/1643647072.4568229/events.out.tfevents.1643647072.node1 +3 -0
- runs/Feb01_00-35-21_node1/events.out.tfevents.1643647072.node1 +3 -0
- runs/Jan31_20-22-53_node1/events.out.tfevents.1643631902.node1 +2 -2
- runs/Jan31_20-22-53_node1/events.out.tfevents.1643632326.node1 +3 -0
- test_results.json +5 -5
- train_results.json +3 -3
- trainer_state.json +59 -167
- training_args.bin +1 -1
- training_config.json +1 -1
all_results.json
CHANGED
@@ -1,18 +1,18 @@
|
|
1 |
{
|
2 |
"epoch": 20.0,
|
3 |
-
"eval_average_metrics": 52.
|
4 |
-
"eval_loss": 0.
|
5 |
-
"eval_matthews_correlation": 52.
|
6 |
-
"eval_runtime":
|
7 |
-
"eval_samples_per_second":
|
8 |
-
"test_average_metrics": 63.
|
9 |
-
"test_loss": 0.
|
10 |
-
"test_matthews_correlation": 63.
|
11 |
-
"test_runtime":
|
12 |
-
"test_samples_per_second":
|
13 |
"train_loss": 0.5282898814179177,
|
14 |
-
"train_runtime":
|
15 |
"train_samples": 8551,
|
16 |
-
"train_samples_per_second":
|
17 |
-
"train_steps_per_second": 4.
|
18 |
}
|
|
|
1 |
{
|
2 |
"epoch": 20.0,
|
3 |
+
"eval_average_metrics": 52.55846445665645,
|
4 |
+
"eval_loss": 0.22525694966316223,
|
5 |
+
"eval_matthews_correlation": 52.55846445665645,
|
6 |
+
"eval_runtime": 1.2083,
|
7 |
+
"eval_samples_per_second": 431.173,
|
8 |
+
"test_average_metrics": 63.756766360647745,
|
9 |
+
"test_loss": 0.18106061220169067,
|
10 |
+
"test_matthews_correlation": 63.756766360647745,
|
11 |
+
"test_runtime": 1.2266,
|
12 |
+
"test_samples_per_second": 425.574,
|
13 |
"train_loss": 0.5282898814179177,
|
14 |
+
"train_runtime": 422.8779,
|
15 |
"train_samples": 8551,
|
16 |
+
"train_samples_per_second": 404.419,
|
17 |
+
"train_steps_per_second": 4.067
|
18 |
}
|
eval_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 20.0,
|
3 |
-
"eval_average_metrics": 52.
|
4 |
-
"eval_loss": 0.
|
5 |
-
"eval_matthews_correlation": 52.
|
6 |
-
"eval_runtime":
|
7 |
-
"eval_samples_per_second":
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 20.0,
|
3 |
+
"eval_average_metrics": 52.55846445665645,
|
4 |
+
"eval_loss": 0.22525694966316223,
|
5 |
+
"eval_matthews_correlation": 52.55846445665645,
|
6 |
+
"eval_runtime": 1.2083,
|
7 |
+
"eval_samples_per_second": 431.173
|
8 |
}
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e3bbc6bb60a0e9a5a611381f99725d8ff5f079fda737349a7ec175298f1b464e
|
3 |
+
size 7551621
|
runs/Feb01_00-35-21_node1/1643647072.4568229/events.out.tfevents.1643647072.node1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2eec98c8506b77fac2f6f93648af3514b1d856bc587d91098be10bbd438852d8
|
3 |
+
size 5011
|
runs/Feb01_00-35-21_node1/events.out.tfevents.1643647072.node1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4fb0a5519639038b689386956ca41fc8f5f1bec052201aaec72142a139e237ed
|
3 |
+
size 4304
|
runs/Jan31_20-22-53_node1/events.out.tfevents.1643631902.node1
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e7b0be464bdd99e7e876ca5647ad2c7836c421cbd20a648d6a2648b1b17b8627
|
3 |
+
size 7801
|
runs/Jan31_20-22-53_node1/events.out.tfevents.1643632326.node1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:67469b42186d9407b3095485ef49514c1f0f317abf04186e198a20bb130ee140
|
3 |
+
size 708
|
test_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 20.0,
|
3 |
-
"test_average_metrics": 63.
|
4 |
-
"test_loss": 0.
|
5 |
-
"test_matthews_correlation": 63.
|
6 |
-
"test_runtime":
|
7 |
-
"test_samples_per_second":
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 20.0,
|
3 |
+
"test_average_metrics": 63.756766360647745,
|
4 |
+
"test_loss": 0.18106061220169067,
|
5 |
+
"test_matthews_correlation": 63.756766360647745,
|
6 |
+
"test_runtime": 1.2266,
|
7 |
+
"test_samples_per_second": 425.574
|
8 |
}
|
train_results.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
"epoch": 20.0,
|
3 |
"train_loss": 0.5282898814179177,
|
4 |
-
"train_runtime":
|
5 |
"train_samples": 8551,
|
6 |
-
"train_samples_per_second":
|
7 |
-
"train_steps_per_second": 4.
|
8 |
}
|
|
|
1 |
{
|
2 |
"epoch": 20.0,
|
3 |
"train_loss": 0.5282898814179177,
|
4 |
+
"train_runtime": 422.8779,
|
5 |
"train_samples": 8551,
|
6 |
+
"train_samples_per_second": 404.419,
|
7 |
+
"train_steps_per_second": 4.067
|
8 |
}
|
trainer_state.json
CHANGED
@@ -1,6 +1,6 @@
|
|
1 |
{
|
2 |
-
"best_metric": 52.
|
3 |
-
"best_model_checkpoint": "outputs/bitfit/t5-base/cola/checkpoint-
|
4 |
"epoch": 20.0,
|
5 |
"global_step": 1720,
|
6 |
"is_hyper_param_search": false,
|
@@ -8,49 +8,22 @@
|
|
8 |
"is_world_process_zero": true,
|
9 |
"log_history": [
|
10 |
{
|
11 |
-
"epoch":
|
12 |
-
"eval_average_metrics":
|
13 |
-
"eval_loss":
|
14 |
-
"eval_matthews_correlation":
|
15 |
-
"eval_runtime": 0.
|
16 |
-
"eval_samples_per_second":
|
17 |
-
"step":
|
18 |
},
|
19 |
{
|
20 |
-
"epoch":
|
21 |
-
"eval_average_metrics":
|
22 |
-
"eval_loss": 0.
|
23 |
-
"eval_matthews_correlation":
|
24 |
-
"eval_runtime": 0.
|
25 |
-
"eval_samples_per_second":
|
26 |
-
"step":
|
27 |
-
},
|
28 |
-
{
|
29 |
-
"epoch": 3.0,
|
30 |
-
"eval_average_metrics": 43.97683870357735,
|
31 |
-
"eval_loss": 0.21786993741989136,
|
32 |
-
"eval_matthews_correlation": 43.97683870357735,
|
33 |
-
"eval_runtime": 0.8315,
|
34 |
-
"eval_samples_per_second": 626.599,
|
35 |
-
"step": 258
|
36 |
-
},
|
37 |
-
{
|
38 |
-
"epoch": 4.0,
|
39 |
-
"eval_average_metrics": 46.78131759250157,
|
40 |
-
"eval_loss": 0.22430144250392914,
|
41 |
-
"eval_matthews_correlation": 46.78131759250157,
|
42 |
-
"eval_runtime": 0.8615,
|
43 |
-
"eval_samples_per_second": 604.782,
|
44 |
-
"step": 344
|
45 |
-
},
|
46 |
-
{
|
47 |
-
"epoch": 5.0,
|
48 |
-
"eval_average_metrics": 49.53306157776009,
|
49 |
-
"eval_loss": 0.21720413863658905,
|
50 |
-
"eval_matthews_correlation": 49.53306157776009,
|
51 |
-
"eval_runtime": 0.8515,
|
52 |
-
"eval_samples_per_second": 611.84,
|
53 |
-
"step": 430
|
54 |
},
|
55 |
{
|
56 |
"epoch": 5.81,
|
@@ -59,58 +32,22 @@
|
|
59 |
"step": 500
|
60 |
},
|
61 |
{
|
62 |
-
"epoch": 6.
|
63 |
-
"eval_average_metrics":
|
64 |
-
"eval_loss": 0.
|
65 |
-
"eval_matthews_correlation":
|
66 |
-
"eval_runtime": 0.
|
67 |
-
"eval_samples_per_second":
|
68 |
-
"step":
|
69 |
-
},
|
70 |
-
{
|
71 |
-
"epoch": 7.0,
|
72 |
-
"eval_average_metrics": 47.85938832793908,
|
73 |
-
"eval_loss": 0.23619267344474792,
|
74 |
-
"eval_matthews_correlation": 47.85938832793908,
|
75 |
-
"eval_runtime": 0.8864,
|
76 |
-
"eval_samples_per_second": 587.749,
|
77 |
-
"step": 602
|
78 |
-
},
|
79 |
-
{
|
80 |
-
"epoch": 8.0,
|
81 |
-
"eval_average_metrics": 49.724093419139464,
|
82 |
-
"eval_loss": 0.2390584498643875,
|
83 |
-
"eval_matthews_correlation": 49.724093419139464,
|
84 |
-
"eval_runtime": 0.8668,
|
85 |
-
"eval_samples_per_second": 601.078,
|
86 |
-
"step": 688
|
87 |
-
},
|
88 |
-
{
|
89 |
-
"epoch": 9.0,
|
90 |
-
"eval_average_metrics": 48.86384938654592,
|
91 |
-
"eval_loss": 0.22121600806713104,
|
92 |
-
"eval_matthews_correlation": 48.86384938654592,
|
93 |
-
"eval_runtime": 0.8908,
|
94 |
-
"eval_samples_per_second": 584.879,
|
95 |
-
"step": 774
|
96 |
-
},
|
97 |
-
{
|
98 |
-
"epoch": 10.0,
|
99 |
-
"eval_average_metrics": 48.08237616875362,
|
100 |
-
"eval_loss": 0.2251322716474533,
|
101 |
-
"eval_matthews_correlation": 48.08237616875362,
|
102 |
-
"eval_runtime": 0.8537,
|
103 |
-
"eval_samples_per_second": 610.303,
|
104 |
-
"step": 860
|
105 |
},
|
106 |
{
|
107 |
-
"epoch":
|
108 |
-
"eval_average_metrics":
|
109 |
-
"eval_loss": 0.
|
110 |
-
"eval_matthews_correlation":
|
111 |
-
"eval_runtime":
|
112 |
-
"eval_samples_per_second":
|
113 |
-
"step":
|
114 |
},
|
115 |
{
|
116 |
"epoch": 11.63,
|
@@ -119,58 +56,31 @@
|
|
119 |
"step": 1000
|
120 |
},
|
121 |
{
|
122 |
-
"epoch":
|
123 |
-
"eval_average_metrics":
|
124 |
-
"eval_loss": 0.
|
125 |
-
"eval_matthews_correlation":
|
126 |
-
"eval_runtime":
|
127 |
-
"eval_samples_per_second":
|
128 |
-
"step":
|
129 |
-
},
|
130 |
-
{
|
131 |
-
"epoch": 13.0,
|
132 |
-
"eval_average_metrics": 50.45736339822007,
|
133 |
-
"eval_loss": 0.22725746035575867,
|
134 |
-
"eval_matthews_correlation": 50.45736339822007,
|
135 |
-
"eval_runtime": 0.848,
|
136 |
-
"eval_samples_per_second": 614.383,
|
137 |
-
"step": 1118
|
138 |
},
|
139 |
{
|
140 |
-
"epoch":
|
141 |
"eval_average_metrics": 51.11915736464516,
|
142 |
-
"eval_loss": 0.
|
143 |
"eval_matthews_correlation": 51.11915736464516,
|
144 |
-
"eval_runtime":
|
145 |
-
"eval_samples_per_second":
|
146 |
-
"step":
|
147 |
},
|
148 |
{
|
149 |
-
"epoch":
|
150 |
-
"eval_average_metrics":
|
151 |
-
"eval_loss": 0.
|
152 |
-
"eval_matthews_correlation":
|
153 |
-
"eval_runtime":
|
154 |
-
"eval_samples_per_second":
|
155 |
-
"step":
|
156 |
-
},
|
157 |
-
{
|
158 |
-
"epoch": 16.0,
|
159 |
-
"eval_average_metrics": 50.78390137665081,
|
160 |
-
"eval_loss": 0.24213995039463043,
|
161 |
-
"eval_matthews_correlation": 50.78390137665081,
|
162 |
-
"eval_runtime": 0.7976,
|
163 |
-
"eval_samples_per_second": 653.201,
|
164 |
-
"step": 1376
|
165 |
-
},
|
166 |
-
{
|
167 |
-
"epoch": 17.0,
|
168 |
-
"eval_average_metrics": 51.452415683714314,
|
169 |
-
"eval_loss": 0.22793160378932953,
|
170 |
-
"eval_matthews_correlation": 51.452415683714314,
|
171 |
-
"eval_runtime": 0.8964,
|
172 |
-
"eval_samples_per_second": 581.215,
|
173 |
-
"step": 1462
|
174 |
},
|
175 |
{
|
176 |
"epoch": 17.44,
|
@@ -179,40 +89,22 @@
|
|
179 |
"step": 1500
|
180 |
},
|
181 |
{
|
182 |
-
"epoch": 18.
|
183 |
-
"eval_average_metrics":
|
184 |
-
"eval_loss": 0.
|
185 |
-
"eval_matthews_correlation":
|
186 |
-
"eval_runtime":
|
187 |
-
"eval_samples_per_second":
|
188 |
-
"step":
|
189 |
-
},
|
190 |
-
{
|
191 |
-
"epoch": 19.0,
|
192 |
-
"eval_average_metrics": 50.9850664467699,
|
193 |
-
"eval_loss": 0.2308470457792282,
|
194 |
-
"eval_matthews_correlation": 50.9850664467699,
|
195 |
-
"eval_runtime": 0.8691,
|
196 |
-
"eval_samples_per_second": 599.475,
|
197 |
-
"step": 1634
|
198 |
-
},
|
199 |
-
{
|
200 |
-
"epoch": 20.0,
|
201 |
-
"eval_average_metrics": 50.9850664467699,
|
202 |
-
"eval_loss": 0.22891275584697723,
|
203 |
-
"eval_matthews_correlation": 50.9850664467699,
|
204 |
-
"eval_runtime": 0.8366,
|
205 |
-
"eval_samples_per_second": 622.759,
|
206 |
-
"step": 1720
|
207 |
},
|
208 |
{
|
209 |
"epoch": 20.0,
|
210 |
"step": 1720,
|
211 |
"total_flos": 6646762904679216.0,
|
212 |
"train_loss": 0.5282898814179177,
|
213 |
-
"train_runtime":
|
214 |
-
"train_samples_per_second":
|
215 |
-
"train_steps_per_second": 4.
|
216 |
}
|
217 |
],
|
218 |
"max_steps": 1720,
|
|
|
1 |
{
|
2 |
+
"best_metric": 52.55846445665645,
|
3 |
+
"best_model_checkpoint": "outputs/bitfit/t5-base/cola/checkpoint-800",
|
4 |
"epoch": 20.0,
|
5 |
"global_step": 1720,
|
6 |
"is_hyper_param_search": false,
|
|
|
8 |
"is_world_process_zero": true,
|
9 |
"log_history": [
|
10 |
{
|
11 |
+
"epoch": 2.33,
|
12 |
+
"eval_average_metrics": 21.053609210794303,
|
13 |
+
"eval_loss": 0.2652588486671448,
|
14 |
+
"eval_matthews_correlation": 21.053609210794303,
|
15 |
+
"eval_runtime": 0.8287,
|
16 |
+
"eval_samples_per_second": 628.717,
|
17 |
+
"step": 200
|
18 |
},
|
19 |
{
|
20 |
+
"epoch": 4.65,
|
21 |
+
"eval_average_metrics": 44.5285953164197,
|
22 |
+
"eval_loss": 0.23253390192985535,
|
23 |
+
"eval_matthews_correlation": 44.5285953164197,
|
24 |
+
"eval_runtime": 0.8258,
|
25 |
+
"eval_samples_per_second": 630.924,
|
26 |
+
"step": 400
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
27 |
},
|
28 |
{
|
29 |
"epoch": 5.81,
|
|
|
32 |
"step": 500
|
33 |
},
|
34 |
{
|
35 |
+
"epoch": 6.98,
|
36 |
+
"eval_average_metrics": 47.511676688521085,
|
37 |
+
"eval_loss": 0.25159838795661926,
|
38 |
+
"eval_matthews_correlation": 47.511676688521085,
|
39 |
+
"eval_runtime": 0.8388,
|
40 |
+
"eval_samples_per_second": 621.09,
|
41 |
+
"step": 600
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
42 |
},
|
43 |
{
|
44 |
+
"epoch": 9.3,
|
45 |
+
"eval_average_metrics": 52.55846445665645,
|
46 |
+
"eval_loss": 0.22525694966316223,
|
47 |
+
"eval_matthews_correlation": 52.55846445665645,
|
48 |
+
"eval_runtime": 1.5205,
|
49 |
+
"eval_samples_per_second": 342.658,
|
50 |
+
"step": 800
|
51 |
},
|
52 |
{
|
53 |
"epoch": 11.63,
|
|
|
56 |
"step": 1000
|
57 |
},
|
58 |
{
|
59 |
+
"epoch": 11.63,
|
60 |
+
"eval_average_metrics": 50.296653045050874,
|
61 |
+
"eval_loss": 0.23001490533351898,
|
62 |
+
"eval_matthews_correlation": 50.296653045050874,
|
63 |
+
"eval_runtime": 1.2989,
|
64 |
+
"eval_samples_per_second": 401.107,
|
65 |
+
"step": 1000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
66 |
},
|
67 |
{
|
68 |
+
"epoch": 13.95,
|
69 |
"eval_average_metrics": 51.11915736464516,
|
70 |
+
"eval_loss": 0.22642329335212708,
|
71 |
"eval_matthews_correlation": 51.11915736464516,
|
72 |
+
"eval_runtime": 1.2231,
|
73 |
+
"eval_samples_per_second": 425.968,
|
74 |
+
"step": 1200
|
75 |
},
|
76 |
{
|
77 |
+
"epoch": 16.28,
|
78 |
+
"eval_average_metrics": 50.87277150375961,
|
79 |
+
"eval_loss": 0.22626039385795593,
|
80 |
+
"eval_matthews_correlation": 50.87277150375961,
|
81 |
+
"eval_runtime": 1.1916,
|
82 |
+
"eval_samples_per_second": 437.245,
|
83 |
+
"step": 1400
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
84 |
},
|
85 |
{
|
86 |
"epoch": 17.44,
|
|
|
89 |
"step": 1500
|
90 |
},
|
91 |
{
|
92 |
+
"epoch": 18.6,
|
93 |
+
"eval_average_metrics": 52.035569173764515,
|
94 |
+
"eval_loss": 0.22766831517219543,
|
95 |
+
"eval_matthews_correlation": 52.035569173764515,
|
96 |
+
"eval_runtime": 1.2214,
|
97 |
+
"eval_samples_per_second": 426.575,
|
98 |
+
"step": 1600
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
99 |
},
|
100 |
{
|
101 |
"epoch": 20.0,
|
102 |
"step": 1720,
|
103 |
"total_flos": 6646762904679216.0,
|
104 |
"train_loss": 0.5282898814179177,
|
105 |
+
"train_runtime": 422.8779,
|
106 |
+
"train_samples_per_second": 404.419,
|
107 |
+
"train_steps_per_second": 4.067
|
108 |
}
|
109 |
],
|
110 |
"max_steps": 1720,
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 3183
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d45981c3869ac88ceb617a2173cfab542cb6a12ef612e4821407250b68788864
|
3 |
size 3183
|
training_config.json
CHANGED
@@ -1 +1 @@
|
|
1 |
-
{"dataset_config_name": ["en"], "delta_type": "
|
|
|
1 |
+
{"bottleneck_dim": 24, "dataset_config_name": ["en"], "delta_type": "adapter", "do_eval": true, "do_test": true, "do_train": true, "eval_dataset_config_name": ["en"], "eval_dataset_name": "cola", "eval_steps": 200, "evaluation_strategy": "steps", "greater_is_better": true, "learning_rate": 0.0003, "load_best_model_at_end": true, "max_source_length": 128, "metric_for_best_model": "average_metrics", "model_name_or_path": "../../../../plm_cache/t5-base", "num_train_epochs": 20, "output_dir": "outputs/bitfit/t5-base/cola", "overwrite_output_dir": true, "per_device_eval_batch_size": 100, "per_device_train_batch_size": 100, "predict_with_generate": true, "push_to_hub": true, "save_steps": 200, "save_strategy": "steps", "save_total_limit": 1, "seed": 42, "split_validation_test": true, "task_name": "cola", "test_dataset_config_name": ["en"], "test_dataset_name": "cola", "tokenizer_name": "../../../../plm_cache/t5-base", "unfrozen_modules": ["deltas", "layer_norm", "final_layer_norm"], "warmup_steps": 500}
|