DewiBrynJones
commited on
Commit
•
c024852
1
Parent(s):
de9bbcb
End of training
Browse files- README.md +3 -1
- all_results.json +9 -9
- eval_results.json +5 -5
- runs/May16_08-14-26_2e698a9c250f/events.out.tfevents.1715850411.2e698a9c250f.233.1 +3 -0
- train_results.json +4 -4
- trainer_state.json +142 -142
README.md
CHANGED
@@ -2,6 +2,8 @@
|
|
2 |
license: apache-2.0
|
3 |
base_model: facebook/wav2vec2-large-xlsr-53
|
4 |
tags:
|
|
|
|
|
5 |
- generated_from_trainer
|
6 |
metrics:
|
7 |
- wer
|
@@ -15,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
|
16 |
# wav2vec2-xlsr-53-ft-btb-cy
|
17 |
|
18 |
-
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on
|
19 |
It achieves the following results on the evaluation set:
|
20 |
- Loss: 0.4589
|
21 |
- Wer: 0.3743
|
|
|
2 |
license: apache-2.0
|
3 |
base_model: facebook/wav2vec2-large-xlsr-53
|
4 |
tags:
|
5 |
+
- automatic-speech-recognition
|
6 |
+
- DewiBrynJones/banc-trawsgrifiadau-bangor-normalized
|
7 |
- generated_from_trainer
|
8 |
metrics:
|
9 |
- wer
|
|
|
17 |
|
18 |
# wav2vec2-xlsr-53-ft-btb-cy
|
19 |
|
20 |
+
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the DEWIBRYNJONES/BANC-TRAWSGRIFIADAU-BANGOR-NORMALIZED - DEFAULT dataset.
|
21 |
It achieves the following results on the evaluation set:
|
22 |
- Loss: 0.4589
|
23 |
- Wer: 0.3743
|
all_results.json
CHANGED
@@ -1,15 +1,15 @@
|
|
1 |
{
|
2 |
"epoch": 3.536067892503536,
|
3 |
-
"eval_loss": 0.
|
4 |
-
"eval_runtime":
|
5 |
"eval_samples": 5656,
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second": 4.
|
8 |
-
"eval_wer": 0.
|
9 |
"total_flos": 9.55169606524761e+18,
|
10 |
-
"train_loss": 1.
|
11 |
-
"train_runtime":
|
12 |
"train_samples": 22621,
|
13 |
-
"train_samples_per_second":
|
14 |
-
"train_steps_per_second": 0.
|
15 |
}
|
|
|
1 |
{
|
2 |
"epoch": 3.536067892503536,
|
3 |
+
"eval_loss": 0.4588949382305145,
|
4 |
+
"eval_runtime": 153.3999,
|
5 |
"eval_samples": 5656,
|
6 |
+
"eval_samples_per_second": 36.871,
|
7 |
+
"eval_steps_per_second": 4.609,
|
8 |
+
"eval_wer": 0.3742697566925595,
|
9 |
"total_flos": 9.55169606524761e+18,
|
10 |
+
"train_loss": 1.2622116821289062,
|
11 |
+
"train_runtime": 6522.8506,
|
12 |
"train_samples": 22621,
|
13 |
+
"train_samples_per_second": 12.265,
|
14 |
+
"train_steps_per_second": 0.383
|
15 |
}
|
eval_results.json
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
{
|
2 |
"epoch": 3.536067892503536,
|
3 |
-
"eval_loss": 0.
|
4 |
-
"eval_runtime":
|
5 |
"eval_samples": 5656,
|
6 |
-
"eval_samples_per_second":
|
7 |
-
"eval_steps_per_second": 4.
|
8 |
-
"eval_wer": 0.
|
9 |
}
|
|
|
1 |
{
|
2 |
"epoch": 3.536067892503536,
|
3 |
+
"eval_loss": 0.4588949382305145,
|
4 |
+
"eval_runtime": 153.3999,
|
5 |
"eval_samples": 5656,
|
6 |
+
"eval_samples_per_second": 36.871,
|
7 |
+
"eval_steps_per_second": 4.609,
|
8 |
+
"eval_wer": 0.3742697566925595
|
9 |
}
|
runs/May16_08-14-26_2e698a9c250f/events.out.tfevents.1715850411.2e698a9c250f.233.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f32c40b2b9131b58fbffd90fb39b2a35a1ffd6bb07bca4c083c735df411d2c55
|
3 |
+
size 406
|
train_results.json
CHANGED
@@ -1,9 +1,9 @@
|
|
1 |
{
|
2 |
"epoch": 3.536067892503536,
|
3 |
"total_flos": 9.55169606524761e+18,
|
4 |
-
"train_loss": 1.
|
5 |
-
"train_runtime":
|
6 |
"train_samples": 22621,
|
7 |
-
"train_samples_per_second":
|
8 |
-
"train_steps_per_second": 0.
|
9 |
}
|
|
|
1 |
{
|
2 |
"epoch": 3.536067892503536,
|
3 |
"total_flos": 9.55169606524761e+18,
|
4 |
+
"train_loss": 1.2622116821289062,
|
5 |
+
"train_runtime": 6522.8506,
|
6 |
"train_samples": 22621,
|
7 |
+
"train_samples_per_second": 12.265,
|
8 |
+
"train_steps_per_second": 0.383
|
9 |
}
|
trainer_state.json
CHANGED
@@ -10,272 +10,272 @@
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.14144271570014144,
|
13 |
-
"eval_loss": 4.
|
14 |
-
"eval_runtime":
|
15 |
-
"eval_samples_per_second":
|
16 |
-
"eval_steps_per_second": 4.
|
17 |
"eval_wer": 1.0,
|
18 |
"step": 100
|
19 |
},
|
20 |
{
|
21 |
"epoch": 0.2828854314002829,
|
22 |
-
"eval_loss": 3.
|
23 |
-
"eval_runtime":
|
24 |
-
"eval_samples_per_second":
|
25 |
-
"eval_steps_per_second": 4.
|
26 |
"eval_wer": 1.0,
|
27 |
"step": 200
|
28 |
},
|
29 |
{
|
30 |
"epoch": 0.4243281471004243,
|
31 |
-
"eval_loss":
|
32 |
-
"eval_runtime":
|
33 |
-
"eval_samples_per_second":
|
34 |
-
"eval_steps_per_second": 4.
|
35 |
-
"eval_wer": 0
|
36 |
"step": 300
|
37 |
},
|
38 |
{
|
39 |
"epoch": 0.5657708628005658,
|
40 |
-
"eval_loss": 1.
|
41 |
-
"eval_runtime":
|
42 |
-
"eval_samples_per_second":
|
43 |
-
"eval_steps_per_second": 4.
|
44 |
-
"eval_wer": 0.
|
45 |
"step": 400
|
46 |
},
|
47 |
{
|
48 |
"epoch": 0.7072135785007072,
|
49 |
-
"grad_norm":
|
50 |
-
"learning_rate": 0.
|
51 |
-
"loss": 3.
|
52 |
"step": 500
|
53 |
},
|
54 |
{
|
55 |
"epoch": 0.7072135785007072,
|
56 |
-
"eval_loss": 1.
|
57 |
-
"eval_runtime":
|
58 |
-
"eval_samples_per_second":
|
59 |
-
"eval_steps_per_second": 4.
|
60 |
-
"eval_wer": 0.
|
61 |
"step": 500
|
62 |
},
|
63 |
{
|
64 |
"epoch": 0.8486562942008486,
|
65 |
-
"eval_loss": 0.
|
66 |
-
"eval_runtime":
|
67 |
-
"eval_samples_per_second":
|
68 |
-
"eval_steps_per_second": 4.
|
69 |
-
"eval_wer": 0.
|
70 |
"step": 600
|
71 |
},
|
72 |
{
|
73 |
"epoch": 0.9900990099009901,
|
74 |
-
"eval_loss": 0.
|
75 |
-
"eval_runtime":
|
76 |
-
"eval_samples_per_second":
|
77 |
-
"eval_steps_per_second": 4.
|
78 |
-
"eval_wer": 0.
|
79 |
"step": 700
|
80 |
},
|
81 |
{
|
82 |
"epoch": 1.1315417256011315,
|
83 |
-
"eval_loss": 0.
|
84 |
-
"eval_runtime":
|
85 |
-
"eval_samples_per_second":
|
86 |
-
"eval_steps_per_second": 4.
|
87 |
-
"eval_wer": 0.
|
88 |
"step": 800
|
89 |
},
|
90 |
{
|
91 |
"epoch": 1.272984441301273,
|
92 |
-
"eval_loss": 0.
|
93 |
-
"eval_runtime":
|
94 |
-
"eval_samples_per_second":
|
95 |
-
"eval_steps_per_second": 4.
|
96 |
-
"eval_wer": 0.
|
97 |
"step": 900
|
98 |
},
|
99 |
{
|
100 |
"epoch": 1.4144271570014144,
|
101 |
-
"grad_norm": 0.
|
102 |
-
"learning_rate": 0.
|
103 |
-
"loss": 0.
|
104 |
"step": 1000
|
105 |
},
|
106 |
{
|
107 |
"epoch": 1.4144271570014144,
|
108 |
-
"eval_loss": 0.
|
109 |
-
"eval_runtime":
|
110 |
-
"eval_samples_per_second":
|
111 |
-
"eval_steps_per_second": 4.
|
112 |
-
"eval_wer": 0.
|
113 |
"step": 1000
|
114 |
},
|
115 |
{
|
116 |
"epoch": 1.5558698727015559,
|
117 |
-
"eval_loss": 0.
|
118 |
-
"eval_runtime":
|
119 |
-
"eval_samples_per_second":
|
120 |
-
"eval_steps_per_second": 4.
|
121 |
-
"eval_wer": 0.
|
122 |
"step": 1100
|
123 |
},
|
124 |
{
|
125 |
"epoch": 1.6973125884016973,
|
126 |
-
"eval_loss": 0.
|
127 |
-
"eval_runtime":
|
128 |
-
"eval_samples_per_second":
|
129 |
-
"eval_steps_per_second": 4.
|
130 |
-
"eval_wer": 0.
|
131 |
"step": 1200
|
132 |
},
|
133 |
{
|
134 |
"epoch": 1.8387553041018387,
|
135 |
-
"eval_loss": 0.
|
136 |
-
"eval_runtime":
|
137 |
-
"eval_samples_per_second":
|
138 |
-
"eval_steps_per_second": 4.
|
139 |
-
"eval_wer": 0.
|
140 |
"step": 1300
|
141 |
},
|
142 |
{
|
143 |
"epoch": 1.9801980198019802,
|
144 |
-
"eval_loss": 0.
|
145 |
-
"eval_runtime":
|
146 |
-
"eval_samples_per_second":
|
147 |
-
"eval_steps_per_second": 4.
|
148 |
-
"eval_wer": 0.
|
149 |
"step": 1400
|
150 |
},
|
151 |
{
|
152 |
"epoch": 2.1216407355021216,
|
153 |
-
"grad_norm":
|
154 |
-
"learning_rate": 0.
|
155 |
-
"loss": 0.
|
156 |
"step": 1500
|
157 |
},
|
158 |
{
|
159 |
"epoch": 2.1216407355021216,
|
160 |
-
"eval_loss": 0.
|
161 |
-
"eval_runtime":
|
162 |
-
"eval_samples_per_second":
|
163 |
-
"eval_steps_per_second": 4.
|
164 |
-
"eval_wer": 0.
|
165 |
"step": 1500
|
166 |
},
|
167 |
{
|
168 |
"epoch": 2.263083451202263,
|
169 |
-
"eval_loss": 0.
|
170 |
-
"eval_runtime":
|
171 |
-
"eval_samples_per_second":
|
172 |
-
"eval_steps_per_second": 4.
|
173 |
-
"eval_wer": 0.
|
174 |
"step": 1600
|
175 |
},
|
176 |
{
|
177 |
"epoch": 2.4045261669024045,
|
178 |
-
"eval_loss": 0.
|
179 |
-
"eval_runtime":
|
180 |
-
"eval_samples_per_second":
|
181 |
-
"eval_steps_per_second": 4.
|
182 |
-
"eval_wer": 0.
|
183 |
"step": 1700
|
184 |
},
|
185 |
{
|
186 |
"epoch": 2.545968882602546,
|
187 |
-
"eval_loss": 0.
|
188 |
-
"eval_runtime":
|
189 |
-
"eval_samples_per_second":
|
190 |
-
"eval_steps_per_second": 4.
|
191 |
-
"eval_wer": 0.
|
192 |
"step": 1800
|
193 |
},
|
194 |
{
|
195 |
"epoch": 2.6874115983026874,
|
196 |
-
"eval_loss": 0.
|
197 |
-
"eval_runtime":
|
198 |
-
"eval_samples_per_second":
|
199 |
-
"eval_steps_per_second": 4.
|
200 |
-
"eval_wer": 0.
|
201 |
"step": 1900
|
202 |
},
|
203 |
{
|
204 |
"epoch": 2.828854314002829,
|
205 |
-
"grad_norm": 0.
|
206 |
-
"learning_rate": 7.
|
207 |
-
"loss": 0.
|
208 |
"step": 2000
|
209 |
},
|
210 |
{
|
211 |
"epoch": 2.828854314002829,
|
212 |
-
"eval_loss": 0.
|
213 |
-
"eval_runtime":
|
214 |
-
"eval_samples_per_second":
|
215 |
-
"eval_steps_per_second": 4.
|
216 |
-
"eval_wer": 0.
|
217 |
"step": 2000
|
218 |
},
|
219 |
{
|
220 |
"epoch": 2.9702970297029703,
|
221 |
-
"eval_loss": 0.
|
222 |
-
"eval_runtime":
|
223 |
-
"eval_samples_per_second":
|
224 |
-
"eval_steps_per_second": 4.
|
225 |
-
"eval_wer": 0.
|
226 |
"step": 2100
|
227 |
},
|
228 |
{
|
229 |
"epoch": 3.1117397454031117,
|
230 |
-
"eval_loss": 0.
|
231 |
-
"eval_runtime":
|
232 |
-
"eval_samples_per_second":
|
233 |
-
"eval_steps_per_second": 4.
|
234 |
-
"eval_wer": 0.
|
235 |
"step": 2200
|
236 |
},
|
237 |
{
|
238 |
"epoch": 3.253182461103253,
|
239 |
-
"eval_loss": 0.
|
240 |
-
"eval_runtime":
|
241 |
-
"eval_samples_per_second":
|
242 |
-
"eval_steps_per_second": 4.
|
243 |
-
"eval_wer": 0.
|
244 |
"step": 2300
|
245 |
},
|
246 |
{
|
247 |
"epoch": 3.3946251768033946,
|
248 |
-
"eval_loss": 0.
|
249 |
-
"eval_runtime":
|
250 |
-
"eval_samples_per_second":
|
251 |
-
"eval_steps_per_second": 4.
|
252 |
-
"eval_wer": 0.
|
253 |
"step": 2400
|
254 |
},
|
255 |
{
|
256 |
"epoch": 3.536067892503536,
|
257 |
-
"grad_norm": 0.
|
258 |
-
"learning_rate":
|
259 |
-
"loss": 0.
|
260 |
"step": 2500
|
261 |
},
|
262 |
{
|
263 |
"epoch": 3.536067892503536,
|
264 |
-
"eval_loss": 0.
|
265 |
-
"eval_runtime":
|
266 |
-
"eval_samples_per_second":
|
267 |
-
"eval_steps_per_second": 4.
|
268 |
-
"eval_wer": 0.
|
269 |
"step": 2500
|
270 |
},
|
271 |
{
|
272 |
"epoch": 3.536067892503536,
|
273 |
"step": 2500,
|
274 |
"total_flos": 9.55169606524761e+18,
|
275 |
-
"train_loss": 1.
|
276 |
-
"train_runtime":
|
277 |
-
"train_samples_per_second":
|
278 |
-
"train_steps_per_second": 0.
|
279 |
}
|
280 |
],
|
281 |
"logging_steps": 500,
|
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.14144271570014144,
|
13 |
+
"eval_loss": 4.035430908203125,
|
14 |
+
"eval_runtime": 153.5895,
|
15 |
+
"eval_samples_per_second": 36.825,
|
16 |
+
"eval_steps_per_second": 4.603,
|
17 |
"eval_wer": 1.0,
|
18 |
"step": 100
|
19 |
},
|
20 |
{
|
21 |
"epoch": 0.2828854314002829,
|
22 |
+
"eval_loss": 3.097731590270996,
|
23 |
+
"eval_runtime": 149.7934,
|
24 |
+
"eval_samples_per_second": 37.759,
|
25 |
+
"eval_steps_per_second": 4.72,
|
26 |
"eval_wer": 1.0,
|
27 |
"step": 200
|
28 |
},
|
29 |
{
|
30 |
"epoch": 0.4243281471004243,
|
31 |
+
"eval_loss": 3.0768725872039795,
|
32 |
+
"eval_runtime": 151.0662,
|
33 |
+
"eval_samples_per_second": 37.441,
|
34 |
+
"eval_steps_per_second": 4.68,
|
35 |
+
"eval_wer": 1.0,
|
36 |
"step": 300
|
37 |
},
|
38 |
{
|
39 |
"epoch": 0.5657708628005658,
|
40 |
+
"eval_loss": 1.373844027519226,
|
41 |
+
"eval_runtime": 150.8632,
|
42 |
+
"eval_samples_per_second": 37.491,
|
43 |
+
"eval_steps_per_second": 4.686,
|
44 |
+
"eval_wer": 0.8913622648777043,
|
45 |
"step": 400
|
46 |
},
|
47 |
{
|
48 |
"epoch": 0.7072135785007072,
|
49 |
+
"grad_norm": 2.2486917972564697,
|
50 |
+
"learning_rate": 0.0002964,
|
51 |
+
"loss": 3.7586,
|
52 |
"step": 500
|
53 |
},
|
54 |
{
|
55 |
"epoch": 0.7072135785007072,
|
56 |
+
"eval_loss": 1.0915191173553467,
|
57 |
+
"eval_runtime": 151.7555,
|
58 |
+
"eval_samples_per_second": 37.27,
|
59 |
+
"eval_steps_per_second": 4.659,
|
60 |
+
"eval_wer": 0.7692431148488156,
|
61 |
"step": 500
|
62 |
},
|
63 |
{
|
64 |
"epoch": 0.8486562942008486,
|
65 |
+
"eval_loss": 0.9360549449920654,
|
66 |
+
"eval_runtime": 152.9408,
|
67 |
+
"eval_samples_per_second": 36.982,
|
68 |
+
"eval_steps_per_second": 4.623,
|
69 |
+
"eval_wer": 0.6854818000898761,
|
70 |
"step": 600
|
71 |
},
|
72 |
{
|
73 |
"epoch": 0.9900990099009901,
|
74 |
+
"eval_loss": 0.8494995832443237,
|
75 |
+
"eval_runtime": 151.5832,
|
76 |
+
"eval_samples_per_second": 37.313,
|
77 |
+
"eval_steps_per_second": 4.664,
|
78 |
+
"eval_wer": 0.6247191371894459,
|
79 |
"step": 700
|
80 |
},
|
81 |
{
|
82 |
"epoch": 1.1315417256011315,
|
83 |
+
"eval_loss": 0.6885886788368225,
|
84 |
+
"eval_runtime": 151.913,
|
85 |
+
"eval_samples_per_second": 37.232,
|
86 |
+
"eval_steps_per_second": 4.654,
|
87 |
+
"eval_wer": 0.5397059767606086,
|
88 |
"step": 800
|
89 |
},
|
90 |
{
|
91 |
"epoch": 1.272984441301273,
|
92 |
+
"eval_loss": 0.6703779101371765,
|
93 |
+
"eval_runtime": 153.2384,
|
94 |
+
"eval_samples_per_second": 36.91,
|
95 |
+
"eval_steps_per_second": 4.614,
|
96 |
+
"eval_wer": 0.531215895230147,
|
97 |
"step": 900
|
98 |
},
|
99 |
{
|
100 |
"epoch": 1.4144271570014144,
|
101 |
+
"grad_norm": 0.8719882369041443,
|
102 |
+
"learning_rate": 0.0002259,
|
103 |
+
"loss": 0.8877,
|
104 |
"step": 1000
|
105 |
},
|
106 |
{
|
107 |
"epoch": 1.4144271570014144,
|
108 |
+
"eval_loss": 0.62369304895401,
|
109 |
+
"eval_runtime": 153.3779,
|
110 |
+
"eval_samples_per_second": 36.876,
|
111 |
+
"eval_steps_per_second": 4.61,
|
112 |
+
"eval_wer": 0.495040765230789,
|
113 |
"step": 1000
|
114 |
},
|
115 |
{
|
116 |
"epoch": 1.5558698727015559,
|
117 |
+
"eval_loss": 0.5992260575294495,
|
118 |
+
"eval_runtime": 152.5492,
|
119 |
+
"eval_samples_per_second": 37.077,
|
120 |
+
"eval_steps_per_second": 4.635,
|
121 |
+
"eval_wer": 0.4767606085895872,
|
122 |
"step": 1100
|
123 |
},
|
124 |
{
|
125 |
"epoch": 1.6973125884016973,
|
126 |
+
"eval_loss": 0.5729629397392273,
|
127 |
+
"eval_runtime": 152.035,
|
128 |
+
"eval_samples_per_second": 37.202,
|
129 |
+
"eval_steps_per_second": 4.65,
|
130 |
+
"eval_wer": 0.4521730756885151,
|
131 |
"step": 1200
|
132 |
},
|
133 |
{
|
134 |
"epoch": 1.8387553041018387,
|
135 |
+
"eval_loss": 0.5504249930381775,
|
136 |
+
"eval_runtime": 152.762,
|
137 |
+
"eval_samples_per_second": 37.025,
|
138 |
+
"eval_steps_per_second": 4.628,
|
139 |
+
"eval_wer": 0.4417891763497464,
|
140 |
"step": 1300
|
141 |
},
|
142 |
{
|
143 |
"epoch": 1.9801980198019802,
|
144 |
+
"eval_loss": 0.5288310050964355,
|
145 |
+
"eval_runtime": 152.1876,
|
146 |
+
"eval_samples_per_second": 37.165,
|
147 |
+
"eval_steps_per_second": 4.646,
|
148 |
+
"eval_wer": 0.42586826731719846,
|
149 |
"step": 1400
|
150 |
},
|
151 |
{
|
152 |
"epoch": 2.1216407355021216,
|
153 |
+
"grad_norm": 0.7933566570281982,
|
154 |
+
"learning_rate": 0.00015134999999999997,
|
155 |
+
"loss": 0.6844,
|
156 |
"step": 1500
|
157 |
},
|
158 |
{
|
159 |
"epoch": 2.1216407355021216,
|
160 |
+
"eval_loss": 0.5165240168571472,
|
161 |
+
"eval_runtime": 152.8573,
|
162 |
+
"eval_samples_per_second": 37.002,
|
163 |
+
"eval_steps_per_second": 4.625,
|
164 |
+
"eval_wer": 0.4216633498106182,
|
165 |
"step": 1500
|
166 |
},
|
167 |
{
|
168 |
"epoch": 2.263083451202263,
|
169 |
+
"eval_loss": 0.5071681141853333,
|
170 |
+
"eval_runtime": 152.7102,
|
171 |
+
"eval_samples_per_second": 37.037,
|
172 |
+
"eval_steps_per_second": 4.63,
|
173 |
+
"eval_wer": 0.41930410220196446,
|
174 |
"step": 1600
|
175 |
},
|
176 |
{
|
177 |
"epoch": 2.4045261669024045,
|
178 |
+
"eval_loss": 0.49842193722724915,
|
179 |
+
"eval_runtime": 153.4952,
|
180 |
+
"eval_samples_per_second": 36.848,
|
181 |
+
"eval_steps_per_second": 4.606,
|
182 |
+
"eval_wer": 0.41546831867496953,
|
183 |
"step": 1700
|
184 |
},
|
185 |
{
|
186 |
"epoch": 2.545968882602546,
|
187 |
+
"eval_loss": 0.4882013499736786,
|
188 |
+
"eval_runtime": 152.4926,
|
189 |
+
"eval_samples_per_second": 37.09,
|
190 |
+
"eval_steps_per_second": 4.636,
|
191 |
+
"eval_wer": 0.4096905694292868,
|
192 |
"step": 1800
|
193 |
},
|
194 |
{
|
195 |
"epoch": 2.6874115983026874,
|
196 |
+
"eval_loss": 0.4803565740585327,
|
197 |
+
"eval_runtime": 152.0827,
|
198 |
+
"eval_samples_per_second": 37.19,
|
199 |
+
"eval_steps_per_second": 4.649,
|
200 |
+
"eval_wer": 0.40800539256596263,
|
201 |
"step": 1900
|
202 |
},
|
203 |
{
|
204 |
"epoch": 2.828854314002829,
|
205 |
+
"grad_norm": 0.7038583159446716,
|
206 |
+
"learning_rate": 7.68e-05,
|
207 |
+
"loss": 0.537,
|
208 |
"step": 2000
|
209 |
},
|
210 |
{
|
211 |
"epoch": 2.828854314002829,
|
212 |
+
"eval_loss": 0.470017671585083,
|
213 |
+
"eval_runtime": 152.5003,
|
214 |
+
"eval_samples_per_second": 37.088,
|
215 |
+
"eval_steps_per_second": 4.636,
|
216 |
+
"eval_wer": 0.39272645567182385,
|
217 |
"step": 2000
|
218 |
},
|
219 |
{
|
220 |
"epoch": 2.9702970297029703,
|
221 |
+
"eval_loss": 0.46773043274879456,
|
222 |
+
"eval_runtime": 152.3208,
|
223 |
+
"eval_samples_per_second": 37.132,
|
224 |
+
"eval_steps_per_second": 4.642,
|
225 |
+
"eval_wer": 0.3884573409514027,
|
226 |
"step": 2100
|
227 |
},
|
228 |
{
|
229 |
"epoch": 3.1117397454031117,
|
230 |
+
"eval_loss": 0.4683248996734619,
|
231 |
+
"eval_runtime": 152.8021,
|
232 |
+
"eval_samples_per_second": 37.015,
|
233 |
+
"eval_steps_per_second": 4.627,
|
234 |
+
"eval_wer": 0.38572895936316365,
|
235 |
"step": 2200
|
236 |
},
|
237 |
{
|
238 |
"epoch": 3.253182461103253,
|
239 |
+
"eval_loss": 0.4618384838104248,
|
240 |
+
"eval_runtime": 152.4829,
|
241 |
+
"eval_samples_per_second": 37.093,
|
242 |
+
"eval_steps_per_second": 4.637,
|
243 |
+
"eval_wer": 0.3792289914617706,
|
244 |
"step": 2300
|
245 |
},
|
246 |
{
|
247 |
"epoch": 3.3946251768033946,
|
248 |
+
"eval_loss": 0.46039119362831116,
|
249 |
+
"eval_runtime": 152.4542,
|
250 |
+
"eval_samples_per_second": 37.1,
|
251 |
+
"eval_steps_per_second": 4.637,
|
252 |
+
"eval_wer": 0.3762759196250883,
|
253 |
"step": 2400
|
254 |
},
|
255 |
{
|
256 |
"epoch": 3.536067892503536,
|
257 |
+
"grad_norm": 0.6644078493118286,
|
258 |
+
"learning_rate": 2.1e-06,
|
259 |
+
"loss": 0.4434,
|
260 |
"step": 2500
|
261 |
},
|
262 |
{
|
263 |
"epoch": 3.536067892503536,
|
264 |
+
"eval_loss": 0.4588949382305145,
|
265 |
+
"eval_runtime": 152.8982,
|
266 |
+
"eval_samples_per_second": 36.992,
|
267 |
+
"eval_steps_per_second": 4.624,
|
268 |
+
"eval_wer": 0.3742697566925595,
|
269 |
"step": 2500
|
270 |
},
|
271 |
{
|
272 |
"epoch": 3.536067892503536,
|
273 |
"step": 2500,
|
274 |
"total_flos": 9.55169606524761e+18,
|
275 |
+
"train_loss": 1.2622116821289062,
|
276 |
+
"train_runtime": 6522.8506,
|
277 |
+
"train_samples_per_second": 12.265,
|
278 |
+
"train_steps_per_second": 0.383
|
279 |
}
|
280 |
],
|
281 |
"logging_steps": 500,
|