marinone94
commited on
Commit
•
3ceb64c
1
Parent(s):
2c31b89
Training in progress, step 2000
Browse files- checkpoint-2000/config.json +41 -0
- checkpoint-2000/optimizer.pt +3 -0
- checkpoint-2000/preprocessor_config.json +0 -0
- checkpoint-2000/pytorch_model.bin +3 -0
- checkpoint-2000/rng_state.pth +3 -0
- checkpoint-2000/scaler.pt +3 -0
- checkpoint-2000/scheduler.pt +3 -0
- checkpoint-2000/trainer_state.json +514 -0
- checkpoint-2000/training_args.bin +3 -0
- pytorch_model.bin +1 -1
checkpoint-2000/config.json
ADDED
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "marinone94/whisper-medium-nordic",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"activation_function": "gelu",
|
5 |
+
"architectures": [
|
6 |
+
"WhisperForConditionalGeneration"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.0,
|
9 |
+
"begin_suppress_tokens": [
|
10 |
+
220,
|
11 |
+
50257
|
12 |
+
],
|
13 |
+
"bos_token_id": 50257,
|
14 |
+
"d_model": 1024,
|
15 |
+
"decoder_attention_heads": 16,
|
16 |
+
"decoder_ffn_dim": 4096,
|
17 |
+
"decoder_layerdrop": 0.0,
|
18 |
+
"decoder_layers": 24,
|
19 |
+
"decoder_start_token_id": 50258,
|
20 |
+
"dropout": 0.0,
|
21 |
+
"encoder_attention_heads": 16,
|
22 |
+
"encoder_ffn_dim": 4096,
|
23 |
+
"encoder_layerdrop": 0.0,
|
24 |
+
"encoder_layers": 24,
|
25 |
+
"eos_token_id": 50257,
|
26 |
+
"forced_decoder_ids": null,
|
27 |
+
"init_std": 0.02,
|
28 |
+
"is_encoder_decoder": true,
|
29 |
+
"max_length": 448,
|
30 |
+
"max_source_positions": 1500,
|
31 |
+
"max_target_positions": 448,
|
32 |
+
"model_type": "whisper",
|
33 |
+
"num_hidden_layers": 24,
|
34 |
+
"num_mel_bins": 80,
|
35 |
+
"pad_token_id": 50257,
|
36 |
+
"scale_embedding": false,
|
37 |
+
"torch_dtype": "float32",
|
38 |
+
"transformers_version": "4.26.0.dev0",
|
39 |
+
"use_cache": false,
|
40 |
+
"vocab_size": 51865
|
41 |
+
}
|
checkpoint-2000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:259711c4e6978872a098051416766fdc14fdd8fac16daedeb90809ff04b25acd
|
3 |
+
size 6111428695
|
checkpoint-2000/preprocessor_config.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-2000/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:523e5675a34a184a100528b8c77ec2ff6d5c5fddb30ef2d034ca8b7b1cdae7e6
|
3 |
+
size 3055754841
|
checkpoint-2000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a0ac7588e56b2f3605270c18191b85d12253a2aa8ac37011f5dd7ee58036e40c
|
3 |
+
size 14575
|
checkpoint-2000/scaler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:93dcbb27405421069b5f31003e3f224646e375ff4c9d7ba4b0d265614f660d0f
|
3 |
+
size 557
|
checkpoint-2000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:37492482747b6e8c7f9b2ddf6b8c932454afa931831c235a6f136e9ca901a936
|
3 |
+
size 627
|
checkpoint-2000/trainer_state.json
ADDED
@@ -0,0 +1,514 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 11.37780883775938,
|
3 |
+
"best_model_checkpoint": "./checkpoint-2000",
|
4 |
+
"epoch": 6.0416,
|
5 |
+
"global_step": 2000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.01,
|
12 |
+
"learning_rate": 9.200000000000001e-07,
|
13 |
+
"loss": 0.5204,
|
14 |
+
"step": 25
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.02,
|
18 |
+
"learning_rate": 1.9200000000000003e-06,
|
19 |
+
"loss": 0.0582,
|
20 |
+
"step": 50
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.03,
|
24 |
+
"learning_rate": 2.92e-06,
|
25 |
+
"loss": 0.0527,
|
26 |
+
"step": 75
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.04,
|
30 |
+
"learning_rate": 3.920000000000001e-06,
|
31 |
+
"loss": 0.0587,
|
32 |
+
"step": 100
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.05,
|
36 |
+
"learning_rate": 4.92e-06,
|
37 |
+
"loss": 0.0659,
|
38 |
+
"step": 125
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.06,
|
42 |
+
"learning_rate": 5.92e-06,
|
43 |
+
"loss": 0.0681,
|
44 |
+
"step": 150
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.07,
|
48 |
+
"learning_rate": 6.92e-06,
|
49 |
+
"loss": 0.07,
|
50 |
+
"step": 175
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.08,
|
54 |
+
"learning_rate": 7.92e-06,
|
55 |
+
"loss": 0.0856,
|
56 |
+
"step": 200
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.09,
|
60 |
+
"learning_rate": 8.920000000000001e-06,
|
61 |
+
"loss": 0.0894,
|
62 |
+
"step": 225
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.1,
|
66 |
+
"learning_rate": 9.920000000000002e-06,
|
67 |
+
"loss": 0.0751,
|
68 |
+
"step": 250
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.11,
|
72 |
+
"learning_rate": 9.89777777777778e-06,
|
73 |
+
"loss": 0.0772,
|
74 |
+
"step": 275
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.12,
|
78 |
+
"learning_rate": 9.786666666666667e-06,
|
79 |
+
"loss": 0.0698,
|
80 |
+
"step": 300
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 1.0,
|
84 |
+
"learning_rate": 9.675555555555555e-06,
|
85 |
+
"loss": 0.0893,
|
86 |
+
"step": 325
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 1.01,
|
90 |
+
"learning_rate": 9.564444444444445e-06,
|
91 |
+
"loss": 0.0782,
|
92 |
+
"step": 350
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 1.02,
|
96 |
+
"learning_rate": 9.453333333333335e-06,
|
97 |
+
"loss": 0.0595,
|
98 |
+
"step": 375
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 1.03,
|
102 |
+
"learning_rate": 9.342222222222223e-06,
|
103 |
+
"loss": 0.0537,
|
104 |
+
"step": 400
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 1.04,
|
108 |
+
"learning_rate": 9.231111111111111e-06,
|
109 |
+
"loss": 0.052,
|
110 |
+
"step": 425
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 1.05,
|
114 |
+
"learning_rate": 9.12e-06,
|
115 |
+
"loss": 0.05,
|
116 |
+
"step": 450
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 1.06,
|
120 |
+
"learning_rate": 9.008888888888889e-06,
|
121 |
+
"loss": 0.0522,
|
122 |
+
"step": 475
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 1.07,
|
126 |
+
"learning_rate": 8.897777777777779e-06,
|
127 |
+
"loss": 0.0467,
|
128 |
+
"step": 500
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 1.08,
|
132 |
+
"learning_rate": 8.786666666666668e-06,
|
133 |
+
"loss": 0.0586,
|
134 |
+
"step": 525
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 1.09,
|
138 |
+
"learning_rate": 8.675555555555556e-06,
|
139 |
+
"loss": 0.0437,
|
140 |
+
"step": 550
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 1.1,
|
144 |
+
"learning_rate": 8.564444444444445e-06,
|
145 |
+
"loss": 0.0395,
|
146 |
+
"step": 575
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 1.11,
|
150 |
+
"learning_rate": 8.453333333333334e-06,
|
151 |
+
"loss": 0.0371,
|
152 |
+
"step": 600
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 1.12,
|
156 |
+
"learning_rate": 8.342222222222222e-06,
|
157 |
+
"loss": 0.0316,
|
158 |
+
"step": 625
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 2.01,
|
162 |
+
"learning_rate": 8.231111111111112e-06,
|
163 |
+
"loss": 0.0412,
|
164 |
+
"step": 650
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 2.02,
|
168 |
+
"learning_rate": 8.120000000000002e-06,
|
169 |
+
"loss": 0.0335,
|
170 |
+
"step": 675
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 2.03,
|
174 |
+
"learning_rate": 8.00888888888889e-06,
|
175 |
+
"loss": 0.0314,
|
176 |
+
"step": 700
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 2.04,
|
180 |
+
"learning_rate": 7.897777777777778e-06,
|
181 |
+
"loss": 0.0329,
|
182 |
+
"step": 725
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 2.05,
|
186 |
+
"learning_rate": 7.786666666666666e-06,
|
187 |
+
"loss": 0.0275,
|
188 |
+
"step": 750
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 2.06,
|
192 |
+
"learning_rate": 7.675555555555556e-06,
|
193 |
+
"loss": 0.028,
|
194 |
+
"step": 775
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 2.07,
|
198 |
+
"learning_rate": 7.564444444444446e-06,
|
199 |
+
"loss": 0.0256,
|
200 |
+
"step": 800
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 2.08,
|
204 |
+
"learning_rate": 7.453333333333334e-06,
|
205 |
+
"loss": 0.026,
|
206 |
+
"step": 825
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 2.09,
|
210 |
+
"learning_rate": 7.342222222222223e-06,
|
211 |
+
"loss": 0.0295,
|
212 |
+
"step": 850
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 2.1,
|
216 |
+
"learning_rate": 7.231111111111112e-06,
|
217 |
+
"loss": 0.0259,
|
218 |
+
"step": 875
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 2.11,
|
222 |
+
"learning_rate": 7.1200000000000004e-06,
|
223 |
+
"loss": 0.0173,
|
224 |
+
"step": 900
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 2.12,
|
228 |
+
"learning_rate": 7.008888888888889e-06,
|
229 |
+
"loss": 0.014,
|
230 |
+
"step": 925
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 3.0,
|
234 |
+
"learning_rate": 6.897777777777779e-06,
|
235 |
+
"loss": 0.0158,
|
236 |
+
"step": 950
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 3.01,
|
240 |
+
"learning_rate": 6.786666666666667e-06,
|
241 |
+
"loss": 0.0193,
|
242 |
+
"step": 975
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 3.02,
|
246 |
+
"learning_rate": 6.675555555555556e-06,
|
247 |
+
"loss": 0.0146,
|
248 |
+
"step": 1000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 3.02,
|
252 |
+
"eval_loss": 0.2546452581882477,
|
253 |
+
"eval_runtime": 2882.0567,
|
254 |
+
"eval_samples_per_second": 1.759,
|
255 |
+
"eval_steps_per_second": 0.11,
|
256 |
+
"eval_wer": 11.942264272658853,
|
257 |
+
"step": 1000
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 3.03,
|
261 |
+
"learning_rate": 6.564444444444446e-06,
|
262 |
+
"loss": 0.0129,
|
263 |
+
"step": 1025
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 3.04,
|
267 |
+
"learning_rate": 6.453333333333334e-06,
|
268 |
+
"loss": 0.0111,
|
269 |
+
"step": 1050
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 3.05,
|
273 |
+
"learning_rate": 6.342222222222223e-06,
|
274 |
+
"loss": 0.0143,
|
275 |
+
"step": 1075
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 3.06,
|
279 |
+
"learning_rate": 6.231111111111111e-06,
|
280 |
+
"loss": 0.0128,
|
281 |
+
"step": 1100
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 3.07,
|
285 |
+
"learning_rate": 6.120000000000001e-06,
|
286 |
+
"loss": 0.0105,
|
287 |
+
"step": 1125
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 3.08,
|
291 |
+
"learning_rate": 6.00888888888889e-06,
|
292 |
+
"loss": 0.012,
|
293 |
+
"step": 1150
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 3.09,
|
297 |
+
"learning_rate": 5.897777777777778e-06,
|
298 |
+
"loss": 0.0133,
|
299 |
+
"step": 1175
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 3.1,
|
303 |
+
"learning_rate": 5.7866666666666674e-06,
|
304 |
+
"loss": 0.0093,
|
305 |
+
"step": 1200
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 3.11,
|
309 |
+
"learning_rate": 5.675555555555556e-06,
|
310 |
+
"loss": 0.0097,
|
311 |
+
"step": 1225
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 3.12,
|
315 |
+
"learning_rate": 5.5644444444444444e-06,
|
316 |
+
"loss": 0.007,
|
317 |
+
"step": 1250
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 4.0,
|
321 |
+
"learning_rate": 5.453333333333334e-06,
|
322 |
+
"loss": 0.0059,
|
323 |
+
"step": 1275
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 4.01,
|
327 |
+
"learning_rate": 5.342222222222223e-06,
|
328 |
+
"loss": 0.01,
|
329 |
+
"step": 1300
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"epoch": 4.02,
|
333 |
+
"learning_rate": 5.231111111111111e-06,
|
334 |
+
"loss": 0.0086,
|
335 |
+
"step": 1325
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 4.03,
|
339 |
+
"learning_rate": 5.12e-06,
|
340 |
+
"loss": 0.0048,
|
341 |
+
"step": 1350
|
342 |
+
},
|
343 |
+
{
|
344 |
+
"epoch": 4.04,
|
345 |
+
"learning_rate": 5.00888888888889e-06,
|
346 |
+
"loss": 0.0058,
|
347 |
+
"step": 1375
|
348 |
+
},
|
349 |
+
{
|
350 |
+
"epoch": 4.05,
|
351 |
+
"learning_rate": 4.897777777777778e-06,
|
352 |
+
"loss": 0.0056,
|
353 |
+
"step": 1400
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 4.06,
|
357 |
+
"learning_rate": 4.786666666666667e-06,
|
358 |
+
"loss": 0.0059,
|
359 |
+
"step": 1425
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 4.07,
|
363 |
+
"learning_rate": 4.675555555555556e-06,
|
364 |
+
"loss": 0.0038,
|
365 |
+
"step": 1450
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 4.08,
|
369 |
+
"learning_rate": 4.564444444444445e-06,
|
370 |
+
"loss": 0.0049,
|
371 |
+
"step": 1475
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 4.09,
|
375 |
+
"learning_rate": 4.453333333333334e-06,
|
376 |
+
"loss": 0.0051,
|
377 |
+
"step": 1500
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 4.1,
|
381 |
+
"learning_rate": 4.3422222222222225e-06,
|
382 |
+
"loss": 0.0044,
|
383 |
+
"step": 1525
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 4.11,
|
387 |
+
"learning_rate": 4.2311111111111114e-06,
|
388 |
+
"loss": 0.0026,
|
389 |
+
"step": 1550
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 4.12,
|
393 |
+
"learning_rate": 4.12e-06,
|
394 |
+
"loss": 0.0036,
|
395 |
+
"step": 1575
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 5.01,
|
399 |
+
"learning_rate": 4.008888888888889e-06,
|
400 |
+
"loss": 0.0035,
|
401 |
+
"step": 1600
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 5.02,
|
405 |
+
"learning_rate": 3.897777777777778e-06,
|
406 |
+
"loss": 0.0024,
|
407 |
+
"step": 1625
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 5.03,
|
411 |
+
"learning_rate": 3.7866666666666667e-06,
|
412 |
+
"loss": 0.0026,
|
413 |
+
"step": 1650
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 5.04,
|
417 |
+
"learning_rate": 3.675555555555556e-06,
|
418 |
+
"loss": 0.0031,
|
419 |
+
"step": 1675
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 5.05,
|
423 |
+
"learning_rate": 3.564444444444445e-06,
|
424 |
+
"loss": 0.0045,
|
425 |
+
"step": 1700
|
426 |
+
},
|
427 |
+
{
|
428 |
+
"epoch": 5.06,
|
429 |
+
"learning_rate": 3.4533333333333334e-06,
|
430 |
+
"loss": 0.0019,
|
431 |
+
"step": 1725
|
432 |
+
},
|
433 |
+
{
|
434 |
+
"epoch": 5.07,
|
435 |
+
"learning_rate": 3.3422222222222224e-06,
|
436 |
+
"loss": 0.0018,
|
437 |
+
"step": 1750
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 5.08,
|
441 |
+
"learning_rate": 3.2311111111111117e-06,
|
442 |
+
"loss": 0.0033,
|
443 |
+
"step": 1775
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 5.09,
|
447 |
+
"learning_rate": 3.12e-06,
|
448 |
+
"loss": 0.0048,
|
449 |
+
"step": 1800
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 5.1,
|
453 |
+
"learning_rate": 3.008888888888889e-06,
|
454 |
+
"loss": 0.0019,
|
455 |
+
"step": 1825
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 5.11,
|
459 |
+
"learning_rate": 2.8977777777777785e-06,
|
460 |
+
"loss": 0.0015,
|
461 |
+
"step": 1850
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 5.12,
|
465 |
+
"learning_rate": 2.786666666666667e-06,
|
466 |
+
"loss": 0.0018,
|
467 |
+
"step": 1875
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 6.0,
|
471 |
+
"learning_rate": 2.675555555555556e-06,
|
472 |
+
"loss": 0.0017,
|
473 |
+
"step": 1900
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 6.01,
|
477 |
+
"learning_rate": 2.5644444444444444e-06,
|
478 |
+
"loss": 0.0017,
|
479 |
+
"step": 1925
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 6.02,
|
483 |
+
"learning_rate": 2.4533333333333333e-06,
|
484 |
+
"loss": 0.0009,
|
485 |
+
"step": 1950
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 6.03,
|
489 |
+
"learning_rate": 2.342222222222222e-06,
|
490 |
+
"loss": 0.0011,
|
491 |
+
"step": 1975
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 6.04,
|
495 |
+
"learning_rate": 2.2311111111111115e-06,
|
496 |
+
"loss": 0.0017,
|
497 |
+
"step": 2000
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 6.04,
|
501 |
+
"eval_loss": 0.29696929454803467,
|
502 |
+
"eval_runtime": 2899.6818,
|
503 |
+
"eval_samples_per_second": 1.748,
|
504 |
+
"eval_steps_per_second": 0.109,
|
505 |
+
"eval_wer": 11.37780883775938,
|
506 |
+
"step": 2000
|
507 |
+
}
|
508 |
+
],
|
509 |
+
"max_steps": 2500,
|
510 |
+
"num_train_epochs": 9223372036854775807,
|
511 |
+
"total_flos": 6.526972504571904e+19,
|
512 |
+
"trial_name": null,
|
513 |
+
"trial_params": null
|
514 |
+
}
|
checkpoint-2000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6e13de8ddcc61da424ec8e8d050cd43e412a6f52aa9be94507835652edf1dedf
|
3 |
+
size 3579
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 3055754841
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:523e5675a34a184a100528b8c77ec2ff6d5c5fddb30ef2d034ca8b7b1cdae7e6
|
3 |
size 3055754841
|