Commit
•
5ab1cc8
1
Parent(s):
7050ceb
Copied from Google Drive
Browse files- checkpoint-1000/config.json +42 -0
- checkpoint-1000/optimizer.pt +3 -0
- checkpoint-1000/preprocessor_config.json +0 -0
- checkpoint-1000/pytorch_model.bin +3 -0
- checkpoint-1000/rng_state.pth +3 -0
- checkpoint-1000/scaler.pt +3 -0
- checkpoint-1000/scheduler.pt +3 -0
- checkpoint-1000/trainer_state.json +265 -0
- checkpoint-1000/training_args.bin +3 -0
- checkpoint-2000/config.json +42 -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
- checkpoint-3000/config.json +42 -0
- checkpoint-3000/optimizer.pt +3 -0
- checkpoint-3000/preprocessor_config.json +0 -0
- checkpoint-3000/pytorch_model.bin +3 -0
- checkpoint-3000/rng_state.pth +3 -0
- checkpoint-3000/scaler.pt +3 -0
- checkpoint-3000/scheduler.pt +3 -0
- checkpoint-3000/trainer_state.json +763 -0
- checkpoint-3000/training_args.bin +3 -0
- checkpoint-4000/config.json +42 -0
- checkpoint-4000/optimizer.pt +3 -0
- checkpoint-4000/preprocessor_config.json +0 -0
- checkpoint-4000/pytorch_model.bin +3 -0
- checkpoint-4000/rng_state.pth +3 -0
- checkpoint-4000/scaler.pt +3 -0
- checkpoint-4000/scheduler.pt +3 -0
- checkpoint-4000/trainer_state.json +1012 -0
- checkpoint-4000/training_args.bin +3 -0
checkpoint-1000/config.json
ADDED
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "openai/whisper-tiny",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"activation_function": "gelu",
|
5 |
+
"architectures": [
|
6 |
+
"WhisperForConditionalGeneration"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.1,
|
9 |
+
"begin_suppress_tokens": [
|
10 |
+
220,
|
11 |
+
50257
|
12 |
+
],
|
13 |
+
"bos_token_id": 50257,
|
14 |
+
"d_model": 384,
|
15 |
+
"decoder_attention_heads": 8,
|
16 |
+
"decoder_ffn_dim": 1536,
|
17 |
+
"decoder_layerdrop": 0.1,
|
18 |
+
"decoder_layers": 4,
|
19 |
+
"decoder_start_token_id": 50258,
|
20 |
+
"dropout": 0.0,
|
21 |
+
"encoder_attention_heads": 8,
|
22 |
+
"encoder_ffn_dim": 1536,
|
23 |
+
"encoder_layerdrop": 0.1,
|
24 |
+
"encoder_layers": 4,
|
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": 4,
|
34 |
+
"num_mel_bins": 80,
|
35 |
+
"pad_token_id": 50257,
|
36 |
+
"scale_embedding": false,
|
37 |
+
"suppress_tokens": [],
|
38 |
+
"torch_dtype": "float32",
|
39 |
+
"transformers_version": "4.26.0.dev0",
|
40 |
+
"use_cache": false,
|
41 |
+
"vocab_size": 51865
|
42 |
+
}
|
checkpoint-1000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:05ffbda7524bc71269a6355327987a8840cff8d9f6418fb003a3acee7a34fbd2
|
3 |
+
size 302181925
|
checkpoint-1000/preprocessor_config.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-1000/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d2cce94786016561db53d460b0096c582df2fbf1a10494d95dc206cdb0962e33
|
3 |
+
size 151097331
|
checkpoint-1000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d47a10318ba2b69f5ec78c7c320068c02cc5838898053c274623df7ef83c49aa
|
3 |
+
size 14439
|
checkpoint-1000/scaler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d1743545be01d128d73f0b9a5573f8bb5feb01bf76961602634dd836db04d040
|
3 |
+
size 559
|
checkpoint-1000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a59a4bfdb165de79183672628b7b3641cf8d090ff8403ca7b04a61f992182382
|
3 |
+
size 623
|
checkpoint-1000/trainer_state.json
ADDED
@@ -0,0 +1,265 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 64.80865224625624,
|
3 |
+
"best_model_checkpoint": "./whisper-tiny-it-6/checkpoint-1000",
|
4 |
+
"epoch": 0.9541984732824428,
|
5 |
+
"global_step": 1000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.02,
|
12 |
+
"learning_rate": 3.8e-06,
|
13 |
+
"loss": 10.1581,
|
14 |
+
"step": 25
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.05,
|
18 |
+
"learning_rate": 8.599999999999999e-06,
|
19 |
+
"loss": 8.1216,
|
20 |
+
"step": 50
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.07,
|
24 |
+
"learning_rate": 1.3600000000000002e-05,
|
25 |
+
"loss": 6.9275,
|
26 |
+
"step": 75
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.1,
|
30 |
+
"learning_rate": 1.86e-05,
|
31 |
+
"loss": 5.6547,
|
32 |
+
"step": 100
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.12,
|
36 |
+
"learning_rate": 2.36e-05,
|
37 |
+
"loss": 5.2734,
|
38 |
+
"step": 125
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.14,
|
42 |
+
"learning_rate": 2.86e-05,
|
43 |
+
"loss": 5.5818,
|
44 |
+
"step": 150
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.17,
|
48 |
+
"learning_rate": 3.3600000000000004e-05,
|
49 |
+
"loss": 4.9292,
|
50 |
+
"step": 175
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.19,
|
54 |
+
"learning_rate": 3.86e-05,
|
55 |
+
"loss": 4.8341,
|
56 |
+
"step": 200
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.21,
|
60 |
+
"learning_rate": 4.36e-05,
|
61 |
+
"loss": 4.1192,
|
62 |
+
"step": 225
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.24,
|
66 |
+
"learning_rate": 4.86e-05,
|
67 |
+
"loss": 4.2109,
|
68 |
+
"step": 250
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.26,
|
72 |
+
"learning_rate": 5.360000000000001e-05,
|
73 |
+
"loss": 3.9181,
|
74 |
+
"step": 275
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.29,
|
78 |
+
"learning_rate": 5.86e-05,
|
79 |
+
"loss": 3.1145,
|
80 |
+
"step": 300
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.31,
|
84 |
+
"learning_rate": 6.36e-05,
|
85 |
+
"loss": 3.4049,
|
86 |
+
"step": 325
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.33,
|
90 |
+
"learning_rate": 6.860000000000001e-05,
|
91 |
+
"loss": 3.458,
|
92 |
+
"step": 350
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.36,
|
96 |
+
"learning_rate": 7.36e-05,
|
97 |
+
"loss": 3.6412,
|
98 |
+
"step": 375
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.38,
|
102 |
+
"learning_rate": 7.860000000000001e-05,
|
103 |
+
"loss": 3.0229,
|
104 |
+
"step": 400
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.41,
|
108 |
+
"learning_rate": 8.36e-05,
|
109 |
+
"loss": 2.9048,
|
110 |
+
"step": 425
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.43,
|
114 |
+
"learning_rate": 8.86e-05,
|
115 |
+
"loss": 2.8555,
|
116 |
+
"step": 450
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.45,
|
120 |
+
"learning_rate": 9.360000000000001e-05,
|
121 |
+
"loss": 2.4833,
|
122 |
+
"step": 475
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 0.48,
|
126 |
+
"learning_rate": 9.86e-05,
|
127 |
+
"loss": 2.7599,
|
128 |
+
"step": 500
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.5,
|
132 |
+
"learning_rate": 9.948571428571429e-05,
|
133 |
+
"loss": 2.8813,
|
134 |
+
"step": 525
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 0.52,
|
138 |
+
"learning_rate": 9.877142857142858e-05,
|
139 |
+
"loss": 2.6932,
|
140 |
+
"step": 550
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 0.55,
|
144 |
+
"learning_rate": 9.805714285714286e-05,
|
145 |
+
"loss": 2.7549,
|
146 |
+
"step": 575
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 0.57,
|
150 |
+
"learning_rate": 9.734285714285714e-05,
|
151 |
+
"loss": 2.7286,
|
152 |
+
"step": 600
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 0.6,
|
156 |
+
"learning_rate": 9.662857142857144e-05,
|
157 |
+
"loss": 2.1775,
|
158 |
+
"step": 625
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.62,
|
162 |
+
"learning_rate": 9.591428571428572e-05,
|
163 |
+
"loss": 2.4533,
|
164 |
+
"step": 650
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.64,
|
168 |
+
"learning_rate": 9.52e-05,
|
169 |
+
"loss": 3.0003,
|
170 |
+
"step": 675
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.67,
|
174 |
+
"learning_rate": 9.448571428571429e-05,
|
175 |
+
"loss": 2.6656,
|
176 |
+
"step": 700
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.69,
|
180 |
+
"learning_rate": 9.377142857142857e-05,
|
181 |
+
"loss": 2.4105,
|
182 |
+
"step": 725
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 0.72,
|
186 |
+
"learning_rate": 9.305714285714286e-05,
|
187 |
+
"loss": 2.2225,
|
188 |
+
"step": 750
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.74,
|
192 |
+
"learning_rate": 9.234285714285714e-05,
|
193 |
+
"loss": 2.6048,
|
194 |
+
"step": 775
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 0.76,
|
198 |
+
"learning_rate": 9.162857142857144e-05,
|
199 |
+
"loss": 2.3541,
|
200 |
+
"step": 800
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 0.79,
|
204 |
+
"learning_rate": 9.091428571428572e-05,
|
205 |
+
"loss": 1.6534,
|
206 |
+
"step": 825
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 0.81,
|
210 |
+
"learning_rate": 9.020000000000001e-05,
|
211 |
+
"loss": 2.6717,
|
212 |
+
"step": 850
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.83,
|
216 |
+
"learning_rate": 8.948571428571429e-05,
|
217 |
+
"loss": 2.1232,
|
218 |
+
"step": 875
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 0.86,
|
222 |
+
"learning_rate": 8.877142857142857e-05,
|
223 |
+
"loss": 2.1955,
|
224 |
+
"step": 900
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 0.88,
|
228 |
+
"learning_rate": 8.805714285714286e-05,
|
229 |
+
"loss": 2.4756,
|
230 |
+
"step": 925
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 0.91,
|
234 |
+
"learning_rate": 8.734285714285714e-05,
|
235 |
+
"loss": 2.0444,
|
236 |
+
"step": 950
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 0.93,
|
240 |
+
"learning_rate": 8.662857142857144e-05,
|
241 |
+
"loss": 2.124,
|
242 |
+
"step": 975
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 0.95,
|
246 |
+
"learning_rate": 8.591428571428572e-05,
|
247 |
+
"loss": 1.7168,
|
248 |
+
"step": 1000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 0.95,
|
252 |
+
"eval_loss": 1.2106691598892212,
|
253 |
+
"eval_runtime": 380.7845,
|
254 |
+
"eval_samples_per_second": 3.939,
|
255 |
+
"eval_steps_per_second": 0.494,
|
256 |
+
"eval_wer": 64.80865224625624,
|
257 |
+
"step": 1000
|
258 |
+
}
|
259 |
+
],
|
260 |
+
"max_steps": 4000,
|
261 |
+
"num_train_epochs": 4,
|
262 |
+
"total_flos": 3.9390216192e+17,
|
263 |
+
"trial_name": null,
|
264 |
+
"trial_params": null
|
265 |
+
}
|
checkpoint-1000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e74aa035d76429a8a28d6b59538373cb3ba9a6d7d359ac8980aafba69d60558d
|
3 |
+
size 3503
|
checkpoint-2000/config.json
ADDED
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "openai/whisper-tiny",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"activation_function": "gelu",
|
5 |
+
"architectures": [
|
6 |
+
"WhisperForConditionalGeneration"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.1,
|
9 |
+
"begin_suppress_tokens": [
|
10 |
+
220,
|
11 |
+
50257
|
12 |
+
],
|
13 |
+
"bos_token_id": 50257,
|
14 |
+
"d_model": 384,
|
15 |
+
"decoder_attention_heads": 8,
|
16 |
+
"decoder_ffn_dim": 1536,
|
17 |
+
"decoder_layerdrop": 0.1,
|
18 |
+
"decoder_layers": 4,
|
19 |
+
"decoder_start_token_id": 50258,
|
20 |
+
"dropout": 0.0,
|
21 |
+
"encoder_attention_heads": 8,
|
22 |
+
"encoder_ffn_dim": 1536,
|
23 |
+
"encoder_layerdrop": 0.1,
|
24 |
+
"encoder_layers": 4,
|
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": 4,
|
34 |
+
"num_mel_bins": 80,
|
35 |
+
"pad_token_id": 50257,
|
36 |
+
"scale_embedding": false,
|
37 |
+
"suppress_tokens": [],
|
38 |
+
"torch_dtype": "float32",
|
39 |
+
"transformers_version": "4.26.0.dev0",
|
40 |
+
"use_cache": false,
|
41 |
+
"vocab_size": 51865
|
42 |
+
}
|
checkpoint-2000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0b9b72311e9b6b96ac639eccd5f5a0fda8a0995c53107f34107a354e66edf501
|
3 |
+
size 302181925
|
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:ad8aacd744c133844bd08ab959b923fb6f1b6c981aee787d9c7db61616d0e084
|
3 |
+
size 151097331
|
checkpoint-2000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:527897972ad0b8c2d327e91ced486740dd032ecf48f1296caeebd4ad20ce1dfb
|
3 |
+
size 14503
|
checkpoint-2000/scaler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cff71612ad265a5e145f561501d3b2478a67ec02fc6f06a4ed870034055aebad
|
3 |
+
size 559
|
checkpoint-2000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8bc1030fa5f85102b95619ee24d130e3788aa2396fd5c6c65068688293450d9a
|
3 |
+
size 623
|
checkpoint-2000/trainer_state.json
ADDED
@@ -0,0 +1,514 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 53.001941209095946,
|
3 |
+
"best_model_checkpoint": "./whisper-tiny-it-6/checkpoint-2000",
|
4 |
+
"epoch": 1.9083969465648853,
|
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": 0.02,
|
12 |
+
"learning_rate": 3.8e-06,
|
13 |
+
"loss": 10.1581,
|
14 |
+
"step": 25
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.05,
|
18 |
+
"learning_rate": 8.599999999999999e-06,
|
19 |
+
"loss": 8.1216,
|
20 |
+
"step": 50
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.07,
|
24 |
+
"learning_rate": 1.3600000000000002e-05,
|
25 |
+
"loss": 6.9275,
|
26 |
+
"step": 75
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.1,
|
30 |
+
"learning_rate": 1.86e-05,
|
31 |
+
"loss": 5.6547,
|
32 |
+
"step": 100
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.12,
|
36 |
+
"learning_rate": 2.36e-05,
|
37 |
+
"loss": 5.2734,
|
38 |
+
"step": 125
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.14,
|
42 |
+
"learning_rate": 2.86e-05,
|
43 |
+
"loss": 5.5818,
|
44 |
+
"step": 150
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.17,
|
48 |
+
"learning_rate": 3.3600000000000004e-05,
|
49 |
+
"loss": 4.9292,
|
50 |
+
"step": 175
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.19,
|
54 |
+
"learning_rate": 3.86e-05,
|
55 |
+
"loss": 4.8341,
|
56 |
+
"step": 200
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.21,
|
60 |
+
"learning_rate": 4.36e-05,
|
61 |
+
"loss": 4.1192,
|
62 |
+
"step": 225
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.24,
|
66 |
+
"learning_rate": 4.86e-05,
|
67 |
+
"loss": 4.2109,
|
68 |
+
"step": 250
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.26,
|
72 |
+
"learning_rate": 5.360000000000001e-05,
|
73 |
+
"loss": 3.9181,
|
74 |
+
"step": 275
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.29,
|
78 |
+
"learning_rate": 5.86e-05,
|
79 |
+
"loss": 3.1145,
|
80 |
+
"step": 300
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.31,
|
84 |
+
"learning_rate": 6.36e-05,
|
85 |
+
"loss": 3.4049,
|
86 |
+
"step": 325
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.33,
|
90 |
+
"learning_rate": 6.860000000000001e-05,
|
91 |
+
"loss": 3.458,
|
92 |
+
"step": 350
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.36,
|
96 |
+
"learning_rate": 7.36e-05,
|
97 |
+
"loss": 3.6412,
|
98 |
+
"step": 375
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.38,
|
102 |
+
"learning_rate": 7.860000000000001e-05,
|
103 |
+
"loss": 3.0229,
|
104 |
+
"step": 400
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.41,
|
108 |
+
"learning_rate": 8.36e-05,
|
109 |
+
"loss": 2.9048,
|
110 |
+
"step": 425
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.43,
|
114 |
+
"learning_rate": 8.86e-05,
|
115 |
+
"loss": 2.8555,
|
116 |
+
"step": 450
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.45,
|
120 |
+
"learning_rate": 9.360000000000001e-05,
|
121 |
+
"loss": 2.4833,
|
122 |
+
"step": 475
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 0.48,
|
126 |
+
"learning_rate": 9.86e-05,
|
127 |
+
"loss": 2.7599,
|
128 |
+
"step": 500
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.5,
|
132 |
+
"learning_rate": 9.948571428571429e-05,
|
133 |
+
"loss": 2.8813,
|
134 |
+
"step": 525
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 0.52,
|
138 |
+
"learning_rate": 9.877142857142858e-05,
|
139 |
+
"loss": 2.6932,
|
140 |
+
"step": 550
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 0.55,
|
144 |
+
"learning_rate": 9.805714285714286e-05,
|
145 |
+
"loss": 2.7549,
|
146 |
+
"step": 575
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 0.57,
|
150 |
+
"learning_rate": 9.734285714285714e-05,
|
151 |
+
"loss": 2.7286,
|
152 |
+
"step": 600
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 0.6,
|
156 |
+
"learning_rate": 9.662857142857144e-05,
|
157 |
+
"loss": 2.1775,
|
158 |
+
"step": 625
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.62,
|
162 |
+
"learning_rate": 9.591428571428572e-05,
|
163 |
+
"loss": 2.4533,
|
164 |
+
"step": 650
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.64,
|
168 |
+
"learning_rate": 9.52e-05,
|
169 |
+
"loss": 3.0003,
|
170 |
+
"step": 675
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.67,
|
174 |
+
"learning_rate": 9.448571428571429e-05,
|
175 |
+
"loss": 2.6656,
|
176 |
+
"step": 700
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.69,
|
180 |
+
"learning_rate": 9.377142857142857e-05,
|
181 |
+
"loss": 2.4105,
|
182 |
+
"step": 725
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 0.72,
|
186 |
+
"learning_rate": 9.305714285714286e-05,
|
187 |
+
"loss": 2.2225,
|
188 |
+
"step": 750
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.74,
|
192 |
+
"learning_rate": 9.234285714285714e-05,
|
193 |
+
"loss": 2.6048,
|
194 |
+
"step": 775
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 0.76,
|
198 |
+
"learning_rate": 9.162857142857144e-05,
|
199 |
+
"loss": 2.3541,
|
200 |
+
"step": 800
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 0.79,
|
204 |
+
"learning_rate": 9.091428571428572e-05,
|
205 |
+
"loss": 1.6534,
|
206 |
+
"step": 825
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 0.81,
|
210 |
+
"learning_rate": 9.020000000000001e-05,
|
211 |
+
"loss": 2.6717,
|
212 |
+
"step": 850
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.83,
|
216 |
+
"learning_rate": 8.948571428571429e-05,
|
217 |
+
"loss": 2.1232,
|
218 |
+
"step": 875
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 0.86,
|
222 |
+
"learning_rate": 8.877142857142857e-05,
|
223 |
+
"loss": 2.1955,
|
224 |
+
"step": 900
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 0.88,
|
228 |
+
"learning_rate": 8.805714285714286e-05,
|
229 |
+
"loss": 2.4756,
|
230 |
+
"step": 925
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 0.91,
|
234 |
+
"learning_rate": 8.734285714285714e-05,
|
235 |
+
"loss": 2.0444,
|
236 |
+
"step": 950
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 0.93,
|
240 |
+
"learning_rate": 8.662857142857144e-05,
|
241 |
+
"loss": 2.124,
|
242 |
+
"step": 975
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 0.95,
|
246 |
+
"learning_rate": 8.591428571428572e-05,
|
247 |
+
"loss": 1.7168,
|
248 |
+
"step": 1000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 0.95,
|
252 |
+
"eval_loss": 1.2106691598892212,
|
253 |
+
"eval_runtime": 380.7845,
|
254 |
+
"eval_samples_per_second": 3.939,
|
255 |
+
"eval_steps_per_second": 0.494,
|
256 |
+
"eval_wer": 64.80865224625624,
|
257 |
+
"step": 1000
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 0.98,
|
261 |
+
"learning_rate": 8.52e-05,
|
262 |
+
"loss": 2.451,
|
263 |
+
"step": 1025
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 1.0,
|
267 |
+
"learning_rate": 8.448571428571429e-05,
|
268 |
+
"loss": 1.4766,
|
269 |
+
"step": 1050
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 1.03,
|
273 |
+
"learning_rate": 8.377142857142858e-05,
|
274 |
+
"loss": 1.6965,
|
275 |
+
"step": 1075
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 1.05,
|
279 |
+
"learning_rate": 8.305714285714285e-05,
|
280 |
+
"loss": 1.4733,
|
281 |
+
"step": 1100
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 1.07,
|
285 |
+
"learning_rate": 8.234285714285714e-05,
|
286 |
+
"loss": 1.9818,
|
287 |
+
"step": 1125
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 1.1,
|
291 |
+
"learning_rate": 8.162857142857144e-05,
|
292 |
+
"loss": 2.0946,
|
293 |
+
"step": 1150
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 1.12,
|
297 |
+
"learning_rate": 8.091428571428572e-05,
|
298 |
+
"loss": 2.3333,
|
299 |
+
"step": 1175
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 1.15,
|
303 |
+
"learning_rate": 8.020000000000001e-05,
|
304 |
+
"loss": 1.7366,
|
305 |
+
"step": 1200
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 1.17,
|
309 |
+
"learning_rate": 7.948571428571429e-05,
|
310 |
+
"loss": 1.8155,
|
311 |
+
"step": 1225
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 1.19,
|
315 |
+
"learning_rate": 7.877142857142857e-05,
|
316 |
+
"loss": 2.1239,
|
317 |
+
"step": 1250
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 1.22,
|
321 |
+
"learning_rate": 7.805714285714286e-05,
|
322 |
+
"loss": 1.6473,
|
323 |
+
"step": 1275
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 1.24,
|
327 |
+
"learning_rate": 7.734285714285714e-05,
|
328 |
+
"loss": 1.1093,
|
329 |
+
"step": 1300
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"epoch": 1.26,
|
333 |
+
"learning_rate": 7.662857142857142e-05,
|
334 |
+
"loss": 1.5792,
|
335 |
+
"step": 1325
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 1.29,
|
339 |
+
"learning_rate": 7.591428571428572e-05,
|
340 |
+
"loss": 1.9768,
|
341 |
+
"step": 1350
|
342 |
+
},
|
343 |
+
{
|
344 |
+
"epoch": 1.31,
|
345 |
+
"learning_rate": 7.52e-05,
|
346 |
+
"loss": 1.5958,
|
347 |
+
"step": 1375
|
348 |
+
},
|
349 |
+
{
|
350 |
+
"epoch": 1.34,
|
351 |
+
"learning_rate": 7.448571428571429e-05,
|
352 |
+
"loss": 1.6682,
|
353 |
+
"step": 1400
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 1.36,
|
357 |
+
"learning_rate": 7.377142857142859e-05,
|
358 |
+
"loss": 1.6654,
|
359 |
+
"step": 1425
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 1.38,
|
363 |
+
"learning_rate": 7.305714285714285e-05,
|
364 |
+
"loss": 1.6971,
|
365 |
+
"step": 1450
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 1.41,
|
369 |
+
"learning_rate": 7.234285714285715e-05,
|
370 |
+
"loss": 1.5318,
|
371 |
+
"step": 1475
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 1.43,
|
375 |
+
"learning_rate": 7.162857142857144e-05,
|
376 |
+
"loss": 1.3164,
|
377 |
+
"step": 1500
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 1.46,
|
381 |
+
"learning_rate": 7.091428571428572e-05,
|
382 |
+
"loss": 2.14,
|
383 |
+
"step": 1525
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 1.48,
|
387 |
+
"learning_rate": 7.02e-05,
|
388 |
+
"loss": 1.9155,
|
389 |
+
"step": 1550
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 1.5,
|
393 |
+
"learning_rate": 6.948571428571429e-05,
|
394 |
+
"loss": 1.4583,
|
395 |
+
"step": 1575
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 1.53,
|
399 |
+
"learning_rate": 6.877142857142857e-05,
|
400 |
+
"loss": 1.5538,
|
401 |
+
"step": 1600
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 1.55,
|
405 |
+
"learning_rate": 6.805714285714287e-05,
|
406 |
+
"loss": 1.1043,
|
407 |
+
"step": 1625
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 1.57,
|
411 |
+
"learning_rate": 6.734285714285715e-05,
|
412 |
+
"loss": 1.8925,
|
413 |
+
"step": 1650
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 1.6,
|
417 |
+
"learning_rate": 6.662857142857143e-05,
|
418 |
+
"loss": 1.7625,
|
419 |
+
"step": 1675
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 1.62,
|
423 |
+
"learning_rate": 6.591428571428572e-05,
|
424 |
+
"loss": 1.082,
|
425 |
+
"step": 1700
|
426 |
+
},
|
427 |
+
{
|
428 |
+
"epoch": 1.65,
|
429 |
+
"learning_rate": 6.52e-05,
|
430 |
+
"loss": 1.3789,
|
431 |
+
"step": 1725
|
432 |
+
},
|
433 |
+
{
|
434 |
+
"epoch": 1.67,
|
435 |
+
"learning_rate": 6.448571428571429e-05,
|
436 |
+
"loss": 1.568,
|
437 |
+
"step": 1750
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 1.69,
|
441 |
+
"learning_rate": 6.377142857142857e-05,
|
442 |
+
"loss": 1.3194,
|
443 |
+
"step": 1775
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 1.72,
|
447 |
+
"learning_rate": 6.305714285714285e-05,
|
448 |
+
"loss": 1.3629,
|
449 |
+
"step": 1800
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 1.74,
|
453 |
+
"learning_rate": 6.234285714285715e-05,
|
454 |
+
"loss": 1.6437,
|
455 |
+
"step": 1825
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 1.77,
|
459 |
+
"learning_rate": 6.162857142857144e-05,
|
460 |
+
"loss": 1.0288,
|
461 |
+
"step": 1850
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 1.79,
|
465 |
+
"learning_rate": 6.091428571428571e-05,
|
466 |
+
"loss": 1.5278,
|
467 |
+
"step": 1875
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 1.81,
|
471 |
+
"learning_rate": 6.02e-05,
|
472 |
+
"loss": 1.2791,
|
473 |
+
"step": 1900
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 1.84,
|
477 |
+
"learning_rate": 5.9485714285714286e-05,
|
478 |
+
"loss": 1.6202,
|
479 |
+
"step": 1925
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 1.86,
|
483 |
+
"learning_rate": 5.877142857142858e-05,
|
484 |
+
"loss": 1.3762,
|
485 |
+
"step": 1950
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 1.88,
|
489 |
+
"learning_rate": 5.805714285714285e-05,
|
490 |
+
"loss": 1.8212,
|
491 |
+
"step": 1975
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 1.91,
|
495 |
+
"learning_rate": 5.734285714285714e-05,
|
496 |
+
"loss": 1.1073,
|
497 |
+
"step": 2000
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 1.91,
|
501 |
+
"eval_loss": 0.9890535473823547,
|
502 |
+
"eval_runtime": 356.3677,
|
503 |
+
"eval_samples_per_second": 4.209,
|
504 |
+
"eval_steps_per_second": 0.528,
|
505 |
+
"eval_wer": 53.001941209095946,
|
506 |
+
"step": 2000
|
507 |
+
}
|
508 |
+
],
|
509 |
+
"max_steps": 4000,
|
510 |
+
"num_train_epochs": 4,
|
511 |
+
"total_flos": 7.8763199164416e+17,
|
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:e74aa035d76429a8a28d6b59538373cb3ba9a6d7d359ac8980aafba69d60558d
|
3 |
+
size 3503
|
checkpoint-3000/config.json
ADDED
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "openai/whisper-tiny",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"activation_function": "gelu",
|
5 |
+
"architectures": [
|
6 |
+
"WhisperForConditionalGeneration"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.1,
|
9 |
+
"begin_suppress_tokens": [
|
10 |
+
220,
|
11 |
+
50257
|
12 |
+
],
|
13 |
+
"bos_token_id": 50257,
|
14 |
+
"d_model": 384,
|
15 |
+
"decoder_attention_heads": 8,
|
16 |
+
"decoder_ffn_dim": 1536,
|
17 |
+
"decoder_layerdrop": 0.1,
|
18 |
+
"decoder_layers": 4,
|
19 |
+
"decoder_start_token_id": 50258,
|
20 |
+
"dropout": 0.0,
|
21 |
+
"encoder_attention_heads": 8,
|
22 |
+
"encoder_ffn_dim": 1536,
|
23 |
+
"encoder_layerdrop": 0.1,
|
24 |
+
"encoder_layers": 4,
|
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": 4,
|
34 |
+
"num_mel_bins": 80,
|
35 |
+
"pad_token_id": 50257,
|
36 |
+
"scale_embedding": false,
|
37 |
+
"suppress_tokens": [],
|
38 |
+
"torch_dtype": "float32",
|
39 |
+
"transformers_version": "4.26.0.dev0",
|
40 |
+
"use_cache": false,
|
41 |
+
"vocab_size": 51865
|
42 |
+
}
|
checkpoint-3000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:659746991c8f7aa15225c342b9b3704ffd96d93c6e0bc3132ed435193c0df415
|
3 |
+
size 302181925
|
checkpoint-3000/preprocessor_config.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-3000/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:835fdb6607c751ea362b408cf3881889fd0886fa4180d9618bf83e08defbe00a
|
3 |
+
size 151097331
|
checkpoint-3000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:924f63f53e3b8ea81d310e923de33f39bc0f1f154a52af5f9b3d2bf1bc468087
|
3 |
+
size 14503
|
checkpoint-3000/scaler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0b5334522442617bda390487fd8b8aa38024ce1f5010660bdca1f02d47facc0d
|
3 |
+
size 559
|
checkpoint-3000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e5cd6756f0068206f0836358868b1cf54c8f5f106d4746d1849166cba117e923
|
3 |
+
size 623
|
checkpoint-3000/trainer_state.json
ADDED
@@ -0,0 +1,763 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 47.76760953965613,
|
3 |
+
"best_model_checkpoint": "./whisper-tiny-it-6/checkpoint-3000",
|
4 |
+
"epoch": 2.8625954198473282,
|
5 |
+
"global_step": 3000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.02,
|
12 |
+
"learning_rate": 3.8e-06,
|
13 |
+
"loss": 10.1581,
|
14 |
+
"step": 25
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.05,
|
18 |
+
"learning_rate": 8.599999999999999e-06,
|
19 |
+
"loss": 8.1216,
|
20 |
+
"step": 50
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.07,
|
24 |
+
"learning_rate": 1.3600000000000002e-05,
|
25 |
+
"loss": 6.9275,
|
26 |
+
"step": 75
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.1,
|
30 |
+
"learning_rate": 1.86e-05,
|
31 |
+
"loss": 5.6547,
|
32 |
+
"step": 100
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.12,
|
36 |
+
"learning_rate": 2.36e-05,
|
37 |
+
"loss": 5.2734,
|
38 |
+
"step": 125
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.14,
|
42 |
+
"learning_rate": 2.86e-05,
|
43 |
+
"loss": 5.5818,
|
44 |
+
"step": 150
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.17,
|
48 |
+
"learning_rate": 3.3600000000000004e-05,
|
49 |
+
"loss": 4.9292,
|
50 |
+
"step": 175
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.19,
|
54 |
+
"learning_rate": 3.86e-05,
|
55 |
+
"loss": 4.8341,
|
56 |
+
"step": 200
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.21,
|
60 |
+
"learning_rate": 4.36e-05,
|
61 |
+
"loss": 4.1192,
|
62 |
+
"step": 225
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.24,
|
66 |
+
"learning_rate": 4.86e-05,
|
67 |
+
"loss": 4.2109,
|
68 |
+
"step": 250
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.26,
|
72 |
+
"learning_rate": 5.360000000000001e-05,
|
73 |
+
"loss": 3.9181,
|
74 |
+
"step": 275
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.29,
|
78 |
+
"learning_rate": 5.86e-05,
|
79 |
+
"loss": 3.1145,
|
80 |
+
"step": 300
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.31,
|
84 |
+
"learning_rate": 6.36e-05,
|
85 |
+
"loss": 3.4049,
|
86 |
+
"step": 325
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.33,
|
90 |
+
"learning_rate": 6.860000000000001e-05,
|
91 |
+
"loss": 3.458,
|
92 |
+
"step": 350
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.36,
|
96 |
+
"learning_rate": 7.36e-05,
|
97 |
+
"loss": 3.6412,
|
98 |
+
"step": 375
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.38,
|
102 |
+
"learning_rate": 7.860000000000001e-05,
|
103 |
+
"loss": 3.0229,
|
104 |
+
"step": 400
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.41,
|
108 |
+
"learning_rate": 8.36e-05,
|
109 |
+
"loss": 2.9048,
|
110 |
+
"step": 425
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.43,
|
114 |
+
"learning_rate": 8.86e-05,
|
115 |
+
"loss": 2.8555,
|
116 |
+
"step": 450
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.45,
|
120 |
+
"learning_rate": 9.360000000000001e-05,
|
121 |
+
"loss": 2.4833,
|
122 |
+
"step": 475
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 0.48,
|
126 |
+
"learning_rate": 9.86e-05,
|
127 |
+
"loss": 2.7599,
|
128 |
+
"step": 500
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.5,
|
132 |
+
"learning_rate": 9.948571428571429e-05,
|
133 |
+
"loss": 2.8813,
|
134 |
+
"step": 525
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 0.52,
|
138 |
+
"learning_rate": 9.877142857142858e-05,
|
139 |
+
"loss": 2.6932,
|
140 |
+
"step": 550
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 0.55,
|
144 |
+
"learning_rate": 9.805714285714286e-05,
|
145 |
+
"loss": 2.7549,
|
146 |
+
"step": 575
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 0.57,
|
150 |
+
"learning_rate": 9.734285714285714e-05,
|
151 |
+
"loss": 2.7286,
|
152 |
+
"step": 600
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 0.6,
|
156 |
+
"learning_rate": 9.662857142857144e-05,
|
157 |
+
"loss": 2.1775,
|
158 |
+
"step": 625
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.62,
|
162 |
+
"learning_rate": 9.591428571428572e-05,
|
163 |
+
"loss": 2.4533,
|
164 |
+
"step": 650
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.64,
|
168 |
+
"learning_rate": 9.52e-05,
|
169 |
+
"loss": 3.0003,
|
170 |
+
"step": 675
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.67,
|
174 |
+
"learning_rate": 9.448571428571429e-05,
|
175 |
+
"loss": 2.6656,
|
176 |
+
"step": 700
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.69,
|
180 |
+
"learning_rate": 9.377142857142857e-05,
|
181 |
+
"loss": 2.4105,
|
182 |
+
"step": 725
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 0.72,
|
186 |
+
"learning_rate": 9.305714285714286e-05,
|
187 |
+
"loss": 2.2225,
|
188 |
+
"step": 750
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.74,
|
192 |
+
"learning_rate": 9.234285714285714e-05,
|
193 |
+
"loss": 2.6048,
|
194 |
+
"step": 775
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 0.76,
|
198 |
+
"learning_rate": 9.162857142857144e-05,
|
199 |
+
"loss": 2.3541,
|
200 |
+
"step": 800
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 0.79,
|
204 |
+
"learning_rate": 9.091428571428572e-05,
|
205 |
+
"loss": 1.6534,
|
206 |
+
"step": 825
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 0.81,
|
210 |
+
"learning_rate": 9.020000000000001e-05,
|
211 |
+
"loss": 2.6717,
|
212 |
+
"step": 850
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.83,
|
216 |
+
"learning_rate": 8.948571428571429e-05,
|
217 |
+
"loss": 2.1232,
|
218 |
+
"step": 875
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 0.86,
|
222 |
+
"learning_rate": 8.877142857142857e-05,
|
223 |
+
"loss": 2.1955,
|
224 |
+
"step": 900
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 0.88,
|
228 |
+
"learning_rate": 8.805714285714286e-05,
|
229 |
+
"loss": 2.4756,
|
230 |
+
"step": 925
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 0.91,
|
234 |
+
"learning_rate": 8.734285714285714e-05,
|
235 |
+
"loss": 2.0444,
|
236 |
+
"step": 950
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 0.93,
|
240 |
+
"learning_rate": 8.662857142857144e-05,
|
241 |
+
"loss": 2.124,
|
242 |
+
"step": 975
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 0.95,
|
246 |
+
"learning_rate": 8.591428571428572e-05,
|
247 |
+
"loss": 1.7168,
|
248 |
+
"step": 1000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 0.95,
|
252 |
+
"eval_loss": 1.2106691598892212,
|
253 |
+
"eval_runtime": 380.7845,
|
254 |
+
"eval_samples_per_second": 3.939,
|
255 |
+
"eval_steps_per_second": 0.494,
|
256 |
+
"eval_wer": 64.80865224625624,
|
257 |
+
"step": 1000
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 0.98,
|
261 |
+
"learning_rate": 8.52e-05,
|
262 |
+
"loss": 2.451,
|
263 |
+
"step": 1025
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 1.0,
|
267 |
+
"learning_rate": 8.448571428571429e-05,
|
268 |
+
"loss": 1.4766,
|
269 |
+
"step": 1050
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 1.03,
|
273 |
+
"learning_rate": 8.377142857142858e-05,
|
274 |
+
"loss": 1.6965,
|
275 |
+
"step": 1075
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 1.05,
|
279 |
+
"learning_rate": 8.305714285714285e-05,
|
280 |
+
"loss": 1.4733,
|
281 |
+
"step": 1100
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 1.07,
|
285 |
+
"learning_rate": 8.234285714285714e-05,
|
286 |
+
"loss": 1.9818,
|
287 |
+
"step": 1125
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 1.1,
|
291 |
+
"learning_rate": 8.162857142857144e-05,
|
292 |
+
"loss": 2.0946,
|
293 |
+
"step": 1150
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 1.12,
|
297 |
+
"learning_rate": 8.091428571428572e-05,
|
298 |
+
"loss": 2.3333,
|
299 |
+
"step": 1175
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 1.15,
|
303 |
+
"learning_rate": 8.020000000000001e-05,
|
304 |
+
"loss": 1.7366,
|
305 |
+
"step": 1200
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 1.17,
|
309 |
+
"learning_rate": 7.948571428571429e-05,
|
310 |
+
"loss": 1.8155,
|
311 |
+
"step": 1225
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 1.19,
|
315 |
+
"learning_rate": 7.877142857142857e-05,
|
316 |
+
"loss": 2.1239,
|
317 |
+
"step": 1250
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 1.22,
|
321 |
+
"learning_rate": 7.805714285714286e-05,
|
322 |
+
"loss": 1.6473,
|
323 |
+
"step": 1275
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 1.24,
|
327 |
+
"learning_rate": 7.734285714285714e-05,
|
328 |
+
"loss": 1.1093,
|
329 |
+
"step": 1300
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"epoch": 1.26,
|
333 |
+
"learning_rate": 7.662857142857142e-05,
|
334 |
+
"loss": 1.5792,
|
335 |
+
"step": 1325
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 1.29,
|
339 |
+
"learning_rate": 7.591428571428572e-05,
|
340 |
+
"loss": 1.9768,
|
341 |
+
"step": 1350
|
342 |
+
},
|
343 |
+
{
|
344 |
+
"epoch": 1.31,
|
345 |
+
"learning_rate": 7.52e-05,
|
346 |
+
"loss": 1.5958,
|
347 |
+
"step": 1375
|
348 |
+
},
|
349 |
+
{
|
350 |
+
"epoch": 1.34,
|
351 |
+
"learning_rate": 7.448571428571429e-05,
|
352 |
+
"loss": 1.6682,
|
353 |
+
"step": 1400
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 1.36,
|
357 |
+
"learning_rate": 7.377142857142859e-05,
|
358 |
+
"loss": 1.6654,
|
359 |
+
"step": 1425
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 1.38,
|
363 |
+
"learning_rate": 7.305714285714285e-05,
|
364 |
+
"loss": 1.6971,
|
365 |
+
"step": 1450
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 1.41,
|
369 |
+
"learning_rate": 7.234285714285715e-05,
|
370 |
+
"loss": 1.5318,
|
371 |
+
"step": 1475
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 1.43,
|
375 |
+
"learning_rate": 7.162857142857144e-05,
|
376 |
+
"loss": 1.3164,
|
377 |
+
"step": 1500
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 1.46,
|
381 |
+
"learning_rate": 7.091428571428572e-05,
|
382 |
+
"loss": 2.14,
|
383 |
+
"step": 1525
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 1.48,
|
387 |
+
"learning_rate": 7.02e-05,
|
388 |
+
"loss": 1.9155,
|
389 |
+
"step": 1550
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 1.5,
|
393 |
+
"learning_rate": 6.948571428571429e-05,
|
394 |
+
"loss": 1.4583,
|
395 |
+
"step": 1575
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 1.53,
|
399 |
+
"learning_rate": 6.877142857142857e-05,
|
400 |
+
"loss": 1.5538,
|
401 |
+
"step": 1600
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 1.55,
|
405 |
+
"learning_rate": 6.805714285714287e-05,
|
406 |
+
"loss": 1.1043,
|
407 |
+
"step": 1625
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 1.57,
|
411 |
+
"learning_rate": 6.734285714285715e-05,
|
412 |
+
"loss": 1.8925,
|
413 |
+
"step": 1650
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 1.6,
|
417 |
+
"learning_rate": 6.662857142857143e-05,
|
418 |
+
"loss": 1.7625,
|
419 |
+
"step": 1675
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 1.62,
|
423 |
+
"learning_rate": 6.591428571428572e-05,
|
424 |
+
"loss": 1.082,
|
425 |
+
"step": 1700
|
426 |
+
},
|
427 |
+
{
|
428 |
+
"epoch": 1.65,
|
429 |
+
"learning_rate": 6.52e-05,
|
430 |
+
"loss": 1.3789,
|
431 |
+
"step": 1725
|
432 |
+
},
|
433 |
+
{
|
434 |
+
"epoch": 1.67,
|
435 |
+
"learning_rate": 6.448571428571429e-05,
|
436 |
+
"loss": 1.568,
|
437 |
+
"step": 1750
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 1.69,
|
441 |
+
"learning_rate": 6.377142857142857e-05,
|
442 |
+
"loss": 1.3194,
|
443 |
+
"step": 1775
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 1.72,
|
447 |
+
"learning_rate": 6.305714285714285e-05,
|
448 |
+
"loss": 1.3629,
|
449 |
+
"step": 1800
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 1.74,
|
453 |
+
"learning_rate": 6.234285714285715e-05,
|
454 |
+
"loss": 1.6437,
|
455 |
+
"step": 1825
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 1.77,
|
459 |
+
"learning_rate": 6.162857142857144e-05,
|
460 |
+
"loss": 1.0288,
|
461 |
+
"step": 1850
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 1.79,
|
465 |
+
"learning_rate": 6.091428571428571e-05,
|
466 |
+
"loss": 1.5278,
|
467 |
+
"step": 1875
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 1.81,
|
471 |
+
"learning_rate": 6.02e-05,
|
472 |
+
"loss": 1.2791,
|
473 |
+
"step": 1900
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 1.84,
|
477 |
+
"learning_rate": 5.9485714285714286e-05,
|
478 |
+
"loss": 1.6202,
|
479 |
+
"step": 1925
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 1.86,
|
483 |
+
"learning_rate": 5.877142857142858e-05,
|
484 |
+
"loss": 1.3762,
|
485 |
+
"step": 1950
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 1.88,
|
489 |
+
"learning_rate": 5.805714285714285e-05,
|
490 |
+
"loss": 1.8212,
|
491 |
+
"step": 1975
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 1.91,
|
495 |
+
"learning_rate": 5.734285714285714e-05,
|
496 |
+
"loss": 1.1073,
|
497 |
+
"step": 2000
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 1.91,
|
501 |
+
"eval_loss": 0.9890535473823547,
|
502 |
+
"eval_runtime": 356.3677,
|
503 |
+
"eval_samples_per_second": 4.209,
|
504 |
+
"eval_steps_per_second": 0.528,
|
505 |
+
"eval_wer": 53.001941209095946,
|
506 |
+
"step": 2000
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 1.93,
|
510 |
+
"learning_rate": 5.662857142857143e-05,
|
511 |
+
"loss": 1.5077,
|
512 |
+
"step": 2025
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 1.96,
|
516 |
+
"learning_rate": 5.591428571428572e-05,
|
517 |
+
"loss": 1.1343,
|
518 |
+
"step": 2050
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 1.98,
|
522 |
+
"learning_rate": 5.520000000000001e-05,
|
523 |
+
"loss": 1.1924,
|
524 |
+
"step": 2075
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 2.0,
|
528 |
+
"learning_rate": 5.448571428571429e-05,
|
529 |
+
"loss": 1.2554,
|
530 |
+
"step": 2100
|
531 |
+
},
|
532 |
+
{
|
533 |
+
"epoch": 2.03,
|
534 |
+
"learning_rate": 5.3771428571428574e-05,
|
535 |
+
"loss": 1.5992,
|
536 |
+
"step": 2125
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 2.05,
|
540 |
+
"learning_rate": 5.305714285714286e-05,
|
541 |
+
"loss": 0.9659,
|
542 |
+
"step": 2150
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 2.08,
|
546 |
+
"learning_rate": 5.234285714285715e-05,
|
547 |
+
"loss": 1.2035,
|
548 |
+
"step": 2175
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 2.1,
|
552 |
+
"learning_rate": 5.162857142857143e-05,
|
553 |
+
"loss": 1.3945,
|
554 |
+
"step": 2200
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 2.12,
|
558 |
+
"learning_rate": 5.0914285714285714e-05,
|
559 |
+
"loss": 1.0145,
|
560 |
+
"step": 2225
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 2.15,
|
564 |
+
"learning_rate": 5.02e-05,
|
565 |
+
"loss": 1.0097,
|
566 |
+
"step": 2250
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 2.17,
|
570 |
+
"learning_rate": 4.9485714285714294e-05,
|
571 |
+
"loss": 0.9375,
|
572 |
+
"step": 2275
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 2.19,
|
576 |
+
"learning_rate": 4.8771428571428574e-05,
|
577 |
+
"loss": 0.9557,
|
578 |
+
"step": 2300
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 2.22,
|
582 |
+
"learning_rate": 4.805714285714286e-05,
|
583 |
+
"loss": 0.9563,
|
584 |
+
"step": 2325
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 2.24,
|
588 |
+
"learning_rate": 4.734285714285715e-05,
|
589 |
+
"loss": 1.2328,
|
590 |
+
"step": 2350
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 2.27,
|
594 |
+
"learning_rate": 4.6628571428571434e-05,
|
595 |
+
"loss": 1.4048,
|
596 |
+
"step": 2375
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 2.29,
|
600 |
+
"learning_rate": 4.5914285714285714e-05,
|
601 |
+
"loss": 0.9962,
|
602 |
+
"step": 2400
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 2.31,
|
606 |
+
"learning_rate": 4.52e-05,
|
607 |
+
"loss": 1.4589,
|
608 |
+
"step": 2425
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 2.34,
|
612 |
+
"learning_rate": 4.448571428571429e-05,
|
613 |
+
"loss": 1.2766,
|
614 |
+
"step": 2450
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 2.36,
|
618 |
+
"learning_rate": 4.3771428571428574e-05,
|
619 |
+
"loss": 1.3044,
|
620 |
+
"step": 2475
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 2.39,
|
624 |
+
"learning_rate": 4.3057142857142854e-05,
|
625 |
+
"loss": 2.5598,
|
626 |
+
"step": 2500
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 2.41,
|
630 |
+
"learning_rate": 4.234285714285715e-05,
|
631 |
+
"loss": 1.2634,
|
632 |
+
"step": 2525
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 2.43,
|
636 |
+
"learning_rate": 4.162857142857143e-05,
|
637 |
+
"loss": 1.2135,
|
638 |
+
"step": 2550
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 2.46,
|
642 |
+
"learning_rate": 4.0914285714285715e-05,
|
643 |
+
"loss": 1.0847,
|
644 |
+
"step": 2575
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 2.48,
|
648 |
+
"learning_rate": 4.02e-05,
|
649 |
+
"loss": 0.8392,
|
650 |
+
"step": 2600
|
651 |
+
},
|
652 |
+
{
|
653 |
+
"epoch": 2.5,
|
654 |
+
"learning_rate": 3.948571428571429e-05,
|
655 |
+
"loss": 1.7331,
|
656 |
+
"step": 2625
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 2.53,
|
660 |
+
"learning_rate": 3.8771428571428575e-05,
|
661 |
+
"loss": 1.468,
|
662 |
+
"step": 2650
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 2.55,
|
666 |
+
"learning_rate": 3.8057142857142855e-05,
|
667 |
+
"loss": 0.9024,
|
668 |
+
"step": 2675
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 2.58,
|
672 |
+
"learning_rate": 3.734285714285715e-05,
|
673 |
+
"loss": 0.8736,
|
674 |
+
"step": 2700
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 2.6,
|
678 |
+
"learning_rate": 3.662857142857143e-05,
|
679 |
+
"loss": 0.8067,
|
680 |
+
"step": 2725
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 2.62,
|
684 |
+
"learning_rate": 3.5914285714285715e-05,
|
685 |
+
"loss": 1.2113,
|
686 |
+
"step": 2750
|
687 |
+
},
|
688 |
+
{
|
689 |
+
"epoch": 2.65,
|
690 |
+
"learning_rate": 3.52e-05,
|
691 |
+
"loss": 1.2159,
|
692 |
+
"step": 2775
|
693 |
+
},
|
694 |
+
{
|
695 |
+
"epoch": 2.67,
|
696 |
+
"learning_rate": 3.448571428571429e-05,
|
697 |
+
"loss": 0.9207,
|
698 |
+
"step": 2800
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 2.7,
|
702 |
+
"learning_rate": 3.377142857142857e-05,
|
703 |
+
"loss": 1.2301,
|
704 |
+
"step": 2825
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 2.72,
|
708 |
+
"learning_rate": 3.305714285714286e-05,
|
709 |
+
"loss": 0.8843,
|
710 |
+
"step": 2850
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 2.74,
|
714 |
+
"learning_rate": 3.234285714285715e-05,
|
715 |
+
"loss": 0.998,
|
716 |
+
"step": 2875
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 2.77,
|
720 |
+
"learning_rate": 3.162857142857143e-05,
|
721 |
+
"loss": 1.0264,
|
722 |
+
"step": 2900
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 2.79,
|
726 |
+
"learning_rate": 3.0914285714285715e-05,
|
727 |
+
"loss": 0.8172,
|
728 |
+
"step": 2925
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 2.81,
|
732 |
+
"learning_rate": 3.02e-05,
|
733 |
+
"loss": 1.0094,
|
734 |
+
"step": 2950
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 2.84,
|
738 |
+
"learning_rate": 2.948571428571429e-05,
|
739 |
+
"loss": 0.9068,
|
740 |
+
"step": 2975
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 2.86,
|
744 |
+
"learning_rate": 2.8771428571428572e-05,
|
745 |
+
"loss": 1.341,
|
746 |
+
"step": 3000
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 2.86,
|
750 |
+
"eval_loss": 0.8741958141326904,
|
751 |
+
"eval_runtime": 363.1523,
|
752 |
+
"eval_samples_per_second": 4.13,
|
753 |
+
"eval_steps_per_second": 0.518,
|
754 |
+
"eval_wer": 47.76760953965613,
|
755 |
+
"step": 3000
|
756 |
+
}
|
757 |
+
],
|
758 |
+
"max_steps": 4000,
|
759 |
+
"num_train_epochs": 4,
|
760 |
+
"total_flos": 1.18136182136832e+18,
|
761 |
+
"trial_name": null,
|
762 |
+
"trial_params": null
|
763 |
+
}
|
checkpoint-3000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e74aa035d76429a8a28d6b59538373cb3ba9a6d7d359ac8980aafba69d60558d
|
3 |
+
size 3503
|
checkpoint-4000/config.json
ADDED
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "openai/whisper-tiny",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"activation_function": "gelu",
|
5 |
+
"architectures": [
|
6 |
+
"WhisperForConditionalGeneration"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.1,
|
9 |
+
"begin_suppress_tokens": [
|
10 |
+
220,
|
11 |
+
50257
|
12 |
+
],
|
13 |
+
"bos_token_id": 50257,
|
14 |
+
"d_model": 384,
|
15 |
+
"decoder_attention_heads": 8,
|
16 |
+
"decoder_ffn_dim": 1536,
|
17 |
+
"decoder_layerdrop": 0.1,
|
18 |
+
"decoder_layers": 4,
|
19 |
+
"decoder_start_token_id": 50258,
|
20 |
+
"dropout": 0.0,
|
21 |
+
"encoder_attention_heads": 8,
|
22 |
+
"encoder_ffn_dim": 1536,
|
23 |
+
"encoder_layerdrop": 0.1,
|
24 |
+
"encoder_layers": 4,
|
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": 4,
|
34 |
+
"num_mel_bins": 80,
|
35 |
+
"pad_token_id": 50257,
|
36 |
+
"scale_embedding": false,
|
37 |
+
"suppress_tokens": [],
|
38 |
+
"torch_dtype": "float32",
|
39 |
+
"transformers_version": "4.26.0.dev0",
|
40 |
+
"use_cache": false,
|
41 |
+
"vocab_size": 51865
|
42 |
+
}
|
checkpoint-4000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d0fa4318310bfc98f73baeae2e684074bcd2eedc9827730328c7be9de0540346
|
3 |
+
size 302181925
|
checkpoint-4000/preprocessor_config.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
checkpoint-4000/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:26156c3c9531f4bca37959e6fbf907c4d6105220f84c4d905b490d6714c18153
|
3 |
+
size 151097331
|
checkpoint-4000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:53027c652b01daa057d590b1432c47adbde9da8c6a366c01e2f6b4d03321e0bc
|
3 |
+
size 14439
|
checkpoint-4000/scaler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a967ef13a7cce16d76f6083e59475c44d465576ce31bf6f085fc5bcd23744fec
|
3 |
+
size 559
|
checkpoint-4000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:21737518bca8588daadbc4aedd4f1035eeb4501b0d990fb3333d3f625118118b
|
3 |
+
size 623
|
checkpoint-4000/trainer_state.json
ADDED
@@ -0,0 +1,1012 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 46.277038269550744,
|
3 |
+
"best_model_checkpoint": "./whisper-tiny-it-6/checkpoint-4000",
|
4 |
+
"epoch": 3.816793893129771,
|
5 |
+
"global_step": 4000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.02,
|
12 |
+
"learning_rate": 3.8e-06,
|
13 |
+
"loss": 10.1581,
|
14 |
+
"step": 25
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.05,
|
18 |
+
"learning_rate": 8.599999999999999e-06,
|
19 |
+
"loss": 8.1216,
|
20 |
+
"step": 50
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.07,
|
24 |
+
"learning_rate": 1.3600000000000002e-05,
|
25 |
+
"loss": 6.9275,
|
26 |
+
"step": 75
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.1,
|
30 |
+
"learning_rate": 1.86e-05,
|
31 |
+
"loss": 5.6547,
|
32 |
+
"step": 100
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.12,
|
36 |
+
"learning_rate": 2.36e-05,
|
37 |
+
"loss": 5.2734,
|
38 |
+
"step": 125
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.14,
|
42 |
+
"learning_rate": 2.86e-05,
|
43 |
+
"loss": 5.5818,
|
44 |
+
"step": 150
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.17,
|
48 |
+
"learning_rate": 3.3600000000000004e-05,
|
49 |
+
"loss": 4.9292,
|
50 |
+
"step": 175
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.19,
|
54 |
+
"learning_rate": 3.86e-05,
|
55 |
+
"loss": 4.8341,
|
56 |
+
"step": 200
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.21,
|
60 |
+
"learning_rate": 4.36e-05,
|
61 |
+
"loss": 4.1192,
|
62 |
+
"step": 225
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.24,
|
66 |
+
"learning_rate": 4.86e-05,
|
67 |
+
"loss": 4.2109,
|
68 |
+
"step": 250
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.26,
|
72 |
+
"learning_rate": 5.360000000000001e-05,
|
73 |
+
"loss": 3.9181,
|
74 |
+
"step": 275
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.29,
|
78 |
+
"learning_rate": 5.86e-05,
|
79 |
+
"loss": 3.1145,
|
80 |
+
"step": 300
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.31,
|
84 |
+
"learning_rate": 6.36e-05,
|
85 |
+
"loss": 3.4049,
|
86 |
+
"step": 325
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.33,
|
90 |
+
"learning_rate": 6.860000000000001e-05,
|
91 |
+
"loss": 3.458,
|
92 |
+
"step": 350
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.36,
|
96 |
+
"learning_rate": 7.36e-05,
|
97 |
+
"loss": 3.6412,
|
98 |
+
"step": 375
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.38,
|
102 |
+
"learning_rate": 7.860000000000001e-05,
|
103 |
+
"loss": 3.0229,
|
104 |
+
"step": 400
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.41,
|
108 |
+
"learning_rate": 8.36e-05,
|
109 |
+
"loss": 2.9048,
|
110 |
+
"step": 425
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.43,
|
114 |
+
"learning_rate": 8.86e-05,
|
115 |
+
"loss": 2.8555,
|
116 |
+
"step": 450
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.45,
|
120 |
+
"learning_rate": 9.360000000000001e-05,
|
121 |
+
"loss": 2.4833,
|
122 |
+
"step": 475
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 0.48,
|
126 |
+
"learning_rate": 9.86e-05,
|
127 |
+
"loss": 2.7599,
|
128 |
+
"step": 500
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.5,
|
132 |
+
"learning_rate": 9.948571428571429e-05,
|
133 |
+
"loss": 2.8813,
|
134 |
+
"step": 525
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 0.52,
|
138 |
+
"learning_rate": 9.877142857142858e-05,
|
139 |
+
"loss": 2.6932,
|
140 |
+
"step": 550
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 0.55,
|
144 |
+
"learning_rate": 9.805714285714286e-05,
|
145 |
+
"loss": 2.7549,
|
146 |
+
"step": 575
|
147 |
+
},
|
148 |
+
{
|
149 |
+
"epoch": 0.57,
|
150 |
+
"learning_rate": 9.734285714285714e-05,
|
151 |
+
"loss": 2.7286,
|
152 |
+
"step": 600
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 0.6,
|
156 |
+
"learning_rate": 9.662857142857144e-05,
|
157 |
+
"loss": 2.1775,
|
158 |
+
"step": 625
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.62,
|
162 |
+
"learning_rate": 9.591428571428572e-05,
|
163 |
+
"loss": 2.4533,
|
164 |
+
"step": 650
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.64,
|
168 |
+
"learning_rate": 9.52e-05,
|
169 |
+
"loss": 3.0003,
|
170 |
+
"step": 675
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.67,
|
174 |
+
"learning_rate": 9.448571428571429e-05,
|
175 |
+
"loss": 2.6656,
|
176 |
+
"step": 700
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.69,
|
180 |
+
"learning_rate": 9.377142857142857e-05,
|
181 |
+
"loss": 2.4105,
|
182 |
+
"step": 725
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 0.72,
|
186 |
+
"learning_rate": 9.305714285714286e-05,
|
187 |
+
"loss": 2.2225,
|
188 |
+
"step": 750
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.74,
|
192 |
+
"learning_rate": 9.234285714285714e-05,
|
193 |
+
"loss": 2.6048,
|
194 |
+
"step": 775
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 0.76,
|
198 |
+
"learning_rate": 9.162857142857144e-05,
|
199 |
+
"loss": 2.3541,
|
200 |
+
"step": 800
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 0.79,
|
204 |
+
"learning_rate": 9.091428571428572e-05,
|
205 |
+
"loss": 1.6534,
|
206 |
+
"step": 825
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 0.81,
|
210 |
+
"learning_rate": 9.020000000000001e-05,
|
211 |
+
"loss": 2.6717,
|
212 |
+
"step": 850
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.83,
|
216 |
+
"learning_rate": 8.948571428571429e-05,
|
217 |
+
"loss": 2.1232,
|
218 |
+
"step": 875
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 0.86,
|
222 |
+
"learning_rate": 8.877142857142857e-05,
|
223 |
+
"loss": 2.1955,
|
224 |
+
"step": 900
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 0.88,
|
228 |
+
"learning_rate": 8.805714285714286e-05,
|
229 |
+
"loss": 2.4756,
|
230 |
+
"step": 925
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 0.91,
|
234 |
+
"learning_rate": 8.734285714285714e-05,
|
235 |
+
"loss": 2.0444,
|
236 |
+
"step": 950
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 0.93,
|
240 |
+
"learning_rate": 8.662857142857144e-05,
|
241 |
+
"loss": 2.124,
|
242 |
+
"step": 975
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 0.95,
|
246 |
+
"learning_rate": 8.591428571428572e-05,
|
247 |
+
"loss": 1.7168,
|
248 |
+
"step": 1000
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 0.95,
|
252 |
+
"eval_loss": 1.2106691598892212,
|
253 |
+
"eval_runtime": 380.7845,
|
254 |
+
"eval_samples_per_second": 3.939,
|
255 |
+
"eval_steps_per_second": 0.494,
|
256 |
+
"eval_wer": 64.80865224625624,
|
257 |
+
"step": 1000
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 0.98,
|
261 |
+
"learning_rate": 8.52e-05,
|
262 |
+
"loss": 2.451,
|
263 |
+
"step": 1025
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 1.0,
|
267 |
+
"learning_rate": 8.448571428571429e-05,
|
268 |
+
"loss": 1.4766,
|
269 |
+
"step": 1050
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 1.03,
|
273 |
+
"learning_rate": 8.377142857142858e-05,
|
274 |
+
"loss": 1.6965,
|
275 |
+
"step": 1075
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 1.05,
|
279 |
+
"learning_rate": 8.305714285714285e-05,
|
280 |
+
"loss": 1.4733,
|
281 |
+
"step": 1100
|
282 |
+
},
|
283 |
+
{
|
284 |
+
"epoch": 1.07,
|
285 |
+
"learning_rate": 8.234285714285714e-05,
|
286 |
+
"loss": 1.9818,
|
287 |
+
"step": 1125
|
288 |
+
},
|
289 |
+
{
|
290 |
+
"epoch": 1.1,
|
291 |
+
"learning_rate": 8.162857142857144e-05,
|
292 |
+
"loss": 2.0946,
|
293 |
+
"step": 1150
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 1.12,
|
297 |
+
"learning_rate": 8.091428571428572e-05,
|
298 |
+
"loss": 2.3333,
|
299 |
+
"step": 1175
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 1.15,
|
303 |
+
"learning_rate": 8.020000000000001e-05,
|
304 |
+
"loss": 1.7366,
|
305 |
+
"step": 1200
|
306 |
+
},
|
307 |
+
{
|
308 |
+
"epoch": 1.17,
|
309 |
+
"learning_rate": 7.948571428571429e-05,
|
310 |
+
"loss": 1.8155,
|
311 |
+
"step": 1225
|
312 |
+
},
|
313 |
+
{
|
314 |
+
"epoch": 1.19,
|
315 |
+
"learning_rate": 7.877142857142857e-05,
|
316 |
+
"loss": 2.1239,
|
317 |
+
"step": 1250
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 1.22,
|
321 |
+
"learning_rate": 7.805714285714286e-05,
|
322 |
+
"loss": 1.6473,
|
323 |
+
"step": 1275
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 1.24,
|
327 |
+
"learning_rate": 7.734285714285714e-05,
|
328 |
+
"loss": 1.1093,
|
329 |
+
"step": 1300
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"epoch": 1.26,
|
333 |
+
"learning_rate": 7.662857142857142e-05,
|
334 |
+
"loss": 1.5792,
|
335 |
+
"step": 1325
|
336 |
+
},
|
337 |
+
{
|
338 |
+
"epoch": 1.29,
|
339 |
+
"learning_rate": 7.591428571428572e-05,
|
340 |
+
"loss": 1.9768,
|
341 |
+
"step": 1350
|
342 |
+
},
|
343 |
+
{
|
344 |
+
"epoch": 1.31,
|
345 |
+
"learning_rate": 7.52e-05,
|
346 |
+
"loss": 1.5958,
|
347 |
+
"step": 1375
|
348 |
+
},
|
349 |
+
{
|
350 |
+
"epoch": 1.34,
|
351 |
+
"learning_rate": 7.448571428571429e-05,
|
352 |
+
"loss": 1.6682,
|
353 |
+
"step": 1400
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 1.36,
|
357 |
+
"learning_rate": 7.377142857142859e-05,
|
358 |
+
"loss": 1.6654,
|
359 |
+
"step": 1425
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 1.38,
|
363 |
+
"learning_rate": 7.305714285714285e-05,
|
364 |
+
"loss": 1.6971,
|
365 |
+
"step": 1450
|
366 |
+
},
|
367 |
+
{
|
368 |
+
"epoch": 1.41,
|
369 |
+
"learning_rate": 7.234285714285715e-05,
|
370 |
+
"loss": 1.5318,
|
371 |
+
"step": 1475
|
372 |
+
},
|
373 |
+
{
|
374 |
+
"epoch": 1.43,
|
375 |
+
"learning_rate": 7.162857142857144e-05,
|
376 |
+
"loss": 1.3164,
|
377 |
+
"step": 1500
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 1.46,
|
381 |
+
"learning_rate": 7.091428571428572e-05,
|
382 |
+
"loss": 2.14,
|
383 |
+
"step": 1525
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 1.48,
|
387 |
+
"learning_rate": 7.02e-05,
|
388 |
+
"loss": 1.9155,
|
389 |
+
"step": 1550
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 1.5,
|
393 |
+
"learning_rate": 6.948571428571429e-05,
|
394 |
+
"loss": 1.4583,
|
395 |
+
"step": 1575
|
396 |
+
},
|
397 |
+
{
|
398 |
+
"epoch": 1.53,
|
399 |
+
"learning_rate": 6.877142857142857e-05,
|
400 |
+
"loss": 1.5538,
|
401 |
+
"step": 1600
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 1.55,
|
405 |
+
"learning_rate": 6.805714285714287e-05,
|
406 |
+
"loss": 1.1043,
|
407 |
+
"step": 1625
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 1.57,
|
411 |
+
"learning_rate": 6.734285714285715e-05,
|
412 |
+
"loss": 1.8925,
|
413 |
+
"step": 1650
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 1.6,
|
417 |
+
"learning_rate": 6.662857142857143e-05,
|
418 |
+
"loss": 1.7625,
|
419 |
+
"step": 1675
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 1.62,
|
423 |
+
"learning_rate": 6.591428571428572e-05,
|
424 |
+
"loss": 1.082,
|
425 |
+
"step": 1700
|
426 |
+
},
|
427 |
+
{
|
428 |
+
"epoch": 1.65,
|
429 |
+
"learning_rate": 6.52e-05,
|
430 |
+
"loss": 1.3789,
|
431 |
+
"step": 1725
|
432 |
+
},
|
433 |
+
{
|
434 |
+
"epoch": 1.67,
|
435 |
+
"learning_rate": 6.448571428571429e-05,
|
436 |
+
"loss": 1.568,
|
437 |
+
"step": 1750
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 1.69,
|
441 |
+
"learning_rate": 6.377142857142857e-05,
|
442 |
+
"loss": 1.3194,
|
443 |
+
"step": 1775
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 1.72,
|
447 |
+
"learning_rate": 6.305714285714285e-05,
|
448 |
+
"loss": 1.3629,
|
449 |
+
"step": 1800
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 1.74,
|
453 |
+
"learning_rate": 6.234285714285715e-05,
|
454 |
+
"loss": 1.6437,
|
455 |
+
"step": 1825
|
456 |
+
},
|
457 |
+
{
|
458 |
+
"epoch": 1.77,
|
459 |
+
"learning_rate": 6.162857142857144e-05,
|
460 |
+
"loss": 1.0288,
|
461 |
+
"step": 1850
|
462 |
+
},
|
463 |
+
{
|
464 |
+
"epoch": 1.79,
|
465 |
+
"learning_rate": 6.091428571428571e-05,
|
466 |
+
"loss": 1.5278,
|
467 |
+
"step": 1875
|
468 |
+
},
|
469 |
+
{
|
470 |
+
"epoch": 1.81,
|
471 |
+
"learning_rate": 6.02e-05,
|
472 |
+
"loss": 1.2791,
|
473 |
+
"step": 1900
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 1.84,
|
477 |
+
"learning_rate": 5.9485714285714286e-05,
|
478 |
+
"loss": 1.6202,
|
479 |
+
"step": 1925
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 1.86,
|
483 |
+
"learning_rate": 5.877142857142858e-05,
|
484 |
+
"loss": 1.3762,
|
485 |
+
"step": 1950
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 1.88,
|
489 |
+
"learning_rate": 5.805714285714285e-05,
|
490 |
+
"loss": 1.8212,
|
491 |
+
"step": 1975
|
492 |
+
},
|
493 |
+
{
|
494 |
+
"epoch": 1.91,
|
495 |
+
"learning_rate": 5.734285714285714e-05,
|
496 |
+
"loss": 1.1073,
|
497 |
+
"step": 2000
|
498 |
+
},
|
499 |
+
{
|
500 |
+
"epoch": 1.91,
|
501 |
+
"eval_loss": 0.9890535473823547,
|
502 |
+
"eval_runtime": 356.3677,
|
503 |
+
"eval_samples_per_second": 4.209,
|
504 |
+
"eval_steps_per_second": 0.528,
|
505 |
+
"eval_wer": 53.001941209095946,
|
506 |
+
"step": 2000
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 1.93,
|
510 |
+
"learning_rate": 5.662857142857143e-05,
|
511 |
+
"loss": 1.5077,
|
512 |
+
"step": 2025
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 1.96,
|
516 |
+
"learning_rate": 5.591428571428572e-05,
|
517 |
+
"loss": 1.1343,
|
518 |
+
"step": 2050
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 1.98,
|
522 |
+
"learning_rate": 5.520000000000001e-05,
|
523 |
+
"loss": 1.1924,
|
524 |
+
"step": 2075
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 2.0,
|
528 |
+
"learning_rate": 5.448571428571429e-05,
|
529 |
+
"loss": 1.2554,
|
530 |
+
"step": 2100
|
531 |
+
},
|
532 |
+
{
|
533 |
+
"epoch": 2.03,
|
534 |
+
"learning_rate": 5.3771428571428574e-05,
|
535 |
+
"loss": 1.5992,
|
536 |
+
"step": 2125
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 2.05,
|
540 |
+
"learning_rate": 5.305714285714286e-05,
|
541 |
+
"loss": 0.9659,
|
542 |
+
"step": 2150
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 2.08,
|
546 |
+
"learning_rate": 5.234285714285715e-05,
|
547 |
+
"loss": 1.2035,
|
548 |
+
"step": 2175
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 2.1,
|
552 |
+
"learning_rate": 5.162857142857143e-05,
|
553 |
+
"loss": 1.3945,
|
554 |
+
"step": 2200
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 2.12,
|
558 |
+
"learning_rate": 5.0914285714285714e-05,
|
559 |
+
"loss": 1.0145,
|
560 |
+
"step": 2225
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 2.15,
|
564 |
+
"learning_rate": 5.02e-05,
|
565 |
+
"loss": 1.0097,
|
566 |
+
"step": 2250
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 2.17,
|
570 |
+
"learning_rate": 4.9485714285714294e-05,
|
571 |
+
"loss": 0.9375,
|
572 |
+
"step": 2275
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 2.19,
|
576 |
+
"learning_rate": 4.8771428571428574e-05,
|
577 |
+
"loss": 0.9557,
|
578 |
+
"step": 2300
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 2.22,
|
582 |
+
"learning_rate": 4.805714285714286e-05,
|
583 |
+
"loss": 0.9563,
|
584 |
+
"step": 2325
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 2.24,
|
588 |
+
"learning_rate": 4.734285714285715e-05,
|
589 |
+
"loss": 1.2328,
|
590 |
+
"step": 2350
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 2.27,
|
594 |
+
"learning_rate": 4.6628571428571434e-05,
|
595 |
+
"loss": 1.4048,
|
596 |
+
"step": 2375
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 2.29,
|
600 |
+
"learning_rate": 4.5914285714285714e-05,
|
601 |
+
"loss": 0.9962,
|
602 |
+
"step": 2400
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 2.31,
|
606 |
+
"learning_rate": 4.52e-05,
|
607 |
+
"loss": 1.4589,
|
608 |
+
"step": 2425
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 2.34,
|
612 |
+
"learning_rate": 4.448571428571429e-05,
|
613 |
+
"loss": 1.2766,
|
614 |
+
"step": 2450
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 2.36,
|
618 |
+
"learning_rate": 4.3771428571428574e-05,
|
619 |
+
"loss": 1.3044,
|
620 |
+
"step": 2475
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 2.39,
|
624 |
+
"learning_rate": 4.3057142857142854e-05,
|
625 |
+
"loss": 2.5598,
|
626 |
+
"step": 2500
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 2.41,
|
630 |
+
"learning_rate": 4.234285714285715e-05,
|
631 |
+
"loss": 1.2634,
|
632 |
+
"step": 2525
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 2.43,
|
636 |
+
"learning_rate": 4.162857142857143e-05,
|
637 |
+
"loss": 1.2135,
|
638 |
+
"step": 2550
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 2.46,
|
642 |
+
"learning_rate": 4.0914285714285715e-05,
|
643 |
+
"loss": 1.0847,
|
644 |
+
"step": 2575
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 2.48,
|
648 |
+
"learning_rate": 4.02e-05,
|
649 |
+
"loss": 0.8392,
|
650 |
+
"step": 2600
|
651 |
+
},
|
652 |
+
{
|
653 |
+
"epoch": 2.5,
|
654 |
+
"learning_rate": 3.948571428571429e-05,
|
655 |
+
"loss": 1.7331,
|
656 |
+
"step": 2625
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 2.53,
|
660 |
+
"learning_rate": 3.8771428571428575e-05,
|
661 |
+
"loss": 1.468,
|
662 |
+
"step": 2650
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 2.55,
|
666 |
+
"learning_rate": 3.8057142857142855e-05,
|
667 |
+
"loss": 0.9024,
|
668 |
+
"step": 2675
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 2.58,
|
672 |
+
"learning_rate": 3.734285714285715e-05,
|
673 |
+
"loss": 0.8736,
|
674 |
+
"step": 2700
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 2.6,
|
678 |
+
"learning_rate": 3.662857142857143e-05,
|
679 |
+
"loss": 0.8067,
|
680 |
+
"step": 2725
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 2.62,
|
684 |
+
"learning_rate": 3.5914285714285715e-05,
|
685 |
+
"loss": 1.2113,
|
686 |
+
"step": 2750
|
687 |
+
},
|
688 |
+
{
|
689 |
+
"epoch": 2.65,
|
690 |
+
"learning_rate": 3.52e-05,
|
691 |
+
"loss": 1.2159,
|
692 |
+
"step": 2775
|
693 |
+
},
|
694 |
+
{
|
695 |
+
"epoch": 2.67,
|
696 |
+
"learning_rate": 3.448571428571429e-05,
|
697 |
+
"loss": 0.9207,
|
698 |
+
"step": 2800
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 2.7,
|
702 |
+
"learning_rate": 3.377142857142857e-05,
|
703 |
+
"loss": 1.2301,
|
704 |
+
"step": 2825
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 2.72,
|
708 |
+
"learning_rate": 3.305714285714286e-05,
|
709 |
+
"loss": 0.8843,
|
710 |
+
"step": 2850
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 2.74,
|
714 |
+
"learning_rate": 3.234285714285715e-05,
|
715 |
+
"loss": 0.998,
|
716 |
+
"step": 2875
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 2.77,
|
720 |
+
"learning_rate": 3.162857142857143e-05,
|
721 |
+
"loss": 1.0264,
|
722 |
+
"step": 2900
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 2.79,
|
726 |
+
"learning_rate": 3.0914285714285715e-05,
|
727 |
+
"loss": 0.8172,
|
728 |
+
"step": 2925
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 2.81,
|
732 |
+
"learning_rate": 3.02e-05,
|
733 |
+
"loss": 1.0094,
|
734 |
+
"step": 2950
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 2.84,
|
738 |
+
"learning_rate": 2.948571428571429e-05,
|
739 |
+
"loss": 0.9068,
|
740 |
+
"step": 2975
|
741 |
+
},
|
742 |
+
{
|
743 |
+
"epoch": 2.86,
|
744 |
+
"learning_rate": 2.8771428571428572e-05,
|
745 |
+
"loss": 1.341,
|
746 |
+
"step": 3000
|
747 |
+
},
|
748 |
+
{
|
749 |
+
"epoch": 2.86,
|
750 |
+
"eval_loss": 0.8741958141326904,
|
751 |
+
"eval_runtime": 363.1523,
|
752 |
+
"eval_samples_per_second": 4.13,
|
753 |
+
"eval_steps_per_second": 0.518,
|
754 |
+
"eval_wer": 47.76760953965613,
|
755 |
+
"step": 3000
|
756 |
+
},
|
757 |
+
{
|
758 |
+
"epoch": 2.89,
|
759 |
+
"learning_rate": 2.805714285714286e-05,
|
760 |
+
"loss": 0.7517,
|
761 |
+
"step": 3025
|
762 |
+
},
|
763 |
+
{
|
764 |
+
"epoch": 2.91,
|
765 |
+
"learning_rate": 2.7342857142857142e-05,
|
766 |
+
"loss": 0.8435,
|
767 |
+
"step": 3050
|
768 |
+
},
|
769 |
+
{
|
770 |
+
"epoch": 2.93,
|
771 |
+
"learning_rate": 2.662857142857143e-05,
|
772 |
+
"loss": 1.2729,
|
773 |
+
"step": 3075
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 2.96,
|
777 |
+
"learning_rate": 2.5914285714285713e-05,
|
778 |
+
"loss": 1.0093,
|
779 |
+
"step": 3100
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 2.98,
|
783 |
+
"learning_rate": 2.5200000000000003e-05,
|
784 |
+
"loss": 1.7558,
|
785 |
+
"step": 3125
|
786 |
+
},
|
787 |
+
{
|
788 |
+
"epoch": 3.01,
|
789 |
+
"learning_rate": 2.4485714285714286e-05,
|
790 |
+
"loss": 1.2849,
|
791 |
+
"step": 3150
|
792 |
+
},
|
793 |
+
{
|
794 |
+
"epoch": 3.03,
|
795 |
+
"learning_rate": 2.3771428571428573e-05,
|
796 |
+
"loss": 0.7461,
|
797 |
+
"step": 3175
|
798 |
+
},
|
799 |
+
{
|
800 |
+
"epoch": 3.05,
|
801 |
+
"learning_rate": 2.3057142857142856e-05,
|
802 |
+
"loss": 0.7609,
|
803 |
+
"step": 3200
|
804 |
+
},
|
805 |
+
{
|
806 |
+
"epoch": 3.08,
|
807 |
+
"learning_rate": 2.2342857142857143e-05,
|
808 |
+
"loss": 1.0514,
|
809 |
+
"step": 3225
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 3.1,
|
813 |
+
"learning_rate": 2.162857142857143e-05,
|
814 |
+
"loss": 0.877,
|
815 |
+
"step": 3250
|
816 |
+
},
|
817 |
+
{
|
818 |
+
"epoch": 3.12,
|
819 |
+
"learning_rate": 2.0914285714285713e-05,
|
820 |
+
"loss": 1.0671,
|
821 |
+
"step": 3275
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 3.15,
|
825 |
+
"learning_rate": 2.0200000000000003e-05,
|
826 |
+
"loss": 0.474,
|
827 |
+
"step": 3300
|
828 |
+
},
|
829 |
+
{
|
830 |
+
"epoch": 3.17,
|
831 |
+
"learning_rate": 1.9485714285714286e-05,
|
832 |
+
"loss": 1.2102,
|
833 |
+
"step": 3325
|
834 |
+
},
|
835 |
+
{
|
836 |
+
"epoch": 3.2,
|
837 |
+
"learning_rate": 1.8771428571428573e-05,
|
838 |
+
"loss": 0.7779,
|
839 |
+
"step": 3350
|
840 |
+
},
|
841 |
+
{
|
842 |
+
"epoch": 3.22,
|
843 |
+
"learning_rate": 1.805714285714286e-05,
|
844 |
+
"loss": 0.9685,
|
845 |
+
"step": 3375
|
846 |
+
},
|
847 |
+
{
|
848 |
+
"epoch": 3.24,
|
849 |
+
"learning_rate": 1.7342857142857143e-05,
|
850 |
+
"loss": 1.103,
|
851 |
+
"step": 3400
|
852 |
+
},
|
853 |
+
{
|
854 |
+
"epoch": 3.27,
|
855 |
+
"learning_rate": 1.662857142857143e-05,
|
856 |
+
"loss": 1.1407,
|
857 |
+
"step": 3425
|
858 |
+
},
|
859 |
+
{
|
860 |
+
"epoch": 3.29,
|
861 |
+
"learning_rate": 1.5914285714285713e-05,
|
862 |
+
"loss": 1.0565,
|
863 |
+
"step": 3450
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 3.32,
|
867 |
+
"learning_rate": 1.52e-05,
|
868 |
+
"loss": 1.184,
|
869 |
+
"step": 3475
|
870 |
+
},
|
871 |
+
{
|
872 |
+
"epoch": 3.34,
|
873 |
+
"learning_rate": 1.4485714285714285e-05,
|
874 |
+
"loss": 0.8219,
|
875 |
+
"step": 3500
|
876 |
+
},
|
877 |
+
{
|
878 |
+
"epoch": 3.36,
|
879 |
+
"learning_rate": 1.3771428571428572e-05,
|
880 |
+
"loss": 1.1516,
|
881 |
+
"step": 3525
|
882 |
+
},
|
883 |
+
{
|
884 |
+
"epoch": 3.39,
|
885 |
+
"learning_rate": 1.3057142857142857e-05,
|
886 |
+
"loss": 0.8686,
|
887 |
+
"step": 3550
|
888 |
+
},
|
889 |
+
{
|
890 |
+
"epoch": 3.41,
|
891 |
+
"learning_rate": 1.2342857142857144e-05,
|
892 |
+
"loss": 0.9356,
|
893 |
+
"step": 3575
|
894 |
+
},
|
895 |
+
{
|
896 |
+
"epoch": 3.44,
|
897 |
+
"learning_rate": 1.1628571428571429e-05,
|
898 |
+
"loss": 0.9117,
|
899 |
+
"step": 3600
|
900 |
+
},
|
901 |
+
{
|
902 |
+
"epoch": 3.46,
|
903 |
+
"learning_rate": 1.0914285714285714e-05,
|
904 |
+
"loss": 0.7603,
|
905 |
+
"step": 3625
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 3.48,
|
909 |
+
"learning_rate": 1.02e-05,
|
910 |
+
"loss": 1.265,
|
911 |
+
"step": 3650
|
912 |
+
},
|
913 |
+
{
|
914 |
+
"epoch": 3.51,
|
915 |
+
"learning_rate": 9.485714285714287e-06,
|
916 |
+
"loss": 1.0381,
|
917 |
+
"step": 3675
|
918 |
+
},
|
919 |
+
{
|
920 |
+
"epoch": 3.53,
|
921 |
+
"learning_rate": 8.771428571428572e-06,
|
922 |
+
"loss": 1.1763,
|
923 |
+
"step": 3700
|
924 |
+
},
|
925 |
+
{
|
926 |
+
"epoch": 3.55,
|
927 |
+
"learning_rate": 8.057142857142857e-06,
|
928 |
+
"loss": 0.7898,
|
929 |
+
"step": 3725
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 3.58,
|
933 |
+
"learning_rate": 7.342857142857143e-06,
|
934 |
+
"loss": 0.9195,
|
935 |
+
"step": 3750
|
936 |
+
},
|
937 |
+
{
|
938 |
+
"epoch": 3.6,
|
939 |
+
"learning_rate": 6.628571428571428e-06,
|
940 |
+
"loss": 1.2321,
|
941 |
+
"step": 3775
|
942 |
+
},
|
943 |
+
{
|
944 |
+
"epoch": 3.63,
|
945 |
+
"learning_rate": 5.914285714285714e-06,
|
946 |
+
"loss": 0.6247,
|
947 |
+
"step": 3800
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 3.65,
|
951 |
+
"learning_rate": 5.2e-06,
|
952 |
+
"loss": 0.6894,
|
953 |
+
"step": 3825
|
954 |
+
},
|
955 |
+
{
|
956 |
+
"epoch": 3.67,
|
957 |
+
"learning_rate": 4.485714285714286e-06,
|
958 |
+
"loss": 0.7708,
|
959 |
+
"step": 3850
|
960 |
+
},
|
961 |
+
{
|
962 |
+
"epoch": 3.7,
|
963 |
+
"learning_rate": 3.7714285714285716e-06,
|
964 |
+
"loss": 0.9571,
|
965 |
+
"step": 3875
|
966 |
+
},
|
967 |
+
{
|
968 |
+
"epoch": 3.72,
|
969 |
+
"learning_rate": 3.0571428571428575e-06,
|
970 |
+
"loss": 0.8388,
|
971 |
+
"step": 3900
|
972 |
+
},
|
973 |
+
{
|
974 |
+
"epoch": 3.75,
|
975 |
+
"learning_rate": 2.342857142857143e-06,
|
976 |
+
"loss": 0.9865,
|
977 |
+
"step": 3925
|
978 |
+
},
|
979 |
+
{
|
980 |
+
"epoch": 3.77,
|
981 |
+
"learning_rate": 1.6285714285714286e-06,
|
982 |
+
"loss": 0.907,
|
983 |
+
"step": 3950
|
984 |
+
},
|
985 |
+
{
|
986 |
+
"epoch": 3.79,
|
987 |
+
"learning_rate": 9.142857142857144e-07,
|
988 |
+
"loss": 1.131,
|
989 |
+
"step": 3975
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 3.82,
|
993 |
+
"learning_rate": 2.0000000000000002e-07,
|
994 |
+
"loss": 0.6761,
|
995 |
+
"step": 4000
|
996 |
+
},
|
997 |
+
{
|
998 |
+
"epoch": 3.82,
|
999 |
+
"eval_loss": 0.8287678360939026,
|
1000 |
+
"eval_runtime": 358.3624,
|
1001 |
+
"eval_samples_per_second": 4.186,
|
1002 |
+
"eval_steps_per_second": 0.525,
|
1003 |
+
"eval_wer": 46.277038269550744,
|
1004 |
+
"step": 4000
|
1005 |
+
}
|
1006 |
+
],
|
1007 |
+
"max_steps": 4000,
|
1008 |
+
"num_train_epochs": 4,
|
1009 |
+
"total_flos": 1.57509165109248e+18,
|
1010 |
+
"trial_name": null,
|
1011 |
+
"trial_params": null
|
1012 |
+
}
|
checkpoint-4000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e74aa035d76429a8a28d6b59538373cb3ba9a6d7d359ac8980aafba69d60558d
|
3 |
+
size 3503
|