First commit
Browse files- config.json +52 -0
- generation_config.json +221 -0
- optimizer.pt +3 -0
- preprocessor_config.json +14 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1471 -0
- training_args.bin +3 -0
config.json
ADDED
@@ -0,0 +1,52 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "openai/whisper-large-v2",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"activation_function": "gelu",
|
5 |
+
"apply_spec_augment": false,
|
6 |
+
"architectures": [
|
7 |
+
"WhisperForConditionalGeneration"
|
8 |
+
],
|
9 |
+
"attention_dropout": 0.0,
|
10 |
+
"begin_suppress_tokens": [
|
11 |
+
220,
|
12 |
+
50257
|
13 |
+
],
|
14 |
+
"bos_token_id": 50257,
|
15 |
+
"classifier_proj_size": 256,
|
16 |
+
"d_model": 1280,
|
17 |
+
"decoder_attention_heads": 20,
|
18 |
+
"decoder_ffn_dim": 5120,
|
19 |
+
"decoder_layerdrop": 0.0,
|
20 |
+
"decoder_layers": 32,
|
21 |
+
"decoder_start_token_id": 50258,
|
22 |
+
"dropout": 0.0,
|
23 |
+
"encoder_attention_heads": 20,
|
24 |
+
"encoder_ffn_dim": 5120,
|
25 |
+
"encoder_layerdrop": 0.0,
|
26 |
+
"encoder_layers": 32,
|
27 |
+
"eos_token_id": 50257,
|
28 |
+
"forced_decoder_ids": null,
|
29 |
+
"init_std": 0.02,
|
30 |
+
"is_encoder_decoder": true,
|
31 |
+
"mask_feature_length": 10,
|
32 |
+
"mask_feature_min_masks": 0,
|
33 |
+
"mask_feature_prob": 0.0,
|
34 |
+
"mask_time_length": 10,
|
35 |
+
"mask_time_min_masks": 2,
|
36 |
+
"mask_time_prob": 0.05,
|
37 |
+
"max_length": 448,
|
38 |
+
"max_source_positions": 1500,
|
39 |
+
"max_target_positions": 448,
|
40 |
+
"median_filter_width": 7,
|
41 |
+
"model_type": "whisper",
|
42 |
+
"num_hidden_layers": 32,
|
43 |
+
"num_mel_bins": 80,
|
44 |
+
"pad_token_id": 50257,
|
45 |
+
"scale_embedding": false,
|
46 |
+
"suppress_tokens": [],
|
47 |
+
"torch_dtype": "float32",
|
48 |
+
"transformers_version": "4.31.0.dev0",
|
49 |
+
"use_cache": true,
|
50 |
+
"use_weighted_layer_sum": false,
|
51 |
+
"vocab_size": 51865
|
52 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,221 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"begin_suppress_tokens": [
|
3 |
+
220,
|
4 |
+
50257
|
5 |
+
],
|
6 |
+
"bos_token_id": 50257,
|
7 |
+
"decoder_start_token_id": 50258,
|
8 |
+
"eos_token_id": 50257,
|
9 |
+
"forced_decoder_ids": [
|
10 |
+
[
|
11 |
+
1,
|
12 |
+
null
|
13 |
+
],
|
14 |
+
[
|
15 |
+
2,
|
16 |
+
50359
|
17 |
+
]
|
18 |
+
],
|
19 |
+
"is_multilingual": true,
|
20 |
+
"lang_to_id": {
|
21 |
+
"<|af|>": 50327,
|
22 |
+
"<|am|>": 50334,
|
23 |
+
"<|ar|>": 50272,
|
24 |
+
"<|as|>": 50350,
|
25 |
+
"<|az|>": 50304,
|
26 |
+
"<|ba|>": 50355,
|
27 |
+
"<|be|>": 50330,
|
28 |
+
"<|bg|>": 50292,
|
29 |
+
"<|bn|>": 50302,
|
30 |
+
"<|bo|>": 50347,
|
31 |
+
"<|br|>": 50309,
|
32 |
+
"<|bs|>": 50315,
|
33 |
+
"<|ca|>": 50270,
|
34 |
+
"<|cs|>": 50283,
|
35 |
+
"<|cy|>": 50297,
|
36 |
+
"<|da|>": 50285,
|
37 |
+
"<|de|>": 50261,
|
38 |
+
"<|el|>": 50281,
|
39 |
+
"<|en|>": 50259,
|
40 |
+
"<|es|>": 50262,
|
41 |
+
"<|et|>": 50307,
|
42 |
+
"<|eu|>": 50310,
|
43 |
+
"<|fa|>": 50300,
|
44 |
+
"<|fi|>": 50277,
|
45 |
+
"<|fo|>": 50338,
|
46 |
+
"<|fr|>": 50265,
|
47 |
+
"<|gl|>": 50319,
|
48 |
+
"<|gu|>": 50333,
|
49 |
+
"<|haw|>": 50352,
|
50 |
+
"<|ha|>": 50354,
|
51 |
+
"<|he|>": 50279,
|
52 |
+
"<|hi|>": 50276,
|
53 |
+
"<|hr|>": 50291,
|
54 |
+
"<|ht|>": 50339,
|
55 |
+
"<|hu|>": 50286,
|
56 |
+
"<|hy|>": 50312,
|
57 |
+
"<|id|>": 50275,
|
58 |
+
"<|is|>": 50311,
|
59 |
+
"<|it|>": 50274,
|
60 |
+
"<|ja|>": 50266,
|
61 |
+
"<|jw|>": 50356,
|
62 |
+
"<|ka|>": 50329,
|
63 |
+
"<|kk|>": 50316,
|
64 |
+
"<|km|>": 50323,
|
65 |
+
"<|kn|>": 50306,
|
66 |
+
"<|ko|>": 50264,
|
67 |
+
"<|la|>": 50294,
|
68 |
+
"<|lb|>": 50345,
|
69 |
+
"<|ln|>": 50353,
|
70 |
+
"<|lo|>": 50336,
|
71 |
+
"<|lt|>": 50293,
|
72 |
+
"<|lv|>": 50301,
|
73 |
+
"<|mg|>": 50349,
|
74 |
+
"<|mi|>": 50295,
|
75 |
+
"<|mk|>": 50308,
|
76 |
+
"<|ml|>": 50296,
|
77 |
+
"<|mn|>": 50314,
|
78 |
+
"<|mr|>": 50320,
|
79 |
+
"<|ms|>": 50282,
|
80 |
+
"<|mt|>": 50343,
|
81 |
+
"<|my|>": 50346,
|
82 |
+
"<|ne|>": 50313,
|
83 |
+
"<|nl|>": 50271,
|
84 |
+
"<|nn|>": 50342,
|
85 |
+
"<|no|>": 50288,
|
86 |
+
"<|oc|>": 50328,
|
87 |
+
"<|pa|>": 50321,
|
88 |
+
"<|pl|>": 50269,
|
89 |
+
"<|ps|>": 50340,
|
90 |
+
"<|pt|>": 50267,
|
91 |
+
"<|ro|>": 50284,
|
92 |
+
"<|ru|>": 50263,
|
93 |
+
"<|sa|>": 50344,
|
94 |
+
"<|sd|>": 50332,
|
95 |
+
"<|si|>": 50322,
|
96 |
+
"<|sk|>": 50298,
|
97 |
+
"<|sl|>": 50305,
|
98 |
+
"<|sn|>": 50324,
|
99 |
+
"<|so|>": 50326,
|
100 |
+
"<|sq|>": 50317,
|
101 |
+
"<|sr|>": 50303,
|
102 |
+
"<|su|>": 50357,
|
103 |
+
"<|sv|>": 50273,
|
104 |
+
"<|sw|>": 50318,
|
105 |
+
"<|ta|>": 50287,
|
106 |
+
"<|te|>": 50299,
|
107 |
+
"<|tg|>": 50331,
|
108 |
+
"<|th|>": 50289,
|
109 |
+
"<|tk|>": 50341,
|
110 |
+
"<|tl|>": 50348,
|
111 |
+
"<|tr|>": 50268,
|
112 |
+
"<|tt|>": 50351,
|
113 |
+
"<|uk|>": 50280,
|
114 |
+
"<|ur|>": 50290,
|
115 |
+
"<|uz|>": 50337,
|
116 |
+
"<|vi|>": 50278,
|
117 |
+
"<|yi|>": 50335,
|
118 |
+
"<|yo|>": 50325,
|
119 |
+
"<|zh|>": 50260
|
120 |
+
},
|
121 |
+
"max_initial_timestamp_index": 1,
|
122 |
+
"max_length": 448,
|
123 |
+
"no_timestamps_token_id": 50363,
|
124 |
+
"pad_token_id": 50257,
|
125 |
+
"return_timestamps": false,
|
126 |
+
"suppress_tokens": [
|
127 |
+
1,
|
128 |
+
2,
|
129 |
+
7,
|
130 |
+
8,
|
131 |
+
9,
|
132 |
+
10,
|
133 |
+
14,
|
134 |
+
25,
|
135 |
+
26,
|
136 |
+
27,
|
137 |
+
28,
|
138 |
+
29,
|
139 |
+
31,
|
140 |
+
58,
|
141 |
+
59,
|
142 |
+
60,
|
143 |
+
61,
|
144 |
+
62,
|
145 |
+
63,
|
146 |
+
90,
|
147 |
+
91,
|
148 |
+
92,
|
149 |
+
93,
|
150 |
+
359,
|
151 |
+
503,
|
152 |
+
522,
|
153 |
+
542,
|
154 |
+
873,
|
155 |
+
893,
|
156 |
+
902,
|
157 |
+
918,
|
158 |
+
922,
|
159 |
+
931,
|
160 |
+
1350,
|
161 |
+
1853,
|
162 |
+
1982,
|
163 |
+
2460,
|
164 |
+
2627,
|
165 |
+
3246,
|
166 |
+
3253,
|
167 |
+
3268,
|
168 |
+
3536,
|
169 |
+
3846,
|
170 |
+
3961,
|
171 |
+
4183,
|
172 |
+
4667,
|
173 |
+
6585,
|
174 |
+
6647,
|
175 |
+
7273,
|
176 |
+
9061,
|
177 |
+
9383,
|
178 |
+
10428,
|
179 |
+
10929,
|
180 |
+
11938,
|
181 |
+
12033,
|
182 |
+
12331,
|
183 |
+
12562,
|
184 |
+
13793,
|
185 |
+
14157,
|
186 |
+
14635,
|
187 |
+
15265,
|
188 |
+
15618,
|
189 |
+
16553,
|
190 |
+
16604,
|
191 |
+
18362,
|
192 |
+
18956,
|
193 |
+
20075,
|
194 |
+
21675,
|
195 |
+
22520,
|
196 |
+
26130,
|
197 |
+
26161,
|
198 |
+
26435,
|
199 |
+
28279,
|
200 |
+
29464,
|
201 |
+
31650,
|
202 |
+
32302,
|
203 |
+
32470,
|
204 |
+
36865,
|
205 |
+
42863,
|
206 |
+
47425,
|
207 |
+
49870,
|
208 |
+
50254,
|
209 |
+
50258,
|
210 |
+
50358,
|
211 |
+
50359,
|
212 |
+
50360,
|
213 |
+
50361,
|
214 |
+
50362
|
215 |
+
],
|
216 |
+
"task_to_id": {
|
217 |
+
"transcribe": 50359,
|
218 |
+
"translate": 50358
|
219 |
+
},
|
220 |
+
"transformers_version": "4.31.0.dev0"
|
221 |
+
}
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c2ad1dd21ed1315593762d2c2942ed5366ada70b8a70e814cc8306a1a8572d09
|
3 |
+
size 3098923455
|
preprocessor_config.json
ADDED
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"chunk_length": 30,
|
3 |
+
"feature_extractor_type": "WhisperFeatureExtractor",
|
4 |
+
"feature_size": 80,
|
5 |
+
"hop_length": 160,
|
6 |
+
"n_fft": 400,
|
7 |
+
"n_samples": 480000,
|
8 |
+
"nb_max_frames": 3000,
|
9 |
+
"padding_side": "right",
|
10 |
+
"padding_value": 0.0,
|
11 |
+
"processor_class": "WhisperProcessor",
|
12 |
+
"return_attention_mask": false,
|
13 |
+
"sampling_rate": 16000
|
14 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5bf8f674335d4f60e3f988d396cf594efb7c3378abee0e4e371d2463990ea2eb
|
3 |
+
size 6173655608
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:66306c16ef060a017984efe93c510600dbe54c1a224178ad15d306e99d5b587d
|
3 |
+
size 14575
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:53c0cfb749124e5d3d5168ae84fde6dddc4d178a5c2c31affdb824296420cd59
|
3 |
+
size 627
|
trainer_state.json
ADDED
@@ -0,0 +1,1471 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 41.11150694691842,
|
3 |
+
"best_model_checkpoint": "whisper_large_v2_adam_8bit/checkpoint-485",
|
4 |
+
"epoch": 17.962962962962962,
|
5 |
+
"global_step": 485,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.19,
|
12 |
+
"learning_rate": 8.000000000000001e-07,
|
13 |
+
"loss": 1.0874,
|
14 |
+
"step": 5
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.19,
|
18 |
+
"eval_loss": 1.0718352794647217,
|
19 |
+
"eval_runtime": 305.4443,
|
20 |
+
"eval_samples_per_second": 0.262,
|
21 |
+
"eval_steps_per_second": 0.131,
|
22 |
+
"eval_wer": 54.82721767011044,
|
23 |
+
"step": 5
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.37,
|
27 |
+
"learning_rate": 1.8000000000000001e-06,
|
28 |
+
"loss": 1.0169,
|
29 |
+
"step": 10
|
30 |
+
},
|
31 |
+
{
|
32 |
+
"epoch": 0.37,
|
33 |
+
"eval_loss": 0.895382285118103,
|
34 |
+
"eval_runtime": 335.7721,
|
35 |
+
"eval_samples_per_second": 0.238,
|
36 |
+
"eval_steps_per_second": 0.119,
|
37 |
+
"eval_wer": 57.14285714285714,
|
38 |
+
"step": 10
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.56,
|
42 |
+
"learning_rate": 2.6e-06,
|
43 |
+
"loss": 0.8095,
|
44 |
+
"step": 15
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.56,
|
48 |
+
"eval_loss": 0.7291234135627747,
|
49 |
+
"eval_runtime": 341.1764,
|
50 |
+
"eval_samples_per_second": 0.234,
|
51 |
+
"eval_steps_per_second": 0.117,
|
52 |
+
"eval_wer": 59.707873174207336,
|
53 |
+
"step": 15
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 0.74,
|
57 |
+
"learning_rate": 3.6000000000000003e-06,
|
58 |
+
"loss": 0.6586,
|
59 |
+
"step": 20
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 0.74,
|
63 |
+
"eval_loss": 0.5784670114517212,
|
64 |
+
"eval_runtime": 208.1203,
|
65 |
+
"eval_samples_per_second": 0.384,
|
66 |
+
"eval_steps_per_second": 0.192,
|
67 |
+
"eval_wer": 83.07801923762024,
|
68 |
+
"step": 20
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.93,
|
72 |
+
"learning_rate": 4.600000000000001e-06,
|
73 |
+
"loss": 0.5849,
|
74 |
+
"step": 25
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.93,
|
78 |
+
"eval_loss": 0.5224391222000122,
|
79 |
+
"eval_runtime": 251.382,
|
80 |
+
"eval_samples_per_second": 0.318,
|
81 |
+
"eval_steps_per_second": 0.159,
|
82 |
+
"eval_wer": 91.73494834342715,
|
83 |
+
"step": 25
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 1.11,
|
87 |
+
"learning_rate": 5.600000000000001e-06,
|
88 |
+
"loss": 0.4623,
|
89 |
+
"step": 30
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 1.11,
|
93 |
+
"eval_loss": 0.48158684372901917,
|
94 |
+
"eval_runtime": 279.2357,
|
95 |
+
"eval_samples_per_second": 0.286,
|
96 |
+
"eval_steps_per_second": 0.143,
|
97 |
+
"eval_wer": 72.99608122550765,
|
98 |
+
"step": 30
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 1.3,
|
102 |
+
"learning_rate": 6.600000000000001e-06,
|
103 |
+
"loss": 0.4459,
|
104 |
+
"step": 35
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 1.3,
|
108 |
+
"eval_loss": 0.4286937713623047,
|
109 |
+
"eval_runtime": 232.3372,
|
110 |
+
"eval_samples_per_second": 0.344,
|
111 |
+
"eval_steps_per_second": 0.172,
|
112 |
+
"eval_wer": 76.0242251514072,
|
113 |
+
"step": 35
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 1.48,
|
117 |
+
"learning_rate": 7.600000000000001e-06,
|
118 |
+
"loss": 0.3496,
|
119 |
+
"step": 40
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 1.48,
|
123 |
+
"eval_loss": 0.4074183404445648,
|
124 |
+
"eval_runtime": 289.554,
|
125 |
+
"eval_samples_per_second": 0.276,
|
126 |
+
"eval_steps_per_second": 0.138,
|
127 |
+
"eval_wer": 48.6284289276808,
|
128 |
+
"step": 40
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 1.67,
|
132 |
+
"learning_rate": 8.6e-06,
|
133 |
+
"loss": 0.3964,
|
134 |
+
"step": 45
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 1.67,
|
138 |
+
"eval_loss": 0.3671049177646637,
|
139 |
+
"eval_runtime": 356.7998,
|
140 |
+
"eval_samples_per_second": 0.224,
|
141 |
+
"eval_steps_per_second": 0.112,
|
142 |
+
"eval_wer": 57.67723548272177,
|
143 |
+
"step": 45
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 1.85,
|
147 |
+
"learning_rate": 9.600000000000001e-06,
|
148 |
+
"loss": 0.3506,
|
149 |
+
"step": 50
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 1.85,
|
153 |
+
"eval_loss": 0.37803885340690613,
|
154 |
+
"eval_runtime": 338.1245,
|
155 |
+
"eval_samples_per_second": 0.237,
|
156 |
+
"eval_steps_per_second": 0.118,
|
157 |
+
"eval_wer": 49.76843605272533,
|
158 |
+
"step": 50
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 2.04,
|
162 |
+
"learning_rate": 9.96842105263158e-06,
|
163 |
+
"loss": 0.377,
|
164 |
+
"step": 55
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 2.04,
|
168 |
+
"eval_loss": 0.35833048820495605,
|
169 |
+
"eval_runtime": 344.9289,
|
170 |
+
"eval_samples_per_second": 0.232,
|
171 |
+
"eval_steps_per_second": 0.116,
|
172 |
+
"eval_wer": 56.21660135375846,
|
173 |
+
"step": 55
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 2.22,
|
177 |
+
"learning_rate": 9.915789473684211e-06,
|
178 |
+
"loss": 0.2766,
|
179 |
+
"step": 60
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 2.22,
|
183 |
+
"eval_loss": 0.37928158044815063,
|
184 |
+
"eval_runtime": 350.8056,
|
185 |
+
"eval_samples_per_second": 0.228,
|
186 |
+
"eval_steps_per_second": 0.114,
|
187 |
+
"eval_wer": 83.0423940149626,
|
188 |
+
"step": 60
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 2.41,
|
192 |
+
"learning_rate": 9.863157894736843e-06,
|
193 |
+
"loss": 0.2522,
|
194 |
+
"step": 65
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 2.41,
|
198 |
+
"eval_loss": 0.3648552894592285,
|
199 |
+
"eval_runtime": 343.605,
|
200 |
+
"eval_samples_per_second": 0.233,
|
201 |
+
"eval_steps_per_second": 0.116,
|
202 |
+
"eval_wer": 69.89668685429284,
|
203 |
+
"step": 65
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 2.59,
|
207 |
+
"learning_rate": 9.810526315789475e-06,
|
208 |
+
"loss": 0.2793,
|
209 |
+
"step": 70
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 2.59,
|
213 |
+
"eval_loss": 0.36223381757736206,
|
214 |
+
"eval_runtime": 325.8454,
|
215 |
+
"eval_samples_per_second": 0.246,
|
216 |
+
"eval_steps_per_second": 0.123,
|
217 |
+
"eval_wer": 48.37905236907731,
|
218 |
+
"step": 70
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 2.78,
|
222 |
+
"learning_rate": 9.757894736842106e-06,
|
223 |
+
"loss": 0.27,
|
224 |
+
"step": 75
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 2.78,
|
228 |
+
"eval_loss": 0.38781291246414185,
|
229 |
+
"eval_runtime": 331.2847,
|
230 |
+
"eval_samples_per_second": 0.241,
|
231 |
+
"eval_steps_per_second": 0.121,
|
232 |
+
"eval_wer": 48.09405058781617,
|
233 |
+
"step": 75
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 2.96,
|
237 |
+
"learning_rate": 9.715789473684212e-06,
|
238 |
+
"loss": 0.2624,
|
239 |
+
"step": 80
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 2.96,
|
243 |
+
"eval_loss": 0.38262635469436646,
|
244 |
+
"eval_runtime": 329.4177,
|
245 |
+
"eval_samples_per_second": 0.243,
|
246 |
+
"eval_steps_per_second": 0.121,
|
247 |
+
"eval_wer": 47.167794798717495,
|
248 |
+
"step": 80
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 3.15,
|
252 |
+
"learning_rate": 9.663157894736843e-06,
|
253 |
+
"loss": 0.225,
|
254 |
+
"step": 85
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 3.15,
|
258 |
+
"eval_loss": 0.38125258684158325,
|
259 |
+
"eval_runtime": 310.0949,
|
260 |
+
"eval_samples_per_second": 0.258,
|
261 |
+
"eval_steps_per_second": 0.129,
|
262 |
+
"eval_wer": 54.39971499821874,
|
263 |
+
"step": 85
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 3.33,
|
267 |
+
"learning_rate": 9.610526315789475e-06,
|
268 |
+
"loss": 0.2062,
|
269 |
+
"step": 90
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 3.33,
|
273 |
+
"eval_loss": 0.3962409198284149,
|
274 |
+
"eval_runtime": 299.3235,
|
275 |
+
"eval_samples_per_second": 0.267,
|
276 |
+
"eval_steps_per_second": 0.134,
|
277 |
+
"eval_wer": 72.92483078019238,
|
278 |
+
"step": 90
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 3.52,
|
282 |
+
"learning_rate": 9.557894736842107e-06,
|
283 |
+
"loss": 0.192,
|
284 |
+
"step": 95
|
285 |
+
},
|
286 |
+
{
|
287 |
+
"epoch": 3.52,
|
288 |
+
"eval_loss": 0.3863430619239807,
|
289 |
+
"eval_runtime": 332.6447,
|
290 |
+
"eval_samples_per_second": 0.24,
|
291 |
+
"eval_steps_per_second": 0.12,
|
292 |
+
"eval_wer": 59.17349483434271,
|
293 |
+
"step": 95
|
294 |
+
},
|
295 |
+
{
|
296 |
+
"epoch": 3.7,
|
297 |
+
"learning_rate": 9.505263157894738e-06,
|
298 |
+
"loss": 0.224,
|
299 |
+
"step": 100
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 3.7,
|
303 |
+
"eval_loss": 0.38708847761154175,
|
304 |
+
"eval_runtime": 330.6389,
|
305 |
+
"eval_samples_per_second": 0.242,
|
306 |
+
"eval_steps_per_second": 0.121,
|
307 |
+
"eval_wer": 54.39971499821874,
|
308 |
+
"step": 100
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 3.89,
|
312 |
+
"learning_rate": 9.452631578947368e-06,
|
313 |
+
"loss": 0.2127,
|
314 |
+
"step": 105
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 3.89,
|
318 |
+
"eval_loss": 0.41648903489112854,
|
319 |
+
"eval_runtime": 357.2142,
|
320 |
+
"eval_samples_per_second": 0.224,
|
321 |
+
"eval_steps_per_second": 0.112,
|
322 |
+
"eval_wer": 70.07481296758104,
|
323 |
+
"step": 105
|
324 |
+
},
|
325 |
+
{
|
326 |
+
"epoch": 4.07,
|
327 |
+
"learning_rate": 9.4e-06,
|
328 |
+
"loss": 0.2033,
|
329 |
+
"step": 110
|
330 |
+
},
|
331 |
+
{
|
332 |
+
"epoch": 4.07,
|
333 |
+
"eval_loss": 0.39047971367836,
|
334 |
+
"eval_runtime": 316.4665,
|
335 |
+
"eval_samples_per_second": 0.253,
|
336 |
+
"eval_steps_per_second": 0.126,
|
337 |
+
"eval_wer": 50.69469184182401,
|
338 |
+
"step": 110
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 4.26,
|
342 |
+
"learning_rate": 9.347368421052633e-06,
|
343 |
+
"loss": 0.1426,
|
344 |
+
"step": 115
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 4.26,
|
348 |
+
"eval_loss": 0.41405850648880005,
|
349 |
+
"eval_runtime": 326.6943,
|
350 |
+
"eval_samples_per_second": 0.245,
|
351 |
+
"eval_steps_per_second": 0.122,
|
352 |
+
"eval_wer": 44.53152832205201,
|
353 |
+
"step": 115
|
354 |
+
},
|
355 |
+
{
|
356 |
+
"epoch": 4.44,
|
357 |
+
"learning_rate": 9.294736842105265e-06,
|
358 |
+
"loss": 0.1737,
|
359 |
+
"step": 120
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 4.44,
|
363 |
+
"eval_loss": 0.4162759780883789,
|
364 |
+
"eval_runtime": 334.7907,
|
365 |
+
"eval_samples_per_second": 0.239,
|
366 |
+
"eval_steps_per_second": 0.119,
|
367 |
+
"eval_wer": 47.52404702529391,
|
368 |
+
"step": 120
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 4.63,
|
372 |
+
"learning_rate": 9.242105263157896e-06,
|
373 |
+
"loss": 0.1404,
|
374 |
+
"step": 125
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 4.63,
|
378 |
+
"eval_loss": 0.4179295599460602,
|
379 |
+
"eval_runtime": 313.7669,
|
380 |
+
"eval_samples_per_second": 0.255,
|
381 |
+
"eval_steps_per_second": 0.127,
|
382 |
+
"eval_wer": 47.951549697185605,
|
383 |
+
"step": 125
|
384 |
+
},
|
385 |
+
{
|
386 |
+
"epoch": 4.81,
|
387 |
+
"learning_rate": 9.189473684210526e-06,
|
388 |
+
"loss": 0.1738,
|
389 |
+
"step": 130
|
390 |
+
},
|
391 |
+
{
|
392 |
+
"epoch": 4.81,
|
393 |
+
"eval_loss": 0.4070858061313629,
|
394 |
+
"eval_runtime": 322.5875,
|
395 |
+
"eval_samples_per_second": 0.248,
|
396 |
+
"eval_steps_per_second": 0.124,
|
397 |
+
"eval_wer": 43.747773423583894,
|
398 |
+
"step": 130
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 5.0,
|
402 |
+
"learning_rate": 9.136842105263158e-06,
|
403 |
+
"loss": 0.1852,
|
404 |
+
"step": 135
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 5.0,
|
408 |
+
"eval_loss": 0.3889687657356262,
|
409 |
+
"eval_runtime": 331.8197,
|
410 |
+
"eval_samples_per_second": 0.241,
|
411 |
+
"eval_steps_per_second": 0.121,
|
412 |
+
"eval_wer": 65.94228713929462,
|
413 |
+
"step": 135
|
414 |
+
},
|
415 |
+
{
|
416 |
+
"epoch": 5.19,
|
417 |
+
"learning_rate": 9.08421052631579e-06,
|
418 |
+
"loss": 0.1111,
|
419 |
+
"step": 140
|
420 |
+
},
|
421 |
+
{
|
422 |
+
"epoch": 5.19,
|
423 |
+
"eval_loss": 0.4467521607875824,
|
424 |
+
"eval_runtime": 340.6568,
|
425 |
+
"eval_samples_per_second": 0.235,
|
426 |
+
"eval_steps_per_second": 0.117,
|
427 |
+
"eval_wer": 47.5952974706092,
|
428 |
+
"step": 140
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 5.37,
|
432 |
+
"learning_rate": 9.031578947368423e-06,
|
433 |
+
"loss": 0.1642,
|
434 |
+
"step": 145
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 5.37,
|
438 |
+
"eval_loss": 0.47300252318382263,
|
439 |
+
"eval_runtime": 330.5579,
|
440 |
+
"eval_samples_per_second": 0.242,
|
441 |
+
"eval_steps_per_second": 0.121,
|
442 |
+
"eval_wer": 53.152832205201285,
|
443 |
+
"step": 145
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 5.56,
|
447 |
+
"learning_rate": 8.978947368421055e-06,
|
448 |
+
"loss": 0.1552,
|
449 |
+
"step": 150
|
450 |
+
},
|
451 |
+
{
|
452 |
+
"epoch": 5.56,
|
453 |
+
"eval_loss": 0.44687384366989136,
|
454 |
+
"eval_runtime": 317.2285,
|
455 |
+
"eval_samples_per_second": 0.252,
|
456 |
+
"eval_steps_per_second": 0.126,
|
457 |
+
"eval_wer": 65.62166013537585,
|
458 |
+
"step": 150
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 5.74,
|
462 |
+
"learning_rate": 8.926315789473685e-06,
|
463 |
+
"loss": 0.1497,
|
464 |
+
"step": 155
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 5.74,
|
468 |
+
"eval_loss": 0.44148778915405273,
|
469 |
+
"eval_runtime": 315.1611,
|
470 |
+
"eval_samples_per_second": 0.254,
|
471 |
+
"eval_steps_per_second": 0.127,
|
472 |
+
"eval_wer": 48.05842536515853,
|
473 |
+
"step": 155
|
474 |
+
},
|
475 |
+
{
|
476 |
+
"epoch": 5.93,
|
477 |
+
"learning_rate": 8.873684210526316e-06,
|
478 |
+
"loss": 0.1419,
|
479 |
+
"step": 160
|
480 |
+
},
|
481 |
+
{
|
482 |
+
"epoch": 5.93,
|
483 |
+
"eval_loss": 0.4392489492893219,
|
484 |
+
"eval_runtime": 336.253,
|
485 |
+
"eval_samples_per_second": 0.238,
|
486 |
+
"eval_steps_per_second": 0.119,
|
487 |
+
"eval_wer": 43.9258995368721,
|
488 |
+
"step": 160
|
489 |
+
},
|
490 |
+
{
|
491 |
+
"epoch": 6.11,
|
492 |
+
"learning_rate": 8.821052631578948e-06,
|
493 |
+
"loss": 0.1176,
|
494 |
+
"step": 165
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 6.11,
|
498 |
+
"eval_loss": 0.44891494512557983,
|
499 |
+
"eval_runtime": 328.7909,
|
500 |
+
"eval_samples_per_second": 0.243,
|
501 |
+
"eval_steps_per_second": 0.122,
|
502 |
+
"eval_wer": 41.1827573922337,
|
503 |
+
"step": 165
|
504 |
+
},
|
505 |
+
{
|
506 |
+
"epoch": 6.3,
|
507 |
+
"learning_rate": 8.76842105263158e-06,
|
508 |
+
"loss": 0.0996,
|
509 |
+
"step": 170
|
510 |
+
},
|
511 |
+
{
|
512 |
+
"epoch": 6.3,
|
513 |
+
"eval_loss": 0.505344569683075,
|
514 |
+
"eval_runtime": 338.587,
|
515 |
+
"eval_samples_per_second": 0.236,
|
516 |
+
"eval_steps_per_second": 0.118,
|
517 |
+
"eval_wer": 46.81154257214107,
|
518 |
+
"step": 170
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 6.48,
|
522 |
+
"learning_rate": 8.715789473684211e-06,
|
523 |
+
"loss": 0.1376,
|
524 |
+
"step": 175
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 6.48,
|
528 |
+
"eval_loss": 0.48324888944625854,
|
529 |
+
"eval_runtime": 326.4874,
|
530 |
+
"eval_samples_per_second": 0.245,
|
531 |
+
"eval_steps_per_second": 0.123,
|
532 |
+
"eval_wer": 52.29782686141788,
|
533 |
+
"step": 175
|
534 |
+
},
|
535 |
+
{
|
536 |
+
"epoch": 6.67,
|
537 |
+
"learning_rate": 8.663157894736843e-06,
|
538 |
+
"loss": 0.1093,
|
539 |
+
"step": 180
|
540 |
+
},
|
541 |
+
{
|
542 |
+
"epoch": 6.67,
|
543 |
+
"eval_loss": 0.4419935643672943,
|
544 |
+
"eval_runtime": 340.888,
|
545 |
+
"eval_samples_per_second": 0.235,
|
546 |
+
"eval_steps_per_second": 0.117,
|
547 |
+
"eval_wer": 54.39971499821874,
|
548 |
+
"step": 180
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 6.85,
|
552 |
+
"learning_rate": 8.610526315789474e-06,
|
553 |
+
"loss": 0.1325,
|
554 |
+
"step": 185
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 6.85,
|
558 |
+
"eval_loss": 0.4463180899620056,
|
559 |
+
"eval_runtime": 333.0453,
|
560 |
+
"eval_samples_per_second": 0.24,
|
561 |
+
"eval_steps_per_second": 0.12,
|
562 |
+
"eval_wer": 45.92091200570004,
|
563 |
+
"step": 185
|
564 |
+
},
|
565 |
+
{
|
566 |
+
"epoch": 7.04,
|
567 |
+
"learning_rate": 8.557894736842106e-06,
|
568 |
+
"loss": 0.1151,
|
569 |
+
"step": 190
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 7.04,
|
573 |
+
"eval_loss": 0.4659479558467865,
|
574 |
+
"eval_runtime": 321.4982,
|
575 |
+
"eval_samples_per_second": 0.249,
|
576 |
+
"eval_steps_per_second": 0.124,
|
577 |
+
"eval_wer": 46.09903811898825,
|
578 |
+
"step": 190
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 7.22,
|
582 |
+
"learning_rate": 8.505263157894738e-06,
|
583 |
+
"loss": 0.0892,
|
584 |
+
"step": 195
|
585 |
+
},
|
586 |
+
{
|
587 |
+
"epoch": 7.22,
|
588 |
+
"eval_loss": 0.4642786979675293,
|
589 |
+
"eval_runtime": 304.1157,
|
590 |
+
"eval_samples_per_second": 0.263,
|
591 |
+
"eval_steps_per_second": 0.132,
|
592 |
+
"eval_wer": 61.9878874242964,
|
593 |
+
"step": 195
|
594 |
+
},
|
595 |
+
{
|
596 |
+
"epoch": 7.41,
|
597 |
+
"learning_rate": 8.45263157894737e-06,
|
598 |
+
"loss": 0.0819,
|
599 |
+
"step": 200
|
600 |
+
},
|
601 |
+
{
|
602 |
+
"epoch": 7.41,
|
603 |
+
"eval_loss": 0.4606548845767975,
|
604 |
+
"eval_runtime": 314.9466,
|
605 |
+
"eval_samples_per_second": 0.254,
|
606 |
+
"eval_steps_per_second": 0.127,
|
607 |
+
"eval_wer": 49.162807267545425,
|
608 |
+
"step": 200
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 7.59,
|
612 |
+
"learning_rate": 8.400000000000001e-06,
|
613 |
+
"loss": 0.0881,
|
614 |
+
"step": 205
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 7.59,
|
618 |
+
"eval_loss": 0.49518561363220215,
|
619 |
+
"eval_runtime": 333.9165,
|
620 |
+
"eval_samples_per_second": 0.24,
|
621 |
+
"eval_steps_per_second": 0.12,
|
622 |
+
"eval_wer": 47.167794798717495,
|
623 |
+
"step": 205
|
624 |
+
},
|
625 |
+
{
|
626 |
+
"epoch": 7.78,
|
627 |
+
"learning_rate": 8.347368421052633e-06,
|
628 |
+
"loss": 0.0902,
|
629 |
+
"step": 210
|
630 |
+
},
|
631 |
+
{
|
632 |
+
"epoch": 7.78,
|
633 |
+
"eval_loss": 0.48233914375305176,
|
634 |
+
"eval_runtime": 323.2853,
|
635 |
+
"eval_samples_per_second": 0.247,
|
636 |
+
"eval_steps_per_second": 0.124,
|
637 |
+
"eval_wer": 47.98717491984325,
|
638 |
+
"step": 210
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 7.96,
|
642 |
+
"learning_rate": 8.294736842105264e-06,
|
643 |
+
"loss": 0.1125,
|
644 |
+
"step": 215
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 7.96,
|
648 |
+
"eval_loss": 0.48526009917259216,
|
649 |
+
"eval_runtime": 331.099,
|
650 |
+
"eval_samples_per_second": 0.242,
|
651 |
+
"eval_steps_per_second": 0.121,
|
652 |
+
"eval_wer": 59.45849661560385,
|
653 |
+
"step": 215
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 8.15,
|
657 |
+
"learning_rate": 8.242105263157896e-06,
|
658 |
+
"loss": 0.0668,
|
659 |
+
"step": 220
|
660 |
+
},
|
661 |
+
{
|
662 |
+
"epoch": 8.15,
|
663 |
+
"eval_loss": 0.5275024175643921,
|
664 |
+
"eval_runtime": 310.2348,
|
665 |
+
"eval_samples_per_second": 0.258,
|
666 |
+
"eval_steps_per_second": 0.129,
|
667 |
+
"eval_wer": 58.42536515853224,
|
668 |
+
"step": 220
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 8.33,
|
672 |
+
"learning_rate": 8.189473684210527e-06,
|
673 |
+
"loss": 0.0723,
|
674 |
+
"step": 225
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 8.33,
|
678 |
+
"eval_loss": 0.532504141330719,
|
679 |
+
"eval_runtime": 331.9501,
|
680 |
+
"eval_samples_per_second": 0.241,
|
681 |
+
"eval_steps_per_second": 0.121,
|
682 |
+
"eval_wer": 46.775917349483436,
|
683 |
+
"step": 225
|
684 |
+
},
|
685 |
+
{
|
686 |
+
"epoch": 8.52,
|
687 |
+
"learning_rate": 8.136842105263159e-06,
|
688 |
+
"loss": 0.0825,
|
689 |
+
"step": 230
|
690 |
+
},
|
691 |
+
{
|
692 |
+
"epoch": 8.52,
|
693 |
+
"eval_loss": 0.5000892877578735,
|
694 |
+
"eval_runtime": 328.9744,
|
695 |
+
"eval_samples_per_second": 0.243,
|
696 |
+
"eval_steps_per_second": 0.122,
|
697 |
+
"eval_wer": 44.21090131813324,
|
698 |
+
"step": 230
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 8.7,
|
702 |
+
"learning_rate": 8.08421052631579e-06,
|
703 |
+
"loss": 0.0859,
|
704 |
+
"step": 235
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 8.7,
|
708 |
+
"eval_loss": 0.5018569827079773,
|
709 |
+
"eval_runtime": 321.0271,
|
710 |
+
"eval_samples_per_second": 0.249,
|
711 |
+
"eval_steps_per_second": 0.125,
|
712 |
+
"eval_wer": 55.68222301389384,
|
713 |
+
"step": 235
|
714 |
+
},
|
715 |
+
{
|
716 |
+
"epoch": 8.89,
|
717 |
+
"learning_rate": 8.03157894736842e-06,
|
718 |
+
"loss": 0.0838,
|
719 |
+
"step": 240
|
720 |
+
},
|
721 |
+
{
|
722 |
+
"epoch": 8.89,
|
723 |
+
"eval_loss": 0.5195188522338867,
|
724 |
+
"eval_runtime": 326.5564,
|
725 |
+
"eval_samples_per_second": 0.245,
|
726 |
+
"eval_steps_per_second": 0.122,
|
727 |
+
"eval_wer": 50.65906661916637,
|
728 |
+
"step": 240
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 9.07,
|
732 |
+
"learning_rate": 7.978947368421052e-06,
|
733 |
+
"loss": 0.0634,
|
734 |
+
"step": 245
|
735 |
+
},
|
736 |
+
{
|
737 |
+
"epoch": 9.07,
|
738 |
+
"eval_loss": 0.5121615529060364,
|
739 |
+
"eval_runtime": 333.3682,
|
740 |
+
"eval_samples_per_second": 0.24,
|
741 |
+
"eval_steps_per_second": 0.12,
|
742 |
+
"eval_wer": 45.42215888849305,
|
743 |
+
"step": 245
|
744 |
+
},
|
745 |
+
{
|
746 |
+
"epoch": 9.26,
|
747 |
+
"learning_rate": 7.926315789473686e-06,
|
748 |
+
"loss": 0.0611,
|
749 |
+
"step": 250
|
750 |
+
},
|
751 |
+
{
|
752 |
+
"epoch": 9.26,
|
753 |
+
"eval_loss": 0.53472900390625,
|
754 |
+
"eval_runtime": 334.7977,
|
755 |
+
"eval_samples_per_second": 0.239,
|
756 |
+
"eval_steps_per_second": 0.119,
|
757 |
+
"eval_wer": 45.38653366583541,
|
758 |
+
"step": 250
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 9.44,
|
762 |
+
"learning_rate": 7.873684210526317e-06,
|
763 |
+
"loss": 0.0742,
|
764 |
+
"step": 255
|
765 |
+
},
|
766 |
+
{
|
767 |
+
"epoch": 9.44,
|
768 |
+
"eval_loss": 0.5380967855453491,
|
769 |
+
"eval_runtime": 341.6365,
|
770 |
+
"eval_samples_per_second": 0.234,
|
771 |
+
"eval_steps_per_second": 0.117,
|
772 |
+
"eval_wer": 46.170288564303526,
|
773 |
+
"step": 255
|
774 |
+
},
|
775 |
+
{
|
776 |
+
"epoch": 9.63,
|
777 |
+
"learning_rate": 7.821052631578949e-06,
|
778 |
+
"loss": 0.0544,
|
779 |
+
"step": 260
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 9.63,
|
783 |
+
"eval_loss": 0.5325397253036499,
|
784 |
+
"eval_runtime": 344.44,
|
785 |
+
"eval_samples_per_second": 0.232,
|
786 |
+
"eval_steps_per_second": 0.116,
|
787 |
+
"eval_wer": 52.08407552547203,
|
788 |
+
"step": 260
|
789 |
+
},
|
790 |
+
{
|
791 |
+
"epoch": 9.81,
|
792 |
+
"learning_rate": 7.768421052631579e-06,
|
793 |
+
"loss": 0.0832,
|
794 |
+
"step": 265
|
795 |
+
},
|
796 |
+
{
|
797 |
+
"epoch": 9.81,
|
798 |
+
"eval_loss": 0.5298484563827515,
|
799 |
+
"eval_runtime": 338.7288,
|
800 |
+
"eval_samples_per_second": 0.236,
|
801 |
+
"eval_steps_per_second": 0.118,
|
802 |
+
"eval_wer": 46.81154257214107,
|
803 |
+
"step": 265
|
804 |
+
},
|
805 |
+
{
|
806 |
+
"epoch": 10.0,
|
807 |
+
"learning_rate": 7.71578947368421e-06,
|
808 |
+
"loss": 0.079,
|
809 |
+
"step": 270
|
810 |
+
},
|
811 |
+
{
|
812 |
+
"epoch": 10.0,
|
813 |
+
"eval_loss": 0.5149779319763184,
|
814 |
+
"eval_runtime": 322.201,
|
815 |
+
"eval_samples_per_second": 0.248,
|
816 |
+
"eval_steps_per_second": 0.124,
|
817 |
+
"eval_wer": 49.447809048806555,
|
818 |
+
"step": 270
|
819 |
+
},
|
820 |
+
{
|
821 |
+
"epoch": 10.19,
|
822 |
+
"learning_rate": 7.663157894736842e-06,
|
823 |
+
"loss": 0.0692,
|
824 |
+
"step": 275
|
825 |
+
},
|
826 |
+
{
|
827 |
+
"epoch": 10.19,
|
828 |
+
"eval_loss": 0.5358998775482178,
|
829 |
+
"eval_runtime": 304.2081,
|
830 |
+
"eval_samples_per_second": 0.263,
|
831 |
+
"eval_steps_per_second": 0.131,
|
832 |
+
"eval_wer": 50.3384396152476,
|
833 |
+
"step": 275
|
834 |
+
},
|
835 |
+
{
|
836 |
+
"epoch": 10.37,
|
837 |
+
"learning_rate": 7.610526315789474e-06,
|
838 |
+
"loss": 0.0549,
|
839 |
+
"step": 280
|
840 |
+
},
|
841 |
+
{
|
842 |
+
"epoch": 10.37,
|
843 |
+
"eval_loss": 0.5207683444023132,
|
844 |
+
"eval_runtime": 333.629,
|
845 |
+
"eval_samples_per_second": 0.24,
|
846 |
+
"eval_steps_per_second": 0.12,
|
847 |
+
"eval_wer": 49.12718204488778,
|
848 |
+
"step": 280
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 10.56,
|
852 |
+
"learning_rate": 7.557894736842106e-06,
|
853 |
+
"loss": 0.0533,
|
854 |
+
"step": 285
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 10.56,
|
858 |
+
"eval_loss": 0.528590202331543,
|
859 |
+
"eval_runtime": 332.7917,
|
860 |
+
"eval_samples_per_second": 0.24,
|
861 |
+
"eval_steps_per_second": 0.12,
|
862 |
+
"eval_wer": 47.02529390808692,
|
863 |
+
"step": 285
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 10.74,
|
867 |
+
"learning_rate": 7.505263157894738e-06,
|
868 |
+
"loss": 0.0821,
|
869 |
+
"step": 290
|
870 |
+
},
|
871 |
+
{
|
872 |
+
"epoch": 10.74,
|
873 |
+
"eval_loss": 0.51960289478302,
|
874 |
+
"eval_runtime": 332.8933,
|
875 |
+
"eval_samples_per_second": 0.24,
|
876 |
+
"eval_steps_per_second": 0.12,
|
877 |
+
"eval_wer": 46.24153900961881,
|
878 |
+
"step": 290
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 10.93,
|
882 |
+
"learning_rate": 7.4526315789473695e-06,
|
883 |
+
"loss": 0.0672,
|
884 |
+
"step": 295
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 10.93,
|
888 |
+
"eval_loss": 0.5139871835708618,
|
889 |
+
"eval_runtime": 342.8201,
|
890 |
+
"eval_samples_per_second": 0.233,
|
891 |
+
"eval_steps_per_second": 0.117,
|
892 |
+
"eval_wer": 46.45529034556466,
|
893 |
+
"step": 295
|
894 |
+
},
|
895 |
+
{
|
896 |
+
"epoch": 11.11,
|
897 |
+
"learning_rate": 7.4e-06,
|
898 |
+
"loss": 0.0694,
|
899 |
+
"step": 300
|
900 |
+
},
|
901 |
+
{
|
902 |
+
"epoch": 11.11,
|
903 |
+
"eval_loss": 0.5785766243934631,
|
904 |
+
"eval_runtime": 334.4012,
|
905 |
+
"eval_samples_per_second": 0.239,
|
906 |
+
"eval_steps_per_second": 0.12,
|
907 |
+
"eval_wer": 44.92340577128607,
|
908 |
+
"step": 300
|
909 |
+
},
|
910 |
+
{
|
911 |
+
"epoch": 11.3,
|
912 |
+
"learning_rate": 7.347368421052632e-06,
|
913 |
+
"loss": 0.09,
|
914 |
+
"step": 305
|
915 |
+
},
|
916 |
+
{
|
917 |
+
"epoch": 11.3,
|
918 |
+
"eval_loss": 0.584839940071106,
|
919 |
+
"eval_runtime": 343.2341,
|
920 |
+
"eval_samples_per_second": 0.233,
|
921 |
+
"eval_steps_per_second": 0.117,
|
922 |
+
"eval_wer": 49.59030993943712,
|
923 |
+
"step": 305
|
924 |
+
},
|
925 |
+
{
|
926 |
+
"epoch": 11.48,
|
927 |
+
"learning_rate": 7.2947368421052636e-06,
|
928 |
+
"loss": 0.0824,
|
929 |
+
"step": 310
|
930 |
+
},
|
931 |
+
{
|
932 |
+
"epoch": 11.48,
|
933 |
+
"eval_loss": 0.5461788773536682,
|
934 |
+
"eval_runtime": 330.7912,
|
935 |
+
"eval_samples_per_second": 0.242,
|
936 |
+
"eval_steps_per_second": 0.121,
|
937 |
+
"eval_wer": 45.24403277520484,
|
938 |
+
"step": 310
|
939 |
+
},
|
940 |
+
{
|
941 |
+
"epoch": 11.67,
|
942 |
+
"learning_rate": 7.242105263157896e-06,
|
943 |
+
"loss": 0.0805,
|
944 |
+
"step": 315
|
945 |
+
},
|
946 |
+
{
|
947 |
+
"epoch": 11.67,
|
948 |
+
"eval_loss": 0.537391185760498,
|
949 |
+
"eval_runtime": 352.6332,
|
950 |
+
"eval_samples_per_second": 0.227,
|
951 |
+
"eval_steps_per_second": 0.113,
|
952 |
+
"eval_wer": 54.25721410758817,
|
953 |
+
"step": 315
|
954 |
+
},
|
955 |
+
{
|
956 |
+
"epoch": 11.85,
|
957 |
+
"learning_rate": 7.189473684210527e-06,
|
958 |
+
"loss": 0.0688,
|
959 |
+
"step": 320
|
960 |
+
},
|
961 |
+
{
|
962 |
+
"epoch": 11.85,
|
963 |
+
"eval_loss": 0.5511082410812378,
|
964 |
+
"eval_runtime": 336.5436,
|
965 |
+
"eval_samples_per_second": 0.238,
|
966 |
+
"eval_steps_per_second": 0.119,
|
967 |
+
"eval_wer": 42.8927680798005,
|
968 |
+
"step": 320
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 12.04,
|
972 |
+
"learning_rate": 7.1368421052631585e-06,
|
973 |
+
"loss": 0.048,
|
974 |
+
"step": 325
|
975 |
+
},
|
976 |
+
{
|
977 |
+
"epoch": 12.04,
|
978 |
+
"eval_loss": 0.5636075139045715,
|
979 |
+
"eval_runtime": 319.3896,
|
980 |
+
"eval_samples_per_second": 0.25,
|
981 |
+
"eval_steps_per_second": 0.125,
|
982 |
+
"eval_wer": 44.81653010331315,
|
983 |
+
"step": 325
|
984 |
+
},
|
985 |
+
{
|
986 |
+
"epoch": 12.22,
|
987 |
+
"learning_rate": 7.08421052631579e-06,
|
988 |
+
"loss": 0.054,
|
989 |
+
"step": 330
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 12.22,
|
993 |
+
"eval_loss": 0.5698742866516113,
|
994 |
+
"eval_runtime": 345.1573,
|
995 |
+
"eval_samples_per_second": 0.232,
|
996 |
+
"eval_steps_per_second": 0.116,
|
997 |
+
"eval_wer": 46.88279301745636,
|
998 |
+
"step": 330
|
999 |
+
},
|
1000 |
+
{
|
1001 |
+
"epoch": 12.41,
|
1002 |
+
"learning_rate": 7.031578947368422e-06,
|
1003 |
+
"loss": 0.0535,
|
1004 |
+
"step": 335
|
1005 |
+
},
|
1006 |
+
{
|
1007 |
+
"epoch": 12.41,
|
1008 |
+
"eval_loss": 0.5786362886428833,
|
1009 |
+
"eval_runtime": 331.7186,
|
1010 |
+
"eval_samples_per_second": 0.241,
|
1011 |
+
"eval_steps_per_second": 0.121,
|
1012 |
+
"eval_wer": 42.60776629853937,
|
1013 |
+
"step": 335
|
1014 |
+
},
|
1015 |
+
{
|
1016 |
+
"epoch": 12.59,
|
1017 |
+
"learning_rate": 6.9789473684210525e-06,
|
1018 |
+
"loss": 0.0644,
|
1019 |
+
"step": 340
|
1020 |
+
},
|
1021 |
+
{
|
1022 |
+
"epoch": 12.59,
|
1023 |
+
"eval_loss": 0.5574811697006226,
|
1024 |
+
"eval_runtime": 339.8046,
|
1025 |
+
"eval_samples_per_second": 0.235,
|
1026 |
+
"eval_steps_per_second": 0.118,
|
1027 |
+
"eval_wer": 45.956537228357675,
|
1028 |
+
"step": 340
|
1029 |
+
},
|
1030 |
+
{
|
1031 |
+
"epoch": 12.78,
|
1032 |
+
"learning_rate": 6.926315789473684e-06,
|
1033 |
+
"loss": 0.0618,
|
1034 |
+
"step": 345
|
1035 |
+
},
|
1036 |
+
{
|
1037 |
+
"epoch": 12.78,
|
1038 |
+
"eval_loss": 0.5477833151817322,
|
1039 |
+
"eval_runtime": 337.626,
|
1040 |
+
"eval_samples_per_second": 0.237,
|
1041 |
+
"eval_steps_per_second": 0.118,
|
1042 |
+
"eval_wer": 46.918418240114,
|
1043 |
+
"step": 345
|
1044 |
+
},
|
1045 |
+
{
|
1046 |
+
"epoch": 12.96,
|
1047 |
+
"learning_rate": 6.873684210526317e-06,
|
1048 |
+
"loss": 0.0832,
|
1049 |
+
"step": 350
|
1050 |
+
},
|
1051 |
+
{
|
1052 |
+
"epoch": 12.96,
|
1053 |
+
"eval_loss": 0.6038811802864075,
|
1054 |
+
"eval_runtime": 336.2086,
|
1055 |
+
"eval_samples_per_second": 0.238,
|
1056 |
+
"eval_steps_per_second": 0.119,
|
1057 |
+
"eval_wer": 44.068400427502674,
|
1058 |
+
"step": 350
|
1059 |
+
},
|
1060 |
+
{
|
1061 |
+
"epoch": 13.15,
|
1062 |
+
"learning_rate": 6.821052631578948e-06,
|
1063 |
+
"loss": 0.0592,
|
1064 |
+
"step": 355
|
1065 |
+
},
|
1066 |
+
{
|
1067 |
+
"epoch": 13.15,
|
1068 |
+
"eval_loss": 0.6127667427062988,
|
1069 |
+
"eval_runtime": 331.5309,
|
1070 |
+
"eval_samples_per_second": 0.241,
|
1071 |
+
"eval_steps_per_second": 0.121,
|
1072 |
+
"eval_wer": 42.32276451727823,
|
1073 |
+
"step": 355
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 13.33,
|
1077 |
+
"learning_rate": 6.76842105263158e-06,
|
1078 |
+
"loss": 0.0619,
|
1079 |
+
"step": 360
|
1080 |
+
},
|
1081 |
+
{
|
1082 |
+
"epoch": 13.33,
|
1083 |
+
"eval_loss": 0.591279149055481,
|
1084 |
+
"eval_runtime": 332.3235,
|
1085 |
+
"eval_samples_per_second": 0.241,
|
1086 |
+
"eval_steps_per_second": 0.12,
|
1087 |
+
"eval_wer": 45.03028143925899,
|
1088 |
+
"step": 360
|
1089 |
+
},
|
1090 |
+
{
|
1091 |
+
"epoch": 13.52,
|
1092 |
+
"learning_rate": 6.715789473684211e-06,
|
1093 |
+
"loss": 0.0998,
|
1094 |
+
"step": 365
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 13.52,
|
1098 |
+
"eval_loss": 0.5318233966827393,
|
1099 |
+
"eval_runtime": 332.2782,
|
1100 |
+
"eval_samples_per_second": 0.241,
|
1101 |
+
"eval_steps_per_second": 0.12,
|
1102 |
+
"eval_wer": 42.75026718916993,
|
1103 |
+
"step": 365
|
1104 |
+
},
|
1105 |
+
{
|
1106 |
+
"epoch": 13.7,
|
1107 |
+
"learning_rate": 6.663157894736842e-06,
|
1108 |
+
"loss": 0.05,
|
1109 |
+
"step": 370
|
1110 |
+
},
|
1111 |
+
{
|
1112 |
+
"epoch": 13.7,
|
1113 |
+
"eval_loss": 0.5774287581443787,
|
1114 |
+
"eval_runtime": 329.9705,
|
1115 |
+
"eval_samples_per_second": 0.242,
|
1116 |
+
"eval_steps_per_second": 0.121,
|
1117 |
+
"eval_wer": 42.82151763448522,
|
1118 |
+
"step": 370
|
1119 |
+
},
|
1120 |
+
{
|
1121 |
+
"epoch": 13.89,
|
1122 |
+
"learning_rate": 6.610526315789474e-06,
|
1123 |
+
"loss": 0.1014,
|
1124 |
+
"step": 375
|
1125 |
+
},
|
1126 |
+
{
|
1127 |
+
"epoch": 13.89,
|
1128 |
+
"eval_loss": 0.5849052667617798,
|
1129 |
+
"eval_runtime": 328.3008,
|
1130 |
+
"eval_samples_per_second": 0.244,
|
1131 |
+
"eval_steps_per_second": 0.122,
|
1132 |
+
"eval_wer": 48.05842536515853,
|
1133 |
+
"step": 375
|
1134 |
+
},
|
1135 |
+
{
|
1136 |
+
"epoch": 14.07,
|
1137 |
+
"learning_rate": 6.557894736842106e-06,
|
1138 |
+
"loss": 0.0494,
|
1139 |
+
"step": 380
|
1140 |
+
},
|
1141 |
+
{
|
1142 |
+
"epoch": 14.07,
|
1143 |
+
"eval_loss": 0.5518860816955566,
|
1144 |
+
"eval_runtime": 318.5974,
|
1145 |
+
"eval_samples_per_second": 0.251,
|
1146 |
+
"eval_steps_per_second": 0.126,
|
1147 |
+
"eval_wer": 48.69967937299608,
|
1148 |
+
"step": 380
|
1149 |
+
},
|
1150 |
+
{
|
1151 |
+
"epoch": 14.26,
|
1152 |
+
"learning_rate": 6.505263157894738e-06,
|
1153 |
+
"loss": 0.0456,
|
1154 |
+
"step": 385
|
1155 |
+
},
|
1156 |
+
{
|
1157 |
+
"epoch": 14.26,
|
1158 |
+
"eval_loss": 0.5957409739494324,
|
1159 |
+
"eval_runtime": 335.9574,
|
1160 |
+
"eval_samples_per_second": 0.238,
|
1161 |
+
"eval_steps_per_second": 0.119,
|
1162 |
+
"eval_wer": 42.75026718916993,
|
1163 |
+
"step": 385
|
1164 |
+
},
|
1165 |
+
{
|
1166 |
+
"epoch": 14.44,
|
1167 |
+
"learning_rate": 6.452631578947369e-06,
|
1168 |
+
"loss": 0.0957,
|
1169 |
+
"step": 390
|
1170 |
+
},
|
1171 |
+
{
|
1172 |
+
"epoch": 14.44,
|
1173 |
+
"eval_loss": 0.5850934982299805,
|
1174 |
+
"eval_runtime": 347.6573,
|
1175 |
+
"eval_samples_per_second": 0.23,
|
1176 |
+
"eval_steps_per_second": 0.115,
|
1177 |
+
"eval_wer": 48.80655504096901,
|
1178 |
+
"step": 390
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 14.63,
|
1182 |
+
"learning_rate": 6.4000000000000006e-06,
|
1183 |
+
"loss": 0.0514,
|
1184 |
+
"step": 395
|
1185 |
+
},
|
1186 |
+
{
|
1187 |
+
"epoch": 14.63,
|
1188 |
+
"eval_loss": 0.6157152652740479,
|
1189 |
+
"eval_runtime": 332.3513,
|
1190 |
+
"eval_samples_per_second": 0.241,
|
1191 |
+
"eval_steps_per_second": 0.12,
|
1192 |
+
"eval_wer": 49.661560384752406,
|
1193 |
+
"step": 395
|
1194 |
+
},
|
1195 |
+
{
|
1196 |
+
"epoch": 14.81,
|
1197 |
+
"learning_rate": 6.347368421052632e-06,
|
1198 |
+
"loss": 0.1045,
|
1199 |
+
"step": 400
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 14.81,
|
1203 |
+
"eval_loss": 0.5892783999443054,
|
1204 |
+
"eval_runtime": 341.0177,
|
1205 |
+
"eval_samples_per_second": 0.235,
|
1206 |
+
"eval_steps_per_second": 0.117,
|
1207 |
+
"eval_wer": 52.6540790879943,
|
1208 |
+
"step": 400
|
1209 |
+
},
|
1210 |
+
{
|
1211 |
+
"epoch": 15.0,
|
1212 |
+
"learning_rate": 6.294736842105264e-06,
|
1213 |
+
"loss": 0.0485,
|
1214 |
+
"step": 405
|
1215 |
+
},
|
1216 |
+
{
|
1217 |
+
"epoch": 15.0,
|
1218 |
+
"eval_loss": 0.5818936228752136,
|
1219 |
+
"eval_runtime": 334.6692,
|
1220 |
+
"eval_samples_per_second": 0.239,
|
1221 |
+
"eval_steps_per_second": 0.12,
|
1222 |
+
"eval_wer": 50.9440684004275,
|
1223 |
+
"step": 405
|
1224 |
+
},
|
1225 |
+
{
|
1226 |
+
"epoch": 15.19,
|
1227 |
+
"learning_rate": 6.242105263157895e-06,
|
1228 |
+
"loss": 0.0705,
|
1229 |
+
"step": 410
|
1230 |
+
},
|
1231 |
+
{
|
1232 |
+
"epoch": 15.19,
|
1233 |
+
"eval_loss": 0.5869933366775513,
|
1234 |
+
"eval_runtime": 338.9314,
|
1235 |
+
"eval_samples_per_second": 0.236,
|
1236 |
+
"eval_steps_per_second": 0.118,
|
1237 |
+
"eval_wer": 45.77841111506947,
|
1238 |
+
"step": 410
|
1239 |
+
},
|
1240 |
+
{
|
1241 |
+
"epoch": 15.37,
|
1242 |
+
"learning_rate": 6.189473684210526e-06,
|
1243 |
+
"loss": 0.0354,
|
1244 |
+
"step": 415
|
1245 |
+
},
|
1246 |
+
{
|
1247 |
+
"epoch": 15.37,
|
1248 |
+
"eval_loss": 0.620836615562439,
|
1249 |
+
"eval_runtime": 340.9346,
|
1250 |
+
"eval_samples_per_second": 0.235,
|
1251 |
+
"eval_steps_per_second": 0.117,
|
1252 |
+
"eval_wer": 47.20342002137513,
|
1253 |
+
"step": 415
|
1254 |
+
},
|
1255 |
+
{
|
1256 |
+
"epoch": 15.56,
|
1257 |
+
"learning_rate": 6.136842105263159e-06,
|
1258 |
+
"loss": 0.0366,
|
1259 |
+
"step": 420
|
1260 |
+
},
|
1261 |
+
{
|
1262 |
+
"epoch": 15.56,
|
1263 |
+
"eval_loss": 0.6018834114074707,
|
1264 |
+
"eval_runtime": 339.0784,
|
1265 |
+
"eval_samples_per_second": 0.236,
|
1266 |
+
"eval_steps_per_second": 0.118,
|
1267 |
+
"eval_wer": 46.02778767367296,
|
1268 |
+
"step": 420
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 15.74,
|
1272 |
+
"learning_rate": 6.08421052631579e-06,
|
1273 |
+
"loss": 0.0422,
|
1274 |
+
"step": 425
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 15.74,
|
1278 |
+
"eval_loss": 0.5831278562545776,
|
1279 |
+
"eval_runtime": 332.3516,
|
1280 |
+
"eval_samples_per_second": 0.241,
|
1281 |
+
"eval_steps_per_second": 0.12,
|
1282 |
+
"eval_wer": 44.92340577128607,
|
1283 |
+
"step": 425
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 15.93,
|
1287 |
+
"learning_rate": 6.031578947368422e-06,
|
1288 |
+
"loss": 0.1133,
|
1289 |
+
"step": 430
|
1290 |
+
},
|
1291 |
+
{
|
1292 |
+
"epoch": 15.93,
|
1293 |
+
"eval_loss": 0.6038104295730591,
|
1294 |
+
"eval_runtime": 333.9842,
|
1295 |
+
"eval_samples_per_second": 0.24,
|
1296 |
+
"eval_steps_per_second": 0.12,
|
1297 |
+
"eval_wer": 44.13965087281795,
|
1298 |
+
"step": 430
|
1299 |
+
},
|
1300 |
+
{
|
1301 |
+
"epoch": 16.11,
|
1302 |
+
"learning_rate": 5.978947368421053e-06,
|
1303 |
+
"loss": 0.0515,
|
1304 |
+
"step": 435
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 16.11,
|
1308 |
+
"eval_loss": 0.6272233724594116,
|
1309 |
+
"eval_runtime": 331.7109,
|
1310 |
+
"eval_samples_per_second": 0.241,
|
1311 |
+
"eval_steps_per_second": 0.121,
|
1312 |
+
"eval_wer": 43.53402208763805,
|
1313 |
+
"step": 435
|
1314 |
+
},
|
1315 |
+
{
|
1316 |
+
"epoch": 16.3,
|
1317 |
+
"learning_rate": 5.9263157894736844e-06,
|
1318 |
+
"loss": 0.0709,
|
1319 |
+
"step": 440
|
1320 |
+
},
|
1321 |
+
{
|
1322 |
+
"epoch": 16.3,
|
1323 |
+
"eval_loss": 0.6419104337692261,
|
1324 |
+
"eval_runtime": 336.7239,
|
1325 |
+
"eval_samples_per_second": 0.238,
|
1326 |
+
"eval_steps_per_second": 0.119,
|
1327 |
+
"eval_wer": 45.92091200570004,
|
1328 |
+
"step": 440
|
1329 |
+
},
|
1330 |
+
{
|
1331 |
+
"epoch": 16.48,
|
1332 |
+
"learning_rate": 5.873684210526316e-06,
|
1333 |
+
"loss": 0.0978,
|
1334 |
+
"step": 445
|
1335 |
+
},
|
1336 |
+
{
|
1337 |
+
"epoch": 16.48,
|
1338 |
+
"eval_loss": 0.6428846120834351,
|
1339 |
+
"eval_runtime": 334.722,
|
1340 |
+
"eval_samples_per_second": 0.239,
|
1341 |
+
"eval_steps_per_second": 0.12,
|
1342 |
+
"eval_wer": 46.59779123619522,
|
1343 |
+
"step": 445
|
1344 |
+
},
|
1345 |
+
{
|
1346 |
+
"epoch": 16.67,
|
1347 |
+
"learning_rate": 5.8210526315789486e-06,
|
1348 |
+
"loss": 0.052,
|
1349 |
+
"step": 450
|
1350 |
+
},
|
1351 |
+
{
|
1352 |
+
"epoch": 16.67,
|
1353 |
+
"eval_loss": 0.6158877611160278,
|
1354 |
+
"eval_runtime": 340.1528,
|
1355 |
+
"eval_samples_per_second": 0.235,
|
1356 |
+
"eval_steps_per_second": 0.118,
|
1357 |
+
"eval_wer": 45.2084075525472,
|
1358 |
+
"step": 450
|
1359 |
+
},
|
1360 |
+
{
|
1361 |
+
"epoch": 16.85,
|
1362 |
+
"learning_rate": 5.76842105263158e-06,
|
1363 |
+
"loss": 0.0508,
|
1364 |
+
"step": 455
|
1365 |
+
},
|
1366 |
+
{
|
1367 |
+
"epoch": 16.85,
|
1368 |
+
"eval_loss": 0.631538987159729,
|
1369 |
+
"eval_runtime": 334.4872,
|
1370 |
+
"eval_samples_per_second": 0.239,
|
1371 |
+
"eval_steps_per_second": 0.12,
|
1372 |
+
"eval_wer": 45.24403277520484,
|
1373 |
+
"step": 455
|
1374 |
+
},
|
1375 |
+
{
|
1376 |
+
"epoch": 17.04,
|
1377 |
+
"learning_rate": 5.715789473684211e-06,
|
1378 |
+
"loss": 0.0453,
|
1379 |
+
"step": 460
|
1380 |
+
},
|
1381 |
+
{
|
1382 |
+
"epoch": 17.04,
|
1383 |
+
"eval_loss": 0.6536934971809387,
|
1384 |
+
"eval_runtime": 327.9543,
|
1385 |
+
"eval_samples_per_second": 0.244,
|
1386 |
+
"eval_steps_per_second": 0.122,
|
1387 |
+
"eval_wer": 48.87780548628429,
|
1388 |
+
"step": 460
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 17.22,
|
1392 |
+
"learning_rate": 5.663157894736843e-06,
|
1393 |
+
"loss": 0.0454,
|
1394 |
+
"step": 465
|
1395 |
+
},
|
1396 |
+
{
|
1397 |
+
"epoch": 17.22,
|
1398 |
+
"eval_loss": 0.6028529405593872,
|
1399 |
+
"eval_runtime": 337.2385,
|
1400 |
+
"eval_samples_per_second": 0.237,
|
1401 |
+
"eval_steps_per_second": 0.119,
|
1402 |
+
"eval_wer": 48.770929818311366,
|
1403 |
+
"step": 465
|
1404 |
+
},
|
1405 |
+
{
|
1406 |
+
"epoch": 17.41,
|
1407 |
+
"learning_rate": 5.610526315789474e-06,
|
1408 |
+
"loss": 0.0662,
|
1409 |
+
"step": 470
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 17.41,
|
1413 |
+
"eval_loss": 0.5731960535049438,
|
1414 |
+
"eval_runtime": 331.8614,
|
1415 |
+
"eval_samples_per_second": 0.241,
|
1416 |
+
"eval_steps_per_second": 0.121,
|
1417 |
+
"eval_wer": 45.885286783042396,
|
1418 |
+
"step": 470
|
1419 |
+
},
|
1420 |
+
{
|
1421 |
+
"epoch": 17.59,
|
1422 |
+
"learning_rate": 5.557894736842105e-06,
|
1423 |
+
"loss": 0.0365,
|
1424 |
+
"step": 475
|
1425 |
+
},
|
1426 |
+
{
|
1427 |
+
"epoch": 17.59,
|
1428 |
+
"eval_loss": 0.6006522178649902,
|
1429 |
+
"eval_runtime": 320.1815,
|
1430 |
+
"eval_samples_per_second": 0.25,
|
1431 |
+
"eval_steps_per_second": 0.125,
|
1432 |
+
"eval_wer": 43.89027431421446,
|
1433 |
+
"step": 475
|
1434 |
+
},
|
1435 |
+
{
|
1436 |
+
"epoch": 17.78,
|
1437 |
+
"learning_rate": 5.505263157894737e-06,
|
1438 |
+
"loss": 0.0728,
|
1439 |
+
"step": 480
|
1440 |
+
},
|
1441 |
+
{
|
1442 |
+
"epoch": 17.78,
|
1443 |
+
"eval_loss": 0.6429422497749329,
|
1444 |
+
"eval_runtime": 333.1015,
|
1445 |
+
"eval_samples_per_second": 0.24,
|
1446 |
+
"eval_steps_per_second": 0.12,
|
1447 |
+
"eval_wer": 44.67402921268258,
|
1448 |
+
"step": 480
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 17.96,
|
1452 |
+
"learning_rate": 5.452631578947369e-06,
|
1453 |
+
"loss": 0.0334,
|
1454 |
+
"step": 485
|
1455 |
+
},
|
1456 |
+
{
|
1457 |
+
"epoch": 17.96,
|
1458 |
+
"eval_loss": 0.648144543170929,
|
1459 |
+
"eval_runtime": 320.2927,
|
1460 |
+
"eval_samples_per_second": 0.25,
|
1461 |
+
"eval_steps_per_second": 0.125,
|
1462 |
+
"eval_wer": 41.11150694691842,
|
1463 |
+
"step": 485
|
1464 |
+
}
|
1465 |
+
],
|
1466 |
+
"max_steps": 1000,
|
1467 |
+
"num_train_epochs": 38,
|
1468 |
+
"total_flos": 1.6475807121408e+19,
|
1469 |
+
"trial_name": null,
|
1470 |
+
"trial_params": null
|
1471 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a0d6b1e93e748ef2d8f0871532589731df0ffddff88c8e6be38cb9ae63ac8b6c
|
3 |
+
size 4155
|