End of training
Browse files- README.md +9 -6
- all_results.json +12 -0
- breeze-dsw-tiny-id.log +14 -0
- eval_results.json +8 -0
- runs/Jan12_23-20-08_knight/events.out.tfevents.1705099256.knight.5818.1 +3 -0
- train_results.json +7 -0
- trainer_state.json +195 -0
README.md
CHANGED
@@ -1,21 +1,24 @@
|
|
1 |
---
|
|
|
|
|
2 |
license: apache-2.0
|
3 |
base_model: openai/whisper-tiny
|
4 |
tags:
|
|
|
5 |
- generated_from_trainer
|
6 |
datasets:
|
7 |
-
- common_voice_16_0
|
8 |
metrics:
|
9 |
- wer
|
10 |
model-index:
|
11 |
-
- name:
|
12 |
results:
|
13 |
- task:
|
14 |
name: Automatic Speech Recognition
|
15 |
type: automatic-speech-recognition
|
16 |
dataset:
|
17 |
-
name: common_voice_16_0
|
18 |
-
type: common_voice_16_0
|
19 |
config: id
|
20 |
split: test
|
21 |
args: id
|
@@ -28,9 +31,9 @@ model-index:
|
|
28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
29 |
should probably proofread and complete it, then remove this comment. -->
|
30 |
|
31 |
-
#
|
32 |
|
33 |
-
This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the common_voice_16_0 dataset.
|
34 |
It achieves the following results on the evaluation set:
|
35 |
- Loss: 0.7109
|
36 |
- Wer: 45.2434
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- id
|
4 |
license: apache-2.0
|
5 |
base_model: openai/whisper-tiny
|
6 |
tags:
|
7 |
+
- whisper-event
|
8 |
- generated_from_trainer
|
9 |
datasets:
|
10 |
+
- mozilla-foundation/common_voice_16_0
|
11 |
metrics:
|
12 |
- wer
|
13 |
model-index:
|
14 |
+
- name: Breeze DSW Indonesian - tiny
|
15 |
results:
|
16 |
- task:
|
17 |
name: Automatic Speech Recognition
|
18 |
type: automatic-speech-recognition
|
19 |
dataset:
|
20 |
+
name: mozilla-foundation/common_voice_16_0 id
|
21 |
+
type: mozilla-foundation/common_voice_16_0
|
22 |
config: id
|
23 |
split: test
|
24 |
args: id
|
|
|
31 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
32 |
should probably proofread and complete it, then remove this comment. -->
|
33 |
|
34 |
+
# Breeze DSW Indonesian - tiny
|
35 |
|
36 |
+
This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the mozilla-foundation/common_voice_16_0 id dataset.
|
37 |
It achieves the following results on the evaluation set:
|
38 |
- Loss: 0.7109
|
39 |
- Wer: 45.2434
|
all_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.03,
|
3 |
+
"eval_loss": 0.7109375,
|
4 |
+
"eval_runtime": 1120.9908,
|
5 |
+
"eval_samples_per_second": 3.249,
|
6 |
+
"eval_steps_per_second": 0.203,
|
7 |
+
"eval_wer": 45.243352654338025,
|
8 |
+
"train_loss": 0.0,
|
9 |
+
"train_runtime": 66.1004,
|
10 |
+
"train_samples_per_second": 968.225,
|
11 |
+
"train_steps_per_second": 30.257
|
12 |
+
}
|
breeze-dsw-tiny-id.log
CHANGED
@@ -480,3 +480,17 @@ Config: alpha=0.000010, betas=(0.900000, 0.999000), weight_decay=0.000000, adam_
|
|
480 |
[2024-01-12 23:22:07,279] [INFO] [engine.py:2988:_get_all_zero_checkpoint_state_dicts] successfully read 1 ZeRO state_dicts for rank 0
|
481 |
[2024-01-12 23:22:07,313] [INFO] [engine.py:2920:_load_zero_checkpoint] loading 1 zero partition checkpoints for rank 0
|
482 |
{'train_runtime': 66.1004, 'train_samples_per_second': 968.225, 'train_steps_per_second': 30.257, 'train_loss': 0.0, 'epoch': 3.03}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
480 |
[2024-01-12 23:22:07,279] [INFO] [engine.py:2988:_get_all_zero_checkpoint_state_dicts] successfully read 1 ZeRO state_dicts for rank 0
|
481 |
[2024-01-12 23:22:07,313] [INFO] [engine.py:2920:_load_zero_checkpoint] loading 1 zero partition checkpoints for rank 0
|
482 |
{'train_runtime': 66.1004, 'train_samples_per_second': 968.225, 'train_steps_per_second': 30.257, 'train_loss': 0.0, 'epoch': 3.03}
|
483 |
+
***** train metrics *****
|
484 |
+
epoch = 3.03
|
485 |
+
train_loss = 0.0
|
486 |
+
train_runtime = 0:01:06.10
|
487 |
+
train_samples_per_second = 968.225
|
488 |
+
train_steps_per_second = 30.257
|
489 |
+
01/12/2024 23:22:15 - INFO - __main__ - *** Evaluate ***
|
490 |
+
***** eval metrics *****
|
491 |
+
epoch = 3.03
|
492 |
+
eval_loss = 0.7109
|
493 |
+
eval_runtime = 0:18:40.99
|
494 |
+
eval_samples_per_second = 3.249
|
495 |
+
eval_steps_per_second = 0.203
|
496 |
+
eval_wer = 45.2434
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.03,
|
3 |
+
"eval_loss": 0.7109375,
|
4 |
+
"eval_runtime": 1120.9908,
|
5 |
+
"eval_samples_per_second": 3.249,
|
6 |
+
"eval_steps_per_second": 0.203,
|
7 |
+
"eval_wer": 45.243352654338025
|
8 |
+
}
|
runs/Jan12_23-20-08_knight/events.out.tfevents.1705099256.knight.5818.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0909b3b36e5be95af082409d55b57c9ca0f9cb13e34f4ff5ebfc90660ada541c
|
3 |
+
size 406
|
train_results.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.03,
|
3 |
+
"train_loss": 0.0,
|
4 |
+
"train_runtime": 66.1004,
|
5 |
+
"train_samples_per_second": 968.225,
|
6 |
+
"train_steps_per_second": 30.257
|
7 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,195 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 45.243352654338025,
|
3 |
+
"best_model_checkpoint": "/cosmos/home/sp-operator/ai/training/models/huggingface/scripts/../breeze-dsw-tiny-id/checkpoint-500",
|
4 |
+
"epoch": 3.032,
|
5 |
+
"eval_steps": 100,
|
6 |
+
"global_step": 500,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.03,
|
13 |
+
"learning_rate": 4.898977360288234e-06,
|
14 |
+
"loss": 1.7847,
|
15 |
+
"step": 25
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.05,
|
19 |
+
"learning_rate": 6.160712527409633e-06,
|
20 |
+
"loss": 1.3566,
|
21 |
+
"step": 50
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.07,
|
25 |
+
"learning_rate": 6.85912902234906e-06,
|
26 |
+
"loss": 1.1407,
|
27 |
+
"step": 75
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.1,
|
31 |
+
"learning_rate": 7.344547104469332e-06,
|
32 |
+
"loss": 0.99,
|
33 |
+
"step": 100
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.1,
|
37 |
+
"eval_loss": 0.8486328125,
|
38 |
+
"eval_runtime": 1294.216,
|
39 |
+
"eval_samples_per_second": 2.814,
|
40 |
+
"eval_steps_per_second": 0.176,
|
41 |
+
"eval_wer": 54.17241696568221,
|
42 |
+
"step": 100
|
43 |
+
},
|
44 |
+
{
|
45 |
+
"epoch": 0.12,
|
46 |
+
"learning_rate": 7.716963756434345e-06,
|
47 |
+
"loss": 0.765,
|
48 |
+
"step": 125
|
49 |
+
},
|
50 |
+
{
|
51 |
+
"epoch": 0.15,
|
52 |
+
"learning_rate": 8.019180844200955e-06,
|
53 |
+
"loss": 0.5849,
|
54 |
+
"step": 150
|
55 |
+
},
|
56 |
+
{
|
57 |
+
"epoch": 1.02,
|
58 |
+
"learning_rate": 8.27351214279797e-06,
|
59 |
+
"loss": 0.7834,
|
60 |
+
"step": 175
|
61 |
+
},
|
62 |
+
{
|
63 |
+
"epoch": 1.04,
|
64 |
+
"learning_rate": 8.49307723936858e-06,
|
65 |
+
"loss": 0.7896,
|
66 |
+
"step": 200
|
67 |
+
},
|
68 |
+
{
|
69 |
+
"epoch": 1.04,
|
70 |
+
"eval_loss": 0.7578125,
|
71 |
+
"eval_runtime": 1161.7981,
|
72 |
+
"eval_samples_per_second": 3.135,
|
73 |
+
"eval_steps_per_second": 0.196,
|
74 |
+
"eval_wer": 48.339313644309506,
|
75 |
+
"step": 200
|
76 |
+
},
|
77 |
+
{
|
78 |
+
"epoch": 1.07,
|
79 |
+
"learning_rate": 8.686247975778677e-06,
|
80 |
+
"loss": 0.8413,
|
81 |
+
"step": 225
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 1.09,
|
85 |
+
"learning_rate": 8.858694625217149e-06,
|
86 |
+
"loss": 0.68,
|
87 |
+
"step": 250
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 1.12,
|
91 |
+
"learning_rate": 9.014436199608479e-06,
|
92 |
+
"loss": 0.5851,
|
93 |
+
"step": 275
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 1.14,
|
97 |
+
"learning_rate": 9.156425255148058e-06,
|
98 |
+
"loss": 0.4164,
|
99 |
+
"step": 300
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 1.14,
|
103 |
+
"eval_loss": 0.73876953125,
|
104 |
+
"eval_runtime": 1191.6636,
|
105 |
+
"eval_samples_per_second": 3.056,
|
106 |
+
"eval_steps_per_second": 0.191,
|
107 |
+
"eval_wer": 49.25936148679732,
|
108 |
+
"step": 300
|
109 |
+
},
|
110 |
+
{
|
111 |
+
"epoch": 2.01,
|
112 |
+
"learning_rate": 9.28689473531776e-06,
|
113 |
+
"loss": 0.5102,
|
114 |
+
"step": 325
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 2.04,
|
118 |
+
"learning_rate": 9.407574351377137e-06,
|
119 |
+
"loss": 0.602,
|
120 |
+
"step": 350
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 2.06,
|
124 |
+
"learning_rate": 9.519831289296397e-06,
|
125 |
+
"loss": 0.6399,
|
126 |
+
"step": 375
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 2.09,
|
130 |
+
"learning_rate": 9.624764935335318e-06,
|
131 |
+
"loss": 0.5456,
|
132 |
+
"step": 400
|
133 |
+
},
|
134 |
+
{
|
135 |
+
"epoch": 2.09,
|
136 |
+
"eval_loss": 0.7177734375,
|
137 |
+
"eval_runtime": 1169.8926,
|
138 |
+
"eval_samples_per_second": 3.113,
|
139 |
+
"eval_steps_per_second": 0.195,
|
140 |
+
"eval_wer": 46.126598583126324,
|
141 |
+
"step": 400
|
142 |
+
},
|
143 |
+
{
|
144 |
+
"epoch": 2.11,
|
145 |
+
"learning_rate": 9.723272550712454e-06,
|
146 |
+
"loss": 0.48,
|
147 |
+
"step": 425
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"epoch": 2.14,
|
151 |
+
"learning_rate": 9.816095971633122e-06,
|
152 |
+
"loss": 0.3325,
|
153 |
+
"step": 450
|
154 |
+
},
|
155 |
+
{
|
156 |
+
"epoch": 3.01,
|
157 |
+
"learning_rate": 9.90385555539545e-06,
|
158 |
+
"loss": 0.3389,
|
159 |
+
"step": 475
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 3.03,
|
163 |
+
"learning_rate": 9.987075336738768e-06,
|
164 |
+
"loss": 0.476,
|
165 |
+
"step": 500
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 3.03,
|
169 |
+
"eval_loss": 0.7109375,
|
170 |
+
"eval_runtime": 1126.9534,
|
171 |
+
"eval_samples_per_second": 3.232,
|
172 |
+
"eval_steps_per_second": 0.202,
|
173 |
+
"eval_wer": 45.243352654338025,
|
174 |
+
"step": 500
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 3.03,
|
178 |
+
"step": 500,
|
179 |
+
"total_flos": 3.922034448169697e+17,
|
180 |
+
"train_loss": 0.0,
|
181 |
+
"train_runtime": 66.1004,
|
182 |
+
"train_samples_per_second": 968.225,
|
183 |
+
"train_steps_per_second": 30.257
|
184 |
+
}
|
185 |
+
],
|
186 |
+
"logging_steps": 25,
|
187 |
+
"max_steps": 2000,
|
188 |
+
"num_input_tokens_seen": 0,
|
189 |
+
"num_train_epochs": 9223372036854775807,
|
190 |
+
"save_steps": 100,
|
191 |
+
"total_flos": 3.922034448169697e+17,
|
192 |
+
"train_batch_size": 32,
|
193 |
+
"trial_name": null,
|
194 |
+
"trial_params": null
|
195 |
+
}
|