Commit
•
55f8a6e
1
Parent(s):
a174ad5
End of training
Browse files- README.md +18 -2
- trainer_state.json +186 -0
README.md
CHANGED
@@ -1,11 +1,27 @@
|
|
1 |
---
|
|
|
|
|
|
|
|
|
2 |
tags:
|
3 |
- generated_from_trainer
|
|
|
|
|
4 |
metrics:
|
5 |
- wer
|
6 |
model-index:
|
7 |
- name: quran-whisper-ar-tiny-1
|
8 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
---
|
10 |
|
11 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -13,7 +29,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
13 |
|
14 |
# quran-whisper-ar-tiny-1
|
15 |
|
16 |
-
This model
|
17 |
It achieves the following results on the evaluation set:
|
18 |
- Loss: 0.5858
|
19 |
- Wer: 164.8352
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- ar
|
4 |
+
license: apache-2.0
|
5 |
+
base_model: openai/whisper-tiny
|
6 |
tags:
|
7 |
- generated_from_trainer
|
8 |
+
datasets:
|
9 |
+
- tarteel-ai/everyayah
|
10 |
metrics:
|
11 |
- wer
|
12 |
model-index:
|
13 |
- name: quran-whisper-ar-tiny-1
|
14 |
+
results:
|
15 |
+
- task:
|
16 |
+
name: Automatic Speech Recognition
|
17 |
+
type: automatic-speech-recognition
|
18 |
+
dataset:
|
19 |
+
name: tarteel-ai/everyayah
|
20 |
+
type: tarteel-ai/everyayah
|
21 |
+
metrics:
|
22 |
+
- name: Wer
|
23 |
+
type: wer
|
24 |
+
value: 164.83516483516482
|
25 |
---
|
26 |
|
27 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
29 |
|
30 |
# quran-whisper-ar-tiny-1
|
31 |
|
32 |
+
This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the tarteel-ai/everyayah dataset.
|
33 |
It achieves the following results on the evaluation set:
|
34 |
- Loss: 0.5858
|
35 |
- Wer: 164.8352
|
trainer_state.json
ADDED
@@ -0,0 +1,186 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 102.19780219780219,
|
3 |
+
"best_model_checkpoint": "./quran-whisper-ar-tiny-1/checkpoint-15",
|
4 |
+
"epoch": 1.0,
|
5 |
+
"eval_steps": 5,
|
6 |
+
"global_step": 45,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.3333333333333333,
|
13 |
+
"grad_norm": Infinity,
|
14 |
+
"learning_rate": 9.333333333333334e-05,
|
15 |
+
"loss": 2.2643,
|
16 |
+
"step": 5
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.3333333333333333,
|
20 |
+
"eval_loss": 4.266593933105469,
|
21 |
+
"eval_runtime": 19.8469,
|
22 |
+
"eval_samples_per_second": 0.504,
|
23 |
+
"eval_steps_per_second": 0.101,
|
24 |
+
"eval_wer": 209.8901098901099,
|
25 |
+
"step": 5
|
26 |
+
},
|
27 |
+
{
|
28 |
+
"epoch": 0.6666666666666666,
|
29 |
+
"grad_norm": 53.0462760925293,
|
30 |
+
"learning_rate": 7.333333333333333e-05,
|
31 |
+
"loss": 3.5798,
|
32 |
+
"step": 10
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.6666666666666666,
|
36 |
+
"eval_loss": 1.9317986965179443,
|
37 |
+
"eval_runtime": 17.1404,
|
38 |
+
"eval_samples_per_second": 0.583,
|
39 |
+
"eval_steps_per_second": 0.117,
|
40 |
+
"eval_wer": 173.62637362637363,
|
41 |
+
"step": 10
|
42 |
+
},
|
43 |
+
{
|
44 |
+
"epoch": 1.0,
|
45 |
+
"grad_norm": 15.332279205322266,
|
46 |
+
"learning_rate": 4e-05,
|
47 |
+
"loss": 1.2993,
|
48 |
+
"step": 15
|
49 |
+
},
|
50 |
+
{
|
51 |
+
"epoch": 1.0,
|
52 |
+
"eval_loss": 1.0968444347381592,
|
53 |
+
"eval_runtime": 14.7686,
|
54 |
+
"eval_samples_per_second": 0.677,
|
55 |
+
"eval_steps_per_second": 0.135,
|
56 |
+
"eval_wer": 102.19780219780219,
|
57 |
+
"step": 15
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 0.6666666666666666,
|
61 |
+
"grad_norm": 14.598529815673828,
|
62 |
+
"learning_rate": 6e-05,
|
63 |
+
"loss": 0.8612,
|
64 |
+
"step": 20
|
65 |
+
},
|
66 |
+
{
|
67 |
+
"epoch": 0.6666666666666666,
|
68 |
+
"eval_loss": 0.9798414707183838,
|
69 |
+
"eval_runtime": 14.0253,
|
70 |
+
"eval_samples_per_second": 0.713,
|
71 |
+
"eval_steps_per_second": 0.143,
|
72 |
+
"eval_wer": 145.05494505494505,
|
73 |
+
"step": 20
|
74 |
+
},
|
75 |
+
{
|
76 |
+
"epoch": 0.8333333333333334,
|
77 |
+
"grad_norm": 10.819196701049805,
|
78 |
+
"learning_rate": 4.3333333333333334e-05,
|
79 |
+
"loss": 0.7235,
|
80 |
+
"step": 25
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.8333333333333334,
|
84 |
+
"eval_loss": 0.7983613610267639,
|
85 |
+
"eval_runtime": 13.4734,
|
86 |
+
"eval_samples_per_second": 0.742,
|
87 |
+
"eval_steps_per_second": 0.148,
|
88 |
+
"eval_wer": 131.86813186813185,
|
89 |
+
"step": 25
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 1.0,
|
93 |
+
"grad_norm": 12.039382934570312,
|
94 |
+
"learning_rate": 2.6666666666666667e-05,
|
95 |
+
"loss": 0.5694,
|
96 |
+
"step": 30
|
97 |
+
},
|
98 |
+
{
|
99 |
+
"epoch": 1.0,
|
100 |
+
"eval_loss": 0.7010310888290405,
|
101 |
+
"eval_runtime": 11.6848,
|
102 |
+
"eval_samples_per_second": 0.856,
|
103 |
+
"eval_steps_per_second": 0.171,
|
104 |
+
"eval_wer": 127.47252747252746,
|
105 |
+
"step": 30
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 0.7777777777777778,
|
109 |
+
"grad_norm": 10.747776985168457,
|
110 |
+
"learning_rate": 4.4444444444444447e-05,
|
111 |
+
"loss": 0.5781,
|
112 |
+
"step": 35
|
113 |
+
},
|
114 |
+
{
|
115 |
+
"epoch": 0.7777777777777778,
|
116 |
+
"eval_loss": 0.7016527652740479,
|
117 |
+
"eval_runtime": 13.6665,
|
118 |
+
"eval_samples_per_second": 0.732,
|
119 |
+
"eval_steps_per_second": 0.146,
|
120 |
+
"eval_wer": 378.021978021978,
|
121 |
+
"step": 35
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.8888888888888888,
|
125 |
+
"grad_norm": 11.89635944366455,
|
126 |
+
"learning_rate": 3.3333333333333335e-05,
|
127 |
+
"loss": 0.5704,
|
128 |
+
"step": 40
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.8888888888888888,
|
132 |
+
"eval_loss": 0.6510931253433228,
|
133 |
+
"eval_runtime": 13.0034,
|
134 |
+
"eval_samples_per_second": 0.769,
|
135 |
+
"eval_steps_per_second": 0.154,
|
136 |
+
"eval_wer": 341.75824175824175,
|
137 |
+
"step": 40
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 1.0,
|
141 |
+
"grad_norm": 8.039495468139648,
|
142 |
+
"learning_rate": 2.2222222222222223e-05,
|
143 |
+
"loss": 0.4247,
|
144 |
+
"step": 45
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 1.0,
|
148 |
+
"eval_loss": 0.5858025550842285,
|
149 |
+
"eval_runtime": 13.6045,
|
150 |
+
"eval_samples_per_second": 0.735,
|
151 |
+
"eval_steps_per_second": 0.147,
|
152 |
+
"eval_wer": 164.83516483516482,
|
153 |
+
"step": 45
|
154 |
+
},
|
155 |
+
{
|
156 |
+
"epoch": 1.0,
|
157 |
+
"step": 45,
|
158 |
+
"total_flos": 1.77255972864e+16,
|
159 |
+
"train_loss": 0.17480638292100695,
|
160 |
+
"train_runtime": 95.2131,
|
161 |
+
"train_samples_per_second": 7.562,
|
162 |
+
"train_steps_per_second": 0.473
|
163 |
+
}
|
164 |
+
],
|
165 |
+
"logging_steps": 5,
|
166 |
+
"max_steps": 45,
|
167 |
+
"num_input_tokens_seen": 0,
|
168 |
+
"num_train_epochs": 9223372036854775807,
|
169 |
+
"save_steps": 5,
|
170 |
+
"stateful_callbacks": {
|
171 |
+
"TrainerControl": {
|
172 |
+
"args": {
|
173 |
+
"should_epoch_stop": false,
|
174 |
+
"should_evaluate": false,
|
175 |
+
"should_log": false,
|
176 |
+
"should_save": true,
|
177 |
+
"should_training_stop": true
|
178 |
+
},
|
179 |
+
"attributes": {}
|
180 |
+
}
|
181 |
+
},
|
182 |
+
"total_flos": 1.77255972864e+16,
|
183 |
+
"train_batch_size": 16,
|
184 |
+
"trial_name": null,
|
185 |
+
"trial_params": null
|
186 |
+
}
|