Hugging Face's logo

juancopi81
/
whisper-medium-es-train-valid-bs-64

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
Spanish
whisper
whisper-event
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
juancopi81 commited on
Commit
18dd42b
·
1 Parent(s): 5ffe685

Training in progress, step 500

Browse files
Files changed (10) hide show
  1. all_results.json +7 -0
  2. config.json +1 -1
  3. pytorch_model.bin +1 -1
  4. run.sh +2 -2
  5. runs/Dec16_17-56-05_132-145-140-45/1671213406.9843092/events.out.tfevents.1671213406.132-145-140-45.3273636.1 +3 -0
  6. runs/Dec16_17-56-05_132-145-140-45/events.out.tfevents.1671213406.132-145-140-45.3273636.0 +3 -0
  7. tokenizer_config.json +1 -1
  8. train_results.json +7 -0
  9. trainer_state.json +154 -0
  10. training_args.bin +1 -1
all_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.03,
3
+ "train_loss": 0.13366424560546875,
4
+ "train_runtime": 7967.969,
5
+ "train_samples_per_second": 4.016,
6
+ "train_steps_per_second": 0.063
7
+ }
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "juancopi81/whisper-medium-es",
3
  "activation_dropout": 0.0,
4
  "activation_function": "gelu",
5
  "architectures": [
 
1
  {
2
+ "_name_or_path": "juancopi81/whisper-medium-es-train-valid-bs-64",
3
  "activation_dropout": 0.0,
4
  "activation_function": "gelu",
5
  "architectures": [
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8c751381bcaefb5fefcb06aedec9f60ca97e286ae1c701235f80134ec506d28f
3
  size 1527847357
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:619bd4c8a11c4fe601524a7c0beb23fb82404bd4aa82d1b0783a1aa0557034b3
3
  size 1527847357
run.sh CHANGED
@@ -1,6 +1,6 @@
1
  deepspeed run_speech_recognition_seq2seq_streaming.py \
2
  --deepspeed="ds_config.json" \
3
- --model_name_or_path="juancopi81/whisper-medium-es" \
4
  --dataset_name="mozilla-foundation/common_voice_11_0" \
5
  --dataset_config_name="es" \
6
  --language="spanish" \
@@ -12,7 +12,7 @@ deepspeed run_speech_recognition_seq2seq_streaming.py \
12
  --per_device_train_batch_size="64" \
13
  --per_device_eval_batch_size="32" \
14
  --logging_steps="25" \
15
- --learning_rate="1e-6" \
16
  --warmup_steps="100" \
17
  --evaluation_strategy="steps" \
18
  --eval_steps="500" \
 
1
  deepspeed run_speech_recognition_seq2seq_streaming.py \
2
  --deepspeed="ds_config.json" \
3
+ --model_name_or_path="juancopi81/whisper-medium-es-train-valid-bs-64" \
4
  --dataset_name="mozilla-foundation/common_voice_11_0" \
5
  --dataset_config_name="es" \
6
  --language="spanish" \
 
12
  --per_device_train_batch_size="64" \
13
  --per_device_eval_batch_size="32" \
14
  --logging_steps="25" \
15
+ --learning_rate="5e-7" \
16
  --warmup_steps="100" \
17
  --evaluation_strategy="steps" \
18
  --eval_steps="500" \
runs/Dec16_17-56-05_132-145-140-45/1671213406.9843092/events.out.tfevents.1671213406.132-145-140-45.3273636.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4fbae1d96d5496d87ae9b9eb5edf336650dea99452c378bbb2738e19803e80e9
3
+ size 5880
runs/Dec16_17-56-05_132-145-140-45/events.out.tfevents.1671213406.132-145-140-45.3273636.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:88e2c5d70f41262f89a071c6ccc1938a932f259a33b4d942adfbcb027f2ddee9
3
+ size 7745
tokenizer_config.json CHANGED
@@ -19,7 +19,7 @@
19
  },
20
  "errors": "replace",
21
  "model_max_length": 1024,
22
- "name_or_path": "juancopi81/whisper-medium-es",
23
  "pad_token": null,
24
  "processor_class": "WhisperProcessor",
25
  "return_attention_mask": false,
 
19
  },
20
  "errors": "replace",
21
  "model_max_length": 1024,
22
+ "name_or_path": "juancopi81/whisper-medium-es-train-valid-bs-64",
23
  "pad_token": null,
24
  "processor_class": "WhisperProcessor",
25
  "return_attention_mask": false,
train_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.03,
3
+ "train_loss": 0.13366424560546875,
4
+ "train_runtime": 7967.969,
5
+ "train_samples_per_second": 4.016,
6
+ "train_steps_per_second": 0.063
7
+ }
trainer_state.json ADDED
@@ -0,0 +1,154 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 5.735798555168651,
3
+ "best_model_checkpoint": "./checkpoint-500",
4
+ "epoch": 1.03,
5
+ "global_step": 500,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.05,
12
+ "learning_rate": 6.611096473669595e-07,
13
+ "loss": 0.6089,
14
+ "step": 25
15
+ },
16
+ {
17
+ "epoch": 0.1,
18
+ "learning_rate": 8.313789158407869e-07,
19
+ "loss": 0.1002,
20
+ "step": 50
21
+ },
22
+ {
23
+ "epoch": 0.15,
24
+ "learning_rate": 9.256291743595375e-07,
25
+ "loss": 0.0973,
26
+ "step": 75
27
+ },
28
+ {
29
+ "epoch": 0.2,
30
+ "learning_rate": 9.91135616519784e-07,
31
+ "loss": 0.096,
32
+ "step": 100
33
+ },
34
+ {
35
+ "epoch": 0.25,
36
+ "learning_rate": 1e-06,
37
+ "loss": 0.1061,
38
+ "step": 125
39
+ },
40
+ {
41
+ "epoch": 0.3,
42
+ "learning_rate": 1e-06,
43
+ "loss": 0.0993,
44
+ "step": 150
45
+ },
46
+ {
47
+ "epoch": 0.35,
48
+ "learning_rate": 1e-06,
49
+ "loss": 0.0994,
50
+ "step": 175
51
+ },
52
+ {
53
+ "epoch": 0.4,
54
+ "learning_rate": 1e-06,
55
+ "loss": 0.0964,
56
+ "step": 200
57
+ },
58
+ {
59
+ "epoch": 0.45,
60
+ "learning_rate": 1e-06,
61
+ "loss": 0.0959,
62
+ "step": 225
63
+ },
64
+ {
65
+ "epoch": 0.5,
66
+ "learning_rate": 1e-06,
67
+ "loss": 0.1011,
68
+ "step": 250
69
+ },
70
+ {
71
+ "epoch": 0.55,
72
+ "learning_rate": 1e-06,
73
+ "loss": 0.0976,
74
+ "step": 275
75
+ },
76
+ {
77
+ "epoch": 0.6,
78
+ "learning_rate": 1e-06,
79
+ "loss": 0.0995,
80
+ "step": 300
81
+ },
82
+ {
83
+ "epoch": 0.65,
84
+ "learning_rate": 1e-06,
85
+ "loss": 0.1059,
86
+ "step": 325
87
+ },
88
+ {
89
+ "epoch": 0.7,
90
+ "learning_rate": 1e-06,
91
+ "loss": 0.1056,
92
+ "step": 350
93
+ },
94
+ {
95
+ "epoch": 0.75,
96
+ "learning_rate": 1e-06,
97
+ "loss": 0.1078,
98
+ "step": 375
99
+ },
100
+ {
101
+ "epoch": 0.8,
102
+ "learning_rate": 1e-06,
103
+ "loss": 0.1122,
104
+ "step": 400
105
+ },
106
+ {
107
+ "epoch": 0.85,
108
+ "learning_rate": 1e-06,
109
+ "loss": 0.1098,
110
+ "step": 425
111
+ },
112
+ {
113
+ "epoch": 0.9,
114
+ "learning_rate": 1e-06,
115
+ "loss": 0.1163,
116
+ "step": 450
117
+ },
118
+ {
119
+ "epoch": 0.95,
120
+ "learning_rate": 1e-06,
121
+ "loss": 0.1209,
122
+ "step": 475
123
+ },
124
+ {
125
+ "epoch": 1.03,
126
+ "learning_rate": 1e-06,
127
+ "loss": 0.1972,
128
+ "step": 500
129
+ },
130
+ {
131
+ "epoch": 1.03,
132
+ "eval_loss": 0.1683349609375,
133
+ "eval_runtime": 3047.5288,
134
+ "eval_samples_per_second": 5.093,
135
+ "eval_steps_per_second": 0.159,
136
+ "eval_wer": 5.735798555168651,
137
+ "step": 500
138
+ },
139
+ {
140
+ "epoch": 1.03,
141
+ "step": 500,
142
+ "total_flos": 3.2659357379539763e+19,
143
+ "train_loss": 0.13366424560546875,
144
+ "train_runtime": 7967.969,
145
+ "train_samples_per_second": 4.016,
146
+ "train_steps_per_second": 0.063
147
+ }
148
+ ],
149
+ "max_steps": 500,
150
+ "num_train_epochs": 9223372036854775807,
151
+ "total_flos": 3.2659357379539763e+19,
152
+ "trial_name": null,
153
+ "trial_params": null
154
+ }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6ee71aa7971a24a5c6f0af1541708e235287f0d50f30593f51de4e017d0a0178
3
  size 4667
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:24d4aa81ef8f537670affe4d5b2c5869ac70b558e4107f25cdf01c52465bbe29
3
  size 4667