cahya commited on
Commit
6d8fafe
1 Parent(s): bf864d5

Training in progress, step 50

Browse files
.gitignore CHANGED
@@ -1 +1,2 @@
1
- checkpoint-*/
 
 
1
+ checkpoint-*/
2
+ wandb/
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7d05055166909e6f0c2c92d51c1d2bcbca00165dbfcc573864d195cd6507063e
3
  size 377694615
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:74b3154675d7dfcbb1295eadc674f84f1852acbe7db2db7de7eefce04b7bad2c
3
  size 377694615
run.sh CHANGED
@@ -1,21 +1,25 @@
 
 
 
 
1
  python run_speech_recognition_ctc.py \
2
  --dataset_name="mozilla-foundation/common_voice_7_0" \
3
  --model_name_or_path="cahya/wav2vec2-base-turkish-artificial" \
4
  --dataset_config_name="tr" \
5
  --output_dir="./" \
6
  --overwrite_output_dir \
7
- --num_train_epochs="1" \
8
- --per_device_train_batch_size="32" \
9
- --per_device_eval_batch_size="2" \
10
  --gradient_accumulation_steps="4" \
11
  --learning_rate="3e-4" \
12
- --warmup_steps="2000" \
13
  --length_column_name="input_length" \
14
  --evaluation_strategy="steps" \
15
  --text_column_name="sentence" \
16
- --save_steps="500" \
17
- --eval_steps="500" \
18
- --logging_steps="100" \
19
  --layerdrop="0.0" \
20
  --activation_dropout="0.1" \
21
  --save_total_limit="3" \
 
1
+ export WANDB_ENTITY=cahya
2
+ export WANDB_LOG_MODEL=true
3
+ export WANDB_PROJECT=xlsr-turkish
4
+
5
  python run_speech_recognition_ctc.py \
6
  --dataset_name="mozilla-foundation/common_voice_7_0" \
7
  --model_name_or_path="cahya/wav2vec2-base-turkish-artificial" \
8
  --dataset_config_name="tr" \
9
  --output_dir="./" \
10
  --overwrite_output_dir \
11
+ --num_train_epochs="100" \
12
+ --per_device_train_batch_size="128" \
13
+ --per_device_eval_batch_size="8" \
14
  --gradient_accumulation_steps="4" \
15
  --learning_rate="3e-4" \
16
+ --warmup_steps="50" \
17
  --length_column_name="input_length" \
18
  --evaluation_strategy="steps" \
19
  --text_column_name="sentence" \
20
+ --save_steps="50" \
21
+ --eval_steps="50" \
22
+ --logging_steps="50" \
23
  --layerdrop="0.0" \
24
  --activation_dropout="0.1" \
25
  --save_total_limit="3" \
tmp.0/config.json DELETED
@@ -1,105 +0,0 @@
1
- {
2
- "_name_or_path": "cahya/wav2vec2-base-turkish-artificial",
3
- "activation_dropout": 0.1,
4
- "adapter_kernel_size": 3,
5
- "adapter_stride": 2,
6
- "add_adapter": false,
7
- "apply_spec_augment": true,
8
- "architectures": [
9
- "Wav2Vec2ForCTC"
10
- ],
11
- "attention_dropout": 0.0,
12
- "bos_token_id": 1,
13
- "classifier_proj_size": 256,
14
- "codevector_dim": 256,
15
- "contrastive_logits_temperature": 0.1,
16
- "conv_bias": false,
17
- "conv_dim": [
18
- 512,
19
- 512,
20
- 512,
21
- 512,
22
- 512,
23
- 512,
24
- 512
25
- ],
26
- "conv_kernel": [
27
- 10,
28
- 3,
29
- 3,
30
- 3,
31
- 3,
32
- 2,
33
- 2
34
- ],
35
- "conv_stride": [
36
- 5,
37
- 2,
38
- 2,
39
- 2,
40
- 2,
41
- 2,
42
- 2
43
- ],
44
- "ctc_loss_reduction": "mean",
45
- "ctc_zero_infinity": true,
46
- "diversity_loss_weight": 0.1,
47
- "do_stable_layer_norm": false,
48
- "eos_token_id": 2,
49
- "feat_extract_activation": "gelu",
50
- "feat_extract_norm": "group",
51
- "feat_proj_dropout": 0.0,
52
- "feat_quantizer_dropout": 0.0,
53
- "final_dropout": 0.0,
54
- "hidden_act": "gelu",
55
- "hidden_dropout": 0.0,
56
- "hidden_size": 768,
57
- "initializer_range": 0.02,
58
- "intermediate_size": 3072,
59
- "layer_norm_eps": 1e-05,
60
- "layerdrop": 0.0,
61
- "mask_feature_length": 64,
62
- "mask_feature_min_masks": 0,
63
- "mask_feature_prob": 0.25,
64
- "mask_time_length": 10,
65
- "mask_time_min_masks": 2,
66
- "mask_time_prob": 0.75,
67
- "model_type": "wav2vec2",
68
- "num_adapter_layers": 3,
69
- "num_attention_heads": 12,
70
- "num_codevector_groups": 2,
71
- "num_codevectors_per_group": 320,
72
- "num_conv_pos_embedding_groups": 16,
73
- "num_conv_pos_embeddings": 128,
74
- "num_feat_extract_layers": 7,
75
- "num_hidden_layers": 12,
76
- "num_negatives": 100,
77
- "output_hidden_size": 768,
78
- "pad_token_id": 39,
79
- "proj_codevector_dim": 256,
80
- "tdnn_dilation": [
81
- 1,
82
- 2,
83
- 3,
84
- 1,
85
- 1
86
- ],
87
- "tdnn_dim": [
88
- 512,
89
- 512,
90
- 512,
91
- 512,
92
- 1500
93
- ],
94
- "tdnn_kernel": [
95
- 5,
96
- 3,
97
- 3,
98
- 1,
99
- 1
100
- ],
101
- "transformers_version": "4.17.0.dev0",
102
- "use_weighted_layer_sum": false,
103
- "vocab_size": 40,
104
- "xvector_output_dim": 512
105
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
tmp.0/preprocessor_config.json DELETED
@@ -1,10 +0,0 @@
1
- {
2
- "do_normalize": true,
3
- "feature_extractor_type": "Wav2Vec2FeatureExtractor",
4
- "feature_size": 1,
5
- "padding_side": "right",
6
- "padding_value": 0.0,
7
- "processor_class": "Wav2Vec2ProcessorWithLM",
8
- "return_attention_mask": true,
9
- "sampling_rate": 16000
10
- }
 
 
 
 
 
 
 
 
 
 
 
tmp.0/special_tokens_map.json DELETED
@@ -1 +0,0 @@
1
- {"unk_token": "[UNK]", "pad_token": "[PAD]"}
 
 
tmp.0/tokenizer_config.json DELETED
@@ -1 +0,0 @@
1
- {"unk_token": "[UNK]", "bos_token": null, "eos_token": null, "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|", "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "./", "tokenizer_class": "Wav2Vec2CTCTokenizer"}
 
 
tmp.0/vocab.json DELETED
@@ -1 +0,0 @@
1
- {"-": 1, "a": 2, "b": 3, "c": 4, "d": 5, "e": 6, "f": 7, "g": 8, "h": 9, "i": 10, "j": 11, "k": 12, "l": 13, "m": 14, "n": 15, "o": 16, "p": 17, "q": 18, "r": 19, "s": 20, "t": 21, "u": 22, "v": 23, "w": 24, "x": 25, "y": 26, "z": 27, "â": 28, "ç": 29, "ë": 30, "î": 31, "ö": 32, "ü": 33, "ğ": 34, "ı": 35, "ş": 36, "̇": 37, "|": 0, "[UNK]": 38, "[PAD]": 39}
 
 
tmp/added_tokens.json DELETED
@@ -1 +0,0 @@
1
- {}
 
 
tmp/alphabet.json DELETED
@@ -1 +0,0 @@
1
- {"labels": [" ", "-", "a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "q", "r", "s", "t", "u", "v", "w", "x", "y", "z", "\u00e2", "\u00e7", "\u00eb", "\u00ee", "\u00f6", "\u00fc", "\u011f", "\u0131", "\u015f", "\u0307", "\u2047", "" ], "is_bpe": false}
 
 
tmp/special_tokens_map.json DELETED
@@ -1 +0,0 @@
1
- {"bos_token": null, "eos_token": null, "unk_token": "[UNK]", "pad_token": "[PAD]"}
 
 
tmp/tokenizer_config.json DELETED
@@ -1 +0,0 @@
1
- {"unk_token": "[UNK]", "bos_token": null, "eos_token": null, "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|", "special_tokens_map_file": "/home/cahya/.cache/huggingface/transformers/2916f85ca1c8db02f6dffab6c5680b457e26471b99cac401f31d9d01fd7af256.a21d51735cf8667bcd610f057e88548d5d6a381401f6b4501a8bc6c1a9dc8498", "tokenizer_file": null, "name_or_path": "./", "processor_class": "Wav2Vec2ProcessorWithLM", "tokenizer_class": "Wav2Vec2CTCTokenizer"}
 
 
tmp/vocab.json DELETED
@@ -1 +0,0 @@
1
- {"(": 1, ")": 2, "-": 3, "a": 4, "b": 5, "c": 6, "d": 7, "e": 8, "f": 9, "g": 10, "h": 11, "i": 12, "j": 13, "k": 14, "l": 15, "m": 16, "n": 17, "o": 18, "p": 19, "q": 20, "r": 21, "s": 22, "t": 23, "u": 24, "v": 25, "w": 26, "x": 27, "y": 28, "z": 29, "\u00e2": 30, "\u00e7": 31, "\u00eb": 32, "\u00ee": 33, "\u00f6": 34, "\u00fc": 35, "\u011f": 36, "\u0131": 37, "\u015f": 38, "\u0307": 39, "|": 0, "[UNK]": 40, "[PAD]": 41}
 
 
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7a10bcc5d960643107542b1720e90ec5f5d76d687e1e4640d575348a7a1a39a6
3
  size 2991
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a546fe0a11ea109369ec2dba50f4765671c1c371cbe15e0062ea7da78f0f82b
3
  size 2991