VladS159 commited on
Commit
a70efe3
·
verified ·
1 Parent(s): 4488d52

End of training

Browse files
README.md ADDED
@@ -0,0 +1,81 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ language:
4
+ - ro
5
+ license: apache-2.0
6
+ base_model: openai/whisper-small
7
+ tags:
8
+ - base_model:adapter:openai/whisper-small
9
+ - lora
10
+ - transformers
11
+ datasets:
12
+ - VladS159/romanian_speech_dataset_with_5_percent_synthetic_data
13
+ metrics:
14
+ - wer
15
+ model-index:
16
+ - name: Whisper Small Ro - PEFT
17
+ results:
18
+ - task:
19
+ type: automatic-speech-recognition
20
+ name: Automatic Speech Recognition
21
+ dataset:
22
+ name: Romanian Speech Dataset + 5% Synthetic
23
+ type: VladS159/romanian_speech_dataset_with_5_percent_synthetic_data
24
+ metrics:
25
+ - type: wer
26
+ value: 106.59810174871058
27
+ name: Wer
28
+ ---
29
+
30
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
31
+ should probably proofread and complete it, then remove this comment. -->
32
+
33
+ # Whisper Small Ro - PEFT
34
+
35
+ This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on the Romanian Speech Dataset + 5% Synthetic dataset.
36
+ It achieves the following results on the evaluation set:
37
+ - Loss: 0.4248
38
+ - Wer: 106.5981
39
+
40
+ ## Model description
41
+
42
+ More information needed
43
+
44
+ ## Intended uses & limitations
45
+
46
+ More information needed
47
+
48
+ ## Training and evaluation data
49
+
50
+ More information needed
51
+
52
+ ## Training procedure
53
+
54
+ ### Training hyperparameters
55
+
56
+ The following hyperparameters were used during training:
57
+ - learning_rate: 0.001
58
+ - train_batch_size: 8
59
+ - eval_batch_size: 8
60
+ - seed: 42
61
+ - optimizer: Use adamw_torch_fused with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
62
+ - lr_scheduler_type: linear
63
+ - lr_scheduler_warmup_steps: 100
64
+ - training_steps: 100
65
+ - mixed_precision_training: Native AMP
66
+
67
+ ### Training results
68
+
69
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
70
+ |:-------------:|:------:|:----:|:---------------:|:--------:|
71
+ | 2.0921 | 0.0113 | 50 | 1.0811 | 95.9014 |
72
+ | 0.6328 | 0.0227 | 100 | 0.4248 | 106.5981 |
73
+
74
+
75
+ ### Framework versions
76
+
77
+ - PEFT 0.18.1.dev0
78
+ - Transformers 4.57.1
79
+ - Pytorch 2.9.1+rocm6.4
80
+ - Datasets 3.6.0
81
+ - Tokenizers 0.22.1
adapter_config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": {
6
+ "base_model_class": "WhisperForConditionalGeneration",
7
+ "parent_library": "transformers.models.whisper.modeling_whisper"
8
+ },
9
+ "base_model_name_or_path": "openai/whisper-small",
10
+ "bias": "none",
11
+ "corda_config": null,
12
+ "ensure_weight_tying": false,
13
+ "eva_config": null,
14
+ "exclude_modules": null,
15
+ "fan_in_fan_out": false,
16
+ "inference_mode": true,
17
+ "init_lora_weights": true,
18
+ "layer_replication": null,
19
+ "layers_pattern": null,
20
+ "layers_to_transform": null,
21
+ "loftq_config": {},
22
+ "lora_alpha": 64,
23
+ "lora_bias": false,
24
+ "lora_dropout": 0.05,
25
+ "megatron_config": null,
26
+ "megatron_core": "megatron.core",
27
+ "modules_to_save": null,
28
+ "peft_type": "LORA",
29
+ "peft_version": "0.18.1.dev0@c5a905d097b3c5ba0caf4cfd39cc8e8aeb0d8f56",
30
+ "qalora_group_size": 16,
31
+ "r": 32,
32
+ "rank_pattern": {},
33
+ "revision": null,
34
+ "target_modules": [
35
+ "q_proj",
36
+ "v_proj"
37
+ ],
38
+ "target_parameters": null,
39
+ "task_type": null,
40
+ "trainable_token_indices": null,
41
+ "use_bdlora": null,
42
+ "use_dora": false,
43
+ "use_qalora": false,
44
+ "use_rslora": false
45
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b0ed1a1c5ca81da7ec6d7bc32f3fb42e4ce07b312144f46b6b2b1d378d883e95
3
+ size 14176064
preprocessor_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "chunk_length": 30,
3
+ "dither": 0.0,
4
+ "feature_extractor_type": "WhisperFeatureExtractor",
5
+ "feature_size": 80,
6
+ "hop_length": 160,
7
+ "n_fft": 400,
8
+ "n_samples": 480000,
9
+ "nb_max_frames": 3000,
10
+ "padding_side": "right",
11
+ "padding_value": 0.0,
12
+ "processor_class": "WhisperProcessor",
13
+ "return_attention_mask": false,
14
+ "sampling_rate": 16000
15
+ }
runs/Dec13_16-06-13_ccafd1f7aa07/events.out.tfevents.1765641974.ccafd1f7aa07.12282.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2200b45720741920b3459ebce43aa72b7166ae15a36c5aaab4563427ce25502d
3
+ size 8099
runs/Dec13_16-21-37_ccafd1f7aa07/events.out.tfevents.1765642898.ccafd1f7aa07.12815.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9aa0b94a211ee1948f1a8a7a3a3f3e3060291e9bcd27f0d12488ae2e7b71c0db
3
+ size 8099
runs/Dec13_16-59-07_ccafd1f7aa07/events.out.tfevents.1765645148.ccafd1f7aa07.13152.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c08cc04d8d27383137667f19af66a163d87d4520263eeca3cc097e95e2ea871
3
+ size 7279
runs/Dec13_17-00-55_ccafd1f7aa07/events.out.tfevents.1765645256.ccafd1f7aa07.13337.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2ee6de8500747ab214dccb1a8f624a1ede27cd3239c039e903bc1259106f2071
3
+ size 7962
runs/Dec13_17-18-54_ccafd1f7aa07/events.out.tfevents.1765646335.ccafd1f7aa07.13703.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62b1fd842acd32d820c29ce366c2f29d04b8eb2c590ffe8ecd758f070b9b8d0e
3
+ size 7763
runs/Dec13_22-39-04_ccafd1f7aa07/events.out.tfevents.1765665544.ccafd1f7aa07.14436.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0d2775d0439477f3b2f5d2038c2a53e15b1f65438a87dc9b6592e9e0d81b4e88
3
+ size 10526
runs/Dec14_10-28-43_ccafd1f7aa07/events.out.tfevents.1765708125.ccafd1f7aa07.16415.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:80d77ca0d16bd325da8a900a8fc90b8c20f68a91b218ff2bc5a98e4b350b2c4a
3
+ size 7495
runs/Dec14_10-37-57_ccafd1f7aa07/events.out.tfevents.1765708678.ccafd1f7aa07.16689.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0ece13614e8fd65cb3376652c06896a098e63d526956dedb6d8b35e5b32760e6
3
+ size 8674
runs/Dec14_11-53-06_ccafd1f7aa07/events.out.tfevents.1765713188.ccafd1f7aa07.17618.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b1f2a10786e2feede0b49daa694f169119afad55d6c1dde9aae41e26cc761bf
3
+ size 8675
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc3f41bed743b1ceaf6086b801eb3f1b90b61f4f968c05a46b93824c58b0ca10
3
+ size 6033