Model save

README.md

@@ -0,0 +1,78 @@
+---
+license: apache-2.0
+base_model: openai/whisper-large-v3
+tags:
+- generated_from_trainer
+datasets:
+- common_voice_16_0
+metrics:
+- wer
+model-index:
+- name: openai/whisper-large-v3
+  results:
+  - task:
+      name: Automatic Speech Recognition
+      type: automatic-speech-recognition
+    dataset:
+      name: common_voice_16_0
+      type: common_voice_16_0
+      config: mn
+      split: test
+      args: mn
+    metrics:
+    - name: Wer
+      type: wer
+      value: 35.22282608695652
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# openai/whisper-large-v3
+
+This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the common_voice_16_0 dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.4691
+- Wer: 35.2228
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0001
+- train_batch_size: 32
+- eval_batch_size: 32
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 500
+- num_epochs: 10.0
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Wer     |
+|:-------------:|:-----:|:----:|:---------------:|:-------:|
+| 0.1552        | 4.35  | 500  | 0.5883          | 51.6576 |
+| 0.007         | 8.7   | 1000 | 0.4691          | 35.2228 |
+
+
+### Framework versions
+
+- Transformers 4.38.0.dev0
+- Pytorch 2.1.2+cu121
+- Datasets 2.16.2.dev0
+- Tokenizers 0.15.0

generation_config.json

@@ -0,0 +1,269 @@
+{
+  "alignment_heads": [
+    [
+      7,
+      0
+    ],
+    [
+      10,
+      17
+    ],
+    [
+      12,
+      18
+    ],
+    [
+      13,
+      12
+    ],
+    [
+      16,
+      1
+    ],
+    [
+      17,
+      14
+    ],
+    [
+      19,
+      11
+    ],
+    [
+      21,
+      4
+    ],
+    [
+      24,
+      1
+    ],
+    [
+      25,
+      6
+    ]
+  ],
+  "begin_suppress_tokens": [
+    220,
+    50257
+  ],
+  "bos_token_id": 50257,
+  "decoder_start_token_id": 50258,
+  "eos_token_id": 50257,
+  "forced_decoder_ids": [
+    [
+      1,
+      50259
+    ],
+    [
+      2,
+      50360
+    ],
+    [
+      3,
+      50364
+    ]
+  ],
+  "is_multilingual": true,
+  "lang_to_id": {
+    "<|af|>": 50327,
+    "<|am|>": 50334,
+    "<|ar|>": 50272,
+    "<|as|>": 50350,
+    "<|az|>": 50304,
+    "<|ba|>": 50355,
+    "<|be|>": 50330,
+    "<|bg|>": 50292,
+    "<|bn|>": 50302,
+    "<|bo|>": 50347,
+    "<|br|>": 50309,
+    "<|bs|>": 50315,
+    "<|ca|>": 50270,
+    "<|cs|>": 50283,
+    "<|cy|>": 50297,
+    "<|da|>": 50285,
+    "<|de|>": 50261,
+    "<|el|>": 50281,
+    "<|en|>": 50259,
+    "<|es|>": 50262,
+    "<|et|>": 50307,
+    "<|eu|>": 50310,
+    "<|fa|>": 50300,
+    "<|fi|>": 50277,
+    "<|fo|>": 50338,
+    "<|fr|>": 50265,
+    "<|gl|>": 50319,
+    "<|gu|>": 50333,
+    "<|haw|>": 50352,
+    "<|ha|>": 50354,
+    "<|he|>": 50279,
+    "<|hi|>": 50276,
+    "<|hr|>": 50291,
+    "<|ht|>": 50339,
+    "<|hu|>": 50286,
+    "<|hy|>": 50312,
+    "<|id|>": 50275,
+    "<|is|>": 50311,
+    "<|it|>": 50274,
+    "<|ja|>": 50266,
+    "<|jw|>": 50356,
+    "<|ka|>": 50329,
+    "<|kk|>": 50316,
+    "<|km|>": 50323,
+    "<|kn|>": 50306,
+    "<|ko|>": 50264,
+    "<|la|>": 50294,
+    "<|lb|>": 50345,
+    "<|ln|>": 50353,
+    "<|lo|>": 50336,
+    "<|lt|>": 50293,
+    "<|lv|>": 50301,
+    "<|mg|>": 50349,
+    "<|mi|>": 50295,
+    "<|mk|>": 50308,
+    "<|ml|>": 50296,
+    "<|mn|>": 50314,
+    "<|mr|>": 50320,
+    "<|ms|>": 50282,
+    "<|mt|>": 50343,
+    "<|my|>": 50346,
+    "<|ne|>": 50313,
+    "<|nl|>": 50271,
+    "<|nn|>": 50342,
+    "<|no|>": 50288,
+    "<|oc|>": 50328,
+    "<|pa|>": 50321,
+    "<|pl|>": 50269,
+    "<|ps|>": 50340,
+    "<|pt|>": 50267,
+    "<|ro|>": 50284,
+    "<|ru|>": 50263,
+    "<|sa|>": 50344,
+    "<|sd|>": 50332,
+    "<|si|>": 50322,
+    "<|sk|>": 50298,
+    "<|sl|>": 50305,
+    "<|sn|>": 50324,
+    "<|so|>": 50326,
+    "<|sq|>": 50317,
+    "<|sr|>": 50303,
+    "<|su|>": 50357,
+    "<|sv|>": 50273,
+    "<|sw|>": 50318,
+    "<|ta|>": 50287,
+    "<|te|>": 50299,
+    "<|tg|>": 50331,
+    "<|th|>": 50289,
+    "<|tk|>": 50341,
+    "<|tl|>": 50348,
+    "<|tr|>": 50268,
+    "<|tt|>": 50351,
+    "<|uk|>": 50280,
+    "<|ur|>": 50290,
+    "<|uz|>": 50337,
+    "<|vi|>": 50278,
+    "<|yi|>": 50335,
+    "<|yo|>": 50325,
+    "<|yue|>": 50358,
+    "<|zh|>": 50260
+  },
+  "max_initial_timestamp_index": 50,
+  "max_length": 448,
+  "no_timestamps_token_id": 50364,
+  "pad_token_id": 50257,
+  "prev_sot_token_id": 50362,
+  "return_timestamps": false,
+  "suppress_tokens": [
+    1,
+    2,
+    7,
+    8,
+    9,
+    10,
+    14,
+    25,
+    26,
+    27,
+    28,
+    29,
+    31,
+    58,
+    59,
+    60,
+    61,
+    62,
+    63,
+    90,
+    91,
+    92,
+    93,
+    359,
+    503,
+    522,
+    542,
+    873,
+    893,
+    902,
+    918,
+    922,
+    931,
+    1350,
+    1853,
+    1982,
+    2460,
+    2627,
+    3246,
+    3253,
+    3268,
+    3536,
+    3846,
+    3961,
+    4183,
+    4667,
+    6585,
+    6647,
+    7273,
+    9061,
+    9383,
+    10428,
+    10929,
+    11938,
+    12033,
+    12331,
+    12562,
+    13793,
+    14157,
+    14635,
+    15265,
+    15618,
+    16553,
+    16604,
+    18362,
+    18956,
+    20075,
+    21675,
+    22520,
+    26130,
+    26161,
+    26435,
+    28279,
+    29464,
+    31650,
+    32302,
+    32470,
+    36865,
+    42863,
+    47425,
+    49870,
+    50254,
+    50258,
+    50359,
+    50360,
+    50361,
+    50362,
+    50363
+  ],
+  "task_to_id": {
+    "transcribe": 50360,
+    "translate": 50359
+  },
+  "transformers_version": "4.38.0.dev0"
+}

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b69de7c66cd658940447aaf1e9c00252cd319bbe68ff5e1d3a59584245adb3b
+size 4993448880

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dba351aa4be5630b41b43a92f6c4b6f88262b48a4b09177eab78730f317b0b44
+size 1180663192

runs/Jan22_19-20-32_hf-dgx-01/events.out.tfevents.1705947751.hf-dgx-01.122354.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:114171a7ae29bb534d4f17691e9b2851aab6d2f9b588bff4d693400a8017cd64
-size 12038
+oid sha256:5452f73cf4d57d016c9a013f21ae4146a5fc33582ab761eacf16adebd0fc10fe
+size 13334

wandb/run-20240122_192233-r12o64xn/files/config.yaml

@@ -84,6 +84,26 @@ _wandb:
       5: 1
       6:
       - 1
+    - 1: train/train_runtime
+      5: 1
+      6:
+      - 1
+    - 1: train/train_samples_per_second
+      5: 1
+      6:
+      - 1
+    - 1: train/train_steps_per_second
+      5: 1
+      6:
+      - 1
+    - 1: train/total_flos
+      5: 1
+      6:
+      - 1
+    - 1: train/train_loss
+      5: 1
+      6:
+      - 1
 vocab_size:
   desc: null
   value: 51866

wandb/run-20240122_192233-r12o64xn/files/output.log

@@ -1173,3 +1173,180 @@ Non-default generation parameters: {'max_length': 448, 'begin_suppress_tokens':
 [INFO|feature_extraction_utils.py:425] 2024-01-22 20:52:57,951 >> Feature extractor saved in ./tmp-checkpoint-1000/preprocessor_config.json
 [INFO|feature_extraction_utils.py:425] 2024-01-22 20:53:29,928 >> Feature extractor saved in ./preprocessor_config.json
 /home/sanchit/hf/lib/python3.8/site-packages/torch/utils/checkpoint.py:429: UserWarning: torch.utils.checkpoint: please pass in use_reentrant=True or use_reentrant=False explicitly. The default value of use_reentrant will be updated to be False in the future. To maintain current behavior, pass use_reentrant=True. It is recommended that you use use_reentrant=False. Refer to docs for more details on the differences between the two variants.
+  warnings.warn(
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ 89%|████████████████████████████████████████████████████████████████████▌        | 1024/1150 [1:32:33<08:53,  4.23s/it]
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ 91%|██████████████████████████████████████████████████████████████████████▎      | 1050/1150 [1:34:12<06:27,  3.88s/it]
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ 93%|███████████████████████████████████████████████████████████████████████▉     | 1074/1150 [1:35:46<04:57,  3.91s/it]
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ 96%|█████████████████████████████████████████████████████████████████████████▌   | 1099/1150 [1:37:23<03:19,  3.90s/it]
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ 98%|███████████████████████████████████████████████████████████████████████████▎ | 1125/1150 [1:39:05<01:37,  3.89s/it]
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+100%|████████████████████████████████████████████████████████████████████████████▉| 1149/1150 [1:40:38<00:03,  3.89s/it]
+{'loss': 0.0019, 'learning_rate': 0.0, 'epoch': 10.0}
+100%|█████████████████████████████████████████████████████████████████████████████| 1150/1150 [1:40:40<00:00,  3.25s/it][INFO|trainer.py:1962] 2024-01-22 21:03:14,468 >>
+Training completed. Do not forget to share your model on huggingface.co/models =)
+100%|█████████████████████████████████████████████████████████████████████████████| 1150/1150 [1:40:40<00:00,  5.25s/it]
+[INFO|trainer.py:2926] 2024-01-22 21:03:14,470 >> Saving model checkpoint to ./
+[WARNING|configuration_utils.py:449] 2024-01-22 21:03:14,470 >> Some non-default generation parameters are set in the model config. These should go into a GenerationConfig file (https://huggingface.co/docs/transformers/generation_strategies#save-a-custom-decoding-strategy-with-your-model) instead. This warning will be raised to an exception in v4.41.
+Non-default generation parameters: {'max_length': 448, 'begin_suppress_tokens': [220, 50257]}
+[INFO|configuration_utils.py:473] 2024-01-22 21:03:14,471 >> Configuration saved in ./config.json
+[INFO|configuration_utils.py:595] 2024-01-22 21:03:14,472 >> Configuration saved in ./generation_config.json
+[INFO|modeling_utils.py:2503] 2024-01-22 21:03:29,570 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 2 checkpoint shards. You can find where each parameters has been saved in the index located at ./model.safetensors.index.json.
+[INFO|feature_extraction_utils.py:425] 2024-01-22 21:03:29,571 >> Feature extractor saved in ./preprocessor_config.json
+[INFO|trainer.py:2926] 2024-01-22 21:03:29,572 >> Saving model checkpoint to ./
+[WARNING|configuration_utils.py:449] 2024-01-22 21:03:29,572 >> Some non-default generation parameters are set in the model config. These should go into a GenerationConfig file (https://huggingface.co/docs/transformers/generation_strategies#save-a-custom-decoding-strategy-with-your-model) instead. This warning will be raised to an exception in v4.41.
+Non-default generation parameters: {'max_length': 448, 'begin_suppress_tokens': [220, 50257]}
+[INFO|configuration_utils.py:473] 2024-01-22 21:03:29,573 >> Configuration saved in ./config.json
+[INFO|configuration_utils.py:595] 2024-01-22 21:03:29,574 >> Configuration saved in ./generation_config.json
+[INFO|modeling_utils.py:2503] 2024-01-22 21:03:46,081 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 2 checkpoint shards. You can find where each parameters has been saved in the index located at ./model.safetensors.index.json.
+[INFO|feature_extraction_utils.py:425] 2024-01-22 21:03:46,082 >> Feature extractor saved in ./preprocessor_config.json
+events.out.tfevents.1705947751.hf-dgx-01.122354.0: 100%|███████████████████████████| 13.3k/13.3k [00:00<00:00, 76.9kB/s]
+run-r12o64xn.wandb: 100%|██████████████████████████████████████████████████████████| 1.43M/1.43M [00:00<00:00, 2.25MB/s]
+model-00001-of-00002.safetensors:   1%|▎                                           | 31.1M/4.99G [00:01<04:59, 16.6MB/s]
+model-00002-of-00002.safetensors:   3%|█▏                                          | 32.4M/1.18G [00:01<01:20, 14.3MB/s]

wandb/run-20240122_192233-r12o64xn/files/wandb-summary.json

@@ -1 +1 @@
-{"train/loss": 0.007, "train/learning_rate": 2.307692307692308e-05, "train/epoch": 8.7, "train/global_step": 1000, "_timestamp": 1705953163.2345467, "_runtime": 5409.929728746414, "_step": 41, "eval/loss": 0.4691177308559418, "eval/wer": 35.22282608695652, "eval/runtime": 661.1419, "eval/samples_per_second": 2.863, "eval/steps_per_second": 0.091}
+{"train/loss": 0.0019, "train/learning_rate": 0.0, "train/epoch": 10.0, "train/global_step": 1150, "_timestamp": 1705953794.4692419, "_runtime": 6041.164423942566, "_step": 48, "eval/loss": 0.4691177308559418, "eval/wer": 35.22282608695652, "eval/runtime": 661.1419, "eval/samples_per_second": 2.863, "eval/steps_per_second": 0.091, "train/train_runtime": 6042.7233, "train/train_samples_per_second": 6.06, "train/train_steps_per_second": 0.19, "train/total_flos": 1.244163879862272e+20, "train/train_loss": 0.1683530541315027}

wandb/run-20240122_192233-r12o64xn/run-r12o64xn.wandb

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:738b6b2c9733f9fb41caa91f903b7dcc49658614dc89ed767162a2c7b35bc8cc
-size 1267654
+oid sha256:bf276ce601ac46c83f848562f06d90164e487c41317f4b00065d5dc58b6f6817
+size 1434848