Uploading model files

.gitattributes

@@ -25,7 +25,6 @@
 *.safetensors filter=lfs diff=lfs merge=lfs -text
 saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
 *.tflite filter=lfs diff=lfs merge=lfs -text
 *.tgz filter=lfs diff=lfs merge=lfs -text
 *.wasm filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1 @@
+checkpoint-*/

README.md

@@ -0,0 +1,81 @@
+---
+license: apache-2.0
+tags:
+- whisper-event
+- generated_from_trainer
+datasets:
+- google/fleurs
+metrics:
+- wer
+model-index:
+- name: Whisper_large_Shona
+  results:
+  - task:
+      name: Automatic Speech Recognition
+      type: automatic-speech-recognition
+    dataset:
+      name: google/fleurs sn_zw
+      type: google/fleurs
+      config: sn_zw
+      split: test
+    metrics:
+    - name: Wer
+      type: wer
+      value: 37.5
+---
+
+<!-- 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. -->
+
+# Whisper_large_Shona
+
+This model is a fine-tuned version of [openai/whisper-large-v2](https://huggingface.co/openai/whisper-large-v2) on the google/fleurs sn_zw dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.9189
+- Wer: 37.5
+
+## 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: 1e-05
+- train_batch_size: 8
+- eval_batch_size: 16
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 4
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 64
+- total_eval_batch_size: 64
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 100
+- training_steps: 1000
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Wer    |
+|:-------------:|:-----:|:----:|:---------------:|:------:|
+| 0.0005        | 41.64 | 500  | 0.8784          | 37.525 |
+| 0.0003        | 83.32 | 1000 | 0.9189          | 37.5   |
+
+
+### Framework versions
+
+- Transformers 4.25.1
+- Pytorch 1.13.0+cu117
+- Datasets 2.7.1
+- Tokenizers 0.13.2

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 83.32,
+    "eval_loss": 0.9189453125,
+    "eval_runtime": 341.1367,
+    "eval_samples_per_second": 2.712,
+    "eval_steps_per_second": 0.044,
+    "eval_wer": 37.5,
+    "train_loss": 0.04746642480790615,
+    "train_runtime": 8896.3772,
+    "train_samples_per_second": 7.194,
+    "train_steps_per_second": 0.112
+}

config.json

@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "openai/whisper-large-v2",
+  "activation_dropout": 0.0,
+  "activation_function": "gelu",
+  "architectures": [
+    "WhisperForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "begin_suppress_tokens": [
+    220,
+    50257
+  ],
+  "bos_token_id": 50257,
+  "d_model": 1280,
+  "decoder_attention_heads": 20,
+  "decoder_ffn_dim": 5120,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 32,
+  "decoder_start_token_id": 50258,
+  "dropout": 0.0,
+  "encoder_attention_heads": 20,
+  "encoder_ffn_dim": 5120,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 32,
+  "eos_token_id": 50257,
+  "forced_decoder_ids": null,
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "max_length": 448,
+  "max_source_positions": 1500,
+  "max_target_positions": 448,
+  "model_type": "whisper",
+  "num_hidden_layers": 32,
+  "num_mel_bins": 80,
+  "pad_token_id": 50257,
+  "scale_embedding": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.25.1",
+  "use_cache": false,
+  "vocab_size": 51865
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 83.32,
+    "eval_loss": 0.9189453125,
+    "eval_runtime": 341.1367,
+    "eval_samples_per_second": 2.712,
+    "eval_steps_per_second": 0.044,
+    "eval_wer": 37.5
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e316e66ee3f638358154275602d59dcb200512cbae4214135cb896f337ae578
+size 3086784957

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 83.32,
+    "train_loss": 0.04746642480790615,
+    "train_runtime": 8896.3772,
+    "train_samples_per_second": 7.194,
+    "train_steps_per_second": 0.112
+}

trainer_state.json

@@ -0,0 +1,283 @@
+{
+  "best_metric": 37.5,
+  "best_model_checkpoint": "./whisper-large-shona/checkpoint-1000",
+  "epoch": 83.32,
+  "global_step": 1000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 2.08,
+      "learning_rate": 6.712113404111031e-06,
+      "loss": 1.3237,
+      "step": 25
+    },
+    {
+      "epoch": 4.16,
+      "learning_rate": 8.360489289678585e-06,
+      "loss": 0.3887,
+      "step": 50
+    },
+    {
+      "epoch": 6.24,
+      "learning_rate": 9.316614300602277e-06,
+      "loss": 0.1151,
+      "step": 75
+    },
+    {
+      "epoch": 8.32,
+      "learning_rate": 9.978175972987748e-06,
+      "loss": 0.0305,
+      "step": 100
+    },
+    {
+      "epoch": 10.4,
+      "learning_rate": 9.733333333333334e-06,
+      "loss": 0.0107,
+      "step": 125
+    },
+    {
+      "epoch": 12.48,
+      "learning_rate": 9.444444444444445e-06,
+      "loss": 0.0059,
+      "step": 150
+    },
+    {
+      "epoch": 14.56,
+      "learning_rate": 9.155555555555557e-06,
+      "loss": 0.0036,
+      "step": 175
+    },
+    {
+      "epoch": 16.64,
+      "learning_rate": 8.866666666666668e-06,
+      "loss": 0.0028,
+      "step": 200
+    },
+    {
+      "epoch": 18.72,
+      "learning_rate": 8.577777777777778e-06,
+      "loss": 0.0018,
+      "step": 225
+    },
+    {
+      "epoch": 20.8,
+      "learning_rate": 8.288888888888889e-06,
+      "loss": 0.0013,
+      "step": 250
+    },
+    {
+      "epoch": 22.88,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 0.0011,
+      "step": 275
+    },
+    {
+      "epoch": 24.96,
+      "learning_rate": 7.711111111111112e-06,
+      "loss": 0.001,
+      "step": 300
+    },
+    {
+      "epoch": 27.08,
+      "learning_rate": 7.422222222222223e-06,
+      "loss": 0.0009,
+      "step": 325
+    },
+    {
+      "epoch": 29.16,
+      "learning_rate": 7.133333333333334e-06,
+      "loss": 0.0008,
+      "step": 350
+    },
+    {
+      "epoch": 31.24,
+      "learning_rate": 6.844444444444445e-06,
+      "loss": 0.0007,
+      "step": 375
+    },
+    {
+      "epoch": 33.32,
+      "learning_rate": 6.555555555555556e-06,
+      "loss": 0.0007,
+      "step": 400
+    },
+    {
+      "epoch": 35.4,
+      "learning_rate": 6.266666666666668e-06,
+      "loss": 0.0006,
+      "step": 425
+    },
+    {
+      "epoch": 37.48,
+      "learning_rate": 5.977777777777778e-06,
+      "loss": 0.0006,
+      "step": 450
+    },
+    {
+      "epoch": 39.56,
+      "learning_rate": 5.688888888888889e-06,
+      "loss": 0.0005,
+      "step": 475
+    },
+    {
+      "epoch": 41.64,
+      "learning_rate": 5.400000000000001e-06,
+      "loss": 0.0005,
+      "step": 500
+    },
+    {
+      "epoch": 41.64,
+      "eval_loss": 0.87841796875,
+      "eval_runtime": 349.183,
+      "eval_samples_per_second": 2.649,
+      "eval_steps_per_second": 0.043,
+      "eval_wer": 37.525,
+      "step": 500
+    },
+    {
+      "epoch": 43.72,
+      "learning_rate": 5.1111111111111115e-06,
+      "loss": 0.0005,
+      "step": 525
+    },
+    {
+      "epoch": 45.8,
+      "learning_rate": 4.822222222222222e-06,
+      "loss": 0.0005,
+      "step": 550
+    },
+    {
+      "epoch": 47.88,
+      "learning_rate": 4.533333333333334e-06,
+      "loss": 0.0004,
+      "step": 575
+    },
+    {
+      "epoch": 49.96,
+      "learning_rate": 4.244444444444445e-06,
+      "loss": 0.0004,
+      "step": 600
+    },
+    {
+      "epoch": 52.08,
+      "learning_rate": 3.944444444444445e-06,
+      "loss": 0.0004,
+      "step": 625
+    },
+    {
+      "epoch": 54.16,
+      "learning_rate": 3.6555555555555562e-06,
+      "loss": 0.0004,
+      "step": 650
+    },
+    {
+      "epoch": 56.24,
+      "learning_rate": 3.366666666666667e-06,
+      "loss": 0.0004,
+      "step": 675
+    },
+    {
+      "epoch": 58.32,
+      "learning_rate": 3.077777777777778e-06,
+      "loss": 0.0004,
+      "step": 700
+    },
+    {
+      "epoch": 60.4,
+      "learning_rate": 2.788888888888889e-06,
+      "loss": 0.0004,
+      "step": 725
+    },
+    {
+      "epoch": 62.48,
+      "learning_rate": 2.5e-06,
+      "loss": 0.0003,
+      "step": 750
+    },
+    {
+      "epoch": 64.56,
+      "learning_rate": 2.2111111111111113e-06,
+      "loss": 0.0003,
+      "step": 775
+    },
+    {
+      "epoch": 66.64,
+      "learning_rate": 1.9222222222222224e-06,
+      "loss": 0.0003,
+      "step": 800
+    },
+    {
+      "epoch": 68.72,
+      "learning_rate": 1.6333333333333335e-06,
+      "loss": 0.0003,
+      "step": 825
+    },
+    {
+      "epoch": 70.8,
+      "learning_rate": 1.3444444444444446e-06,
+      "loss": 0.0003,
+      "step": 850
+    },
+    {
+      "epoch": 72.88,
+      "learning_rate": 1.0555555555555557e-06,
+      "loss": 0.0003,
+      "step": 875
+    },
+    {
+      "epoch": 74.96,
+      "learning_rate": 7.666666666666667e-07,
+      "loss": 0.0003,
+      "step": 900
+    },
+    {
+      "epoch": 77.08,
+      "learning_rate": 4.777777777777778e-07,
+      "loss": 0.0003,
+      "step": 925
+    },
+    {
+      "epoch": 79.16,
+      "learning_rate": 1.888888888888889e-07,
+      "loss": 0.0003,
+      "step": 950
+    },
+    {
+      "epoch": 81.24,
+      "learning_rate": 0.0,
+      "loss": 0.0003,
+      "step": 975
+    },
+    {
+      "epoch": 83.32,
+      "learning_rate": 0.0,
+      "loss": 0.0003,
+      "step": 1000
+    },
+    {
+      "epoch": 83.32,
+      "eval_loss": 0.9189453125,
+      "eval_runtime": 348.727,
+      "eval_samples_per_second": 2.653,
+      "eval_steps_per_second": 0.043,
+      "eval_wer": 37.5,
+      "step": 1000
+    },
+    {
+      "epoch": 83.32,
+      "step": 1000,
+      "total_flos": 1.387025210307332e+20,
+      "train_loss": 0.04746642480790615,
+      "train_runtime": 8896.3772,
+      "train_samples_per_second": 7.194,
+      "train_steps_per_second": 0.112
+    }
+  ],
+  "max_steps": 1000,
+  "num_train_epochs": 84,
+  "total_flos": 1.387025210307332e+20,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d996b847634ae48932d879d37157d549f850ad46291d2c119cc1c52412cc681
+size 4667