End of training

README.md

@@ -0,0 +1,82 @@
+---
+base_model: masoudmzb/wav2vec2-xlsr-multilingual-53-fa
+tags:
+- generated_from_trainer
+metrics:
+- wer
+model-index:
+- name: wav2vec2-large-xlsr-persian-asr-shemo_lnxdx
+  results: []
+---
+
+<!-- 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. -->
+
+# wav2vec2-large-xlsr-persian-asr-shemo_lnxdx
+
+This model is a fine-tuned version of [masoudmzb/wav2vec2-xlsr-multilingual-53-fa](https://huggingface.co/masoudmzb/wav2vec2-xlsr-multilingual-53-fa) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.7064
+- Wer: 0.3344
+
+## 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: 8
+- seed: 42
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 16
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 500
+- training_steps: 2000
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Wer    |
+|:-------------:|:-----:|:----:|:---------------:|:------:|
+| 1.9455        | 0.62  | 100  | 1.4247          | 0.4831 |
+| 1.4824        | 1.25  | 200  | 1.1107          | 0.4331 |
+| 1.2516        | 1.88  | 300  | 0.9141          | 0.4136 |
+| 1.0859        | 2.5   | 400  | 0.8360          | 0.3975 |
+| 1.0357        | 3.12  | 500  | 0.8097          | 0.3814 |
+| 1.0472        | 3.75  | 600  | 0.7550          | 0.3753 |
+| 0.9963        | 4.38  | 700  | 0.7533          | 0.3636 |
+| 0.9767        | 5.0   | 800  | 0.7424          | 0.3589 |
+| 0.9667        | 5.62  | 900  | 0.7360          | 0.3516 |
+| 0.9385        | 6.25  | 1000 | 0.7355          | 0.3487 |
+| 0.9805        | 6.88  | 1100 | 0.7237          | 0.3464 |
+| 0.976         | 7.5   | 1200 | 0.7078          | 0.3455 |
+| 0.88          | 8.12  | 1300 | 0.7229          | 0.3438 |
+| 0.9421        | 8.75  | 1400 | 0.7180          | 0.3432 |
+| 0.9584        | 9.38  | 1500 | 0.7059          | 0.3364 |
+| 0.88          | 10.0  | 1600 | 0.7106          | 0.3364 |
+| 0.9113        | 10.62 | 1700 | 0.7125          | 0.3344 |
+| 0.912         | 11.25 | 1800 | 0.7091          | 0.3353 |
+| 0.9607        | 11.88 | 1900 | 0.7066          | 0.3344 |
+| 0.8974        | 12.5  | 2000 | 0.7064          | 0.3344 |
+
+
+### Framework versions
+
+- Transformers 4.35.2
+- Pytorch 2.1.0+cu118
+- Datasets 2.15.0
+- Tokenizers 0.15.0

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 12.5,
+    "total_flos": 9.605345431463184e+18,
+    "train_loss": 1.0512529983520509,
+    "train_runtime": 9036.2889,
+    "train_samples": 2554,
+    "train_samples_per_second": 3.541,
+    "train_steps_per_second": 0.221
+}

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "lnxdx/wav2vec2-large-xlsr-persian-asr-shemo_sm74",
+  "_name_or_path": "masoudmzb/wav2vec2-xlsr-multilingual-53-fa",
   "activation_dropout": 0.09216,
   "adapter_attn_dim": null,
   "adapter_kernel_size": 3,
@@ -9,7 +9,7 @@
   "architectures": [
     "Wav2Vec2ForCTC"
   ],
-  "attention_dropout": 0.0,
+  "attention_dropout": 0.1,
   "bos_token_id": 1,
   "classifier_proj_size": 256,
   "codevector_dim": 256,
@@ -50,16 +50,16 @@
   "feat_extract_activation": "gelu",
   "feat_extract_dropout": 0.0,
   "feat_extract_norm": "layer",
-  "feat_proj_dropout": 0.0,
+  "feat_proj_dropout": 0.1,
   "feat_quantizer_dropout": 0.0,
   "final_dropout": 0.0,
   "hidden_act": "gelu",
-  "hidden_dropout": 0.0,
+  "hidden_dropout": 0.1,
   "hidden_size": 1024,
   "initializer_range": 0.02,
   "intermediate_size": 4096,
   "layer_norm_eps": 1e-05,
-  "layerdrop": 0.0,
+  "layerdrop": 0.1,
   "mask_channel_length": 10,
   "mask_channel_min_space": 1,
   "mask_channel_other": 0.0,
@@ -72,7 +72,7 @@
   "mask_time_min_masks": 2,
   "mask_time_min_space": 1,
   "mask_time_other": 0.0,
-  "mask_time_prob": 0.05,
+  "mask_time_prob": 0.075,
   "mask_time_selection": "static",
   "model_type": "wav2vec2",
   "num_adapter_layers": 3,

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a211c63d80f38da4e47e2b53fa296f26ee404f07aa1c8db93fc80a31ee10e15e
+oid sha256:97091bd4b726eeb78e17650e14a2c752540d9ed803467afde90d1a9a185fb735
 size 1261971480

special_tokens_map.json

@@ -1,30 +1,6 @@
 {
-  "bos_token": {
-    "content": "<s>",
-    "lstrip": true,
-    "normalized": false,
-    "rstrip": true,
-    "single_word": false
-  },
-  "eos_token": {
-    "content": "</s>",
-    "lstrip": true,
-    "normalized": false,
-    "rstrip": true,
-    "single_word": false
-  },
-  "pad_token": {
-    "content": "<pad>",
-    "lstrip": true,
-    "normalized": false,
-    "rstrip": true,
-    "single_word": false
-  },
-  "unk_token": {
-    "content": "<unk>",
-    "lstrip": true,
-    "normalized": false,
-    "rstrip": true,
-    "single_word": false
-  }
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
 }

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 12.5,
+    "total_flos": 9.605345431463184e+18,
+    "train_loss": 1.0512529983520509,
+    "train_runtime": 9036.2889,
+    "train_samples": 2554,
+    "train_samples_per_second": 3.541,
+    "train_steps_per_second": 0.221
+}

trainer_state.json

@@ -0,0 +1,328 @@
+{
+  "best_metric": 0.33440420560747663,
+  "best_model_checkpoint": "/content/drive/MyDrive/wav2vec2-large-xlsr-persian-asr-shemo_lnxdx/checkpoint-1700",
+  "epoch": 12.5,
+  "eval_steps": 100,
+  "global_step": 2000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.62,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 1.9455,
+      "step": 100
+    },
+    {
+      "epoch": 0.62,
+      "eval_loss": 1.4246947765350342,
+      "eval_runtime": 33.9505,
+      "eval_samples_per_second": 8.365,
+      "eval_steps_per_second": 1.06,
+      "eval_wer": 0.4830607476635514,
+      "step": 100
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 1.4824,
+      "step": 200
+    },
+    {
+      "epoch": 1.25,
+      "eval_loss": 1.110669493675232,
+      "eval_runtime": 35.7852,
+      "eval_samples_per_second": 7.936,
+      "eval_steps_per_second": 1.006,
+      "eval_wer": 0.43311915887850466,
+      "step": 200
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6e-06,
+      "loss": 1.2516,
+      "step": 300
+    },
+    {
+      "epoch": 1.88,
+      "eval_loss": 0.9141018390655518,
+      "eval_runtime": 33.4012,
+      "eval_samples_per_second": 8.503,
+      "eval_steps_per_second": 1.078,
+      "eval_wer": 0.4135514018691589,
+      "step": 300
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 7.980000000000002e-06,
+      "loss": 1.0859,
+      "step": 400
+    },
+    {
+      "epoch": 2.5,
+      "eval_loss": 0.8359942436218262,
+      "eval_runtime": 33.9773,
+      "eval_samples_per_second": 8.359,
+      "eval_steps_per_second": 1.06,
+      "eval_wer": 0.3974883177570093,
+      "step": 400
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 9.980000000000001e-06,
+      "loss": 1.0357,
+      "step": 500
+    },
+    {
+      "epoch": 3.12,
+      "eval_loss": 0.8096792101860046,
+      "eval_runtime": 33.8733,
+      "eval_samples_per_second": 8.384,
+      "eval_steps_per_second": 1.063,
+      "eval_wer": 0.3814252336448598,
+      "step": 500
+    },
+    {
+      "epoch": 3.75,
+      "learning_rate": 9.340000000000002e-06,
+      "loss": 1.0472,
+      "step": 600
+    },
+    {
+      "epoch": 3.75,
+      "eval_loss": 0.7550356984138489,
+      "eval_runtime": 34.119,
+      "eval_samples_per_second": 8.324,
+      "eval_steps_per_second": 1.055,
+      "eval_wer": 0.37529205607476634,
+      "step": 600
+    },
+    {
+      "epoch": 4.38,
+      "learning_rate": 8.673333333333334e-06,
+      "loss": 0.9963,
+      "step": 700
+    },
+    {
+      "epoch": 4.38,
+      "eval_loss": 0.7533305287361145,
+      "eval_runtime": 36.7992,
+      "eval_samples_per_second": 7.718,
+      "eval_steps_per_second": 0.978,
+      "eval_wer": 0.36360981308411217,
+      "step": 700
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 8.006666666666667e-06,
+      "loss": 0.9767,
+      "step": 800
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 0.7423538565635681,
+      "eval_runtime": 33.6729,
+      "eval_samples_per_second": 8.434,
+      "eval_steps_per_second": 1.069,
+      "eval_wer": 0.3589369158878505,
+      "step": 800
+    },
+    {
+      "epoch": 5.62,
+      "learning_rate": 7.340000000000001e-06,
+      "loss": 0.9667,
+      "step": 900
+    },
+    {
+      "epoch": 5.62,
+      "eval_loss": 0.736015796661377,
+      "eval_runtime": 34.1214,
+      "eval_samples_per_second": 8.323,
+      "eval_steps_per_second": 1.055,
+      "eval_wer": 0.3516355140186916,
+      "step": 900
+    },
+    {
+      "epoch": 6.25,
+      "learning_rate": 6.680000000000001e-06,
+      "loss": 0.9385,
+      "step": 1000
+    },
+    {
+      "epoch": 6.25,
+      "eval_loss": 0.7354543209075928,
+      "eval_runtime": 33.4234,
+      "eval_samples_per_second": 8.497,
+      "eval_steps_per_second": 1.077,
+      "eval_wer": 0.348714953271028,
+      "step": 1000
+    },
+    {
+      "epoch": 6.88,
+      "learning_rate": 6.013333333333335e-06,
+      "loss": 0.9805,
+      "step": 1100
+    },
+    {
+      "epoch": 6.88,
+      "eval_loss": 0.7237350940704346,
+      "eval_runtime": 36.5425,
+      "eval_samples_per_second": 7.772,
+      "eval_steps_per_second": 0.985,
+      "eval_wer": 0.3463785046728972,
+      "step": 1100
+    },
+    {
+      "epoch": 7.5,
+      "learning_rate": 5.346666666666667e-06,
+      "loss": 0.976,
+      "step": 1200
+    },
+    {
+      "epoch": 7.5,
+      "eval_loss": 0.7077719569206238,
+      "eval_runtime": 33.9188,
+      "eval_samples_per_second": 8.373,
+      "eval_steps_per_second": 1.061,
+      "eval_wer": 0.3455023364485981,
+      "step": 1200
+    },
+    {
+      "epoch": 8.12,
+      "learning_rate": 4.680000000000001e-06,
+      "loss": 0.88,
+      "step": 1300
+    },
+    {
+      "epoch": 8.12,
+      "eval_loss": 0.7229148745536804,
+      "eval_runtime": 33.6191,
+      "eval_samples_per_second": 8.448,
+      "eval_steps_per_second": 1.071,
+      "eval_wer": 0.34375,
+      "step": 1300
+    },
+    {
+      "epoch": 8.75,
+      "learning_rate": 4.013333333333334e-06,
+      "loss": 0.9421,
+      "step": 1400
+    },
+    {
+      "epoch": 8.75,
+      "eval_loss": 0.7180063128471375,
+      "eval_runtime": 33.4233,
+      "eval_samples_per_second": 8.497,
+      "eval_steps_per_second": 1.077,
+      "eval_wer": 0.3431658878504673,
+      "step": 1400
+    },
+    {
+      "epoch": 9.38,
+      "learning_rate": 3.346666666666667e-06,
+      "loss": 0.9584,
+      "step": 1500
+    },
+    {
+      "epoch": 9.38,
+      "eval_loss": 0.705901563167572,
+      "eval_runtime": 33.1792,
+      "eval_samples_per_second": 8.56,
+      "eval_steps_per_second": 1.085,
+      "eval_wer": 0.3364485981308411,
+      "step": 1500
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 2.68e-06,
+      "loss": 0.88,
+      "step": 1600
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 0.7106391191482544,
+      "eval_runtime": 35.2342,
+      "eval_samples_per_second": 8.06,
+      "eval_steps_per_second": 1.022,
+      "eval_wer": 0.3364485981308411,
+      "step": 1600
+    },
+    {
+      "epoch": 10.62,
+      "learning_rate": 2.0133333333333337e-06,
+      "loss": 0.9113,
+      "step": 1700
+    },
+    {
+      "epoch": 10.62,
+      "eval_loss": 0.7124722599983215,
+      "eval_runtime": 34.7063,
+      "eval_samples_per_second": 8.183,
+      "eval_steps_per_second": 1.037,
+      "eval_wer": 0.33440420560747663,
+      "step": 1700
+    },
+    {
+      "epoch": 11.25,
+      "learning_rate": 1.3466666666666668e-06,
+      "loss": 0.912,
+      "step": 1800
+    },
+    {
+      "epoch": 11.25,
+      "eval_loss": 0.7090508937835693,
+      "eval_runtime": 33.4477,
+      "eval_samples_per_second": 8.491,
+      "eval_steps_per_second": 1.076,
+      "eval_wer": 0.3352803738317757,
+      "step": 1800
+    },
+    {
+      "epoch": 11.88,
+      "learning_rate": 6.800000000000001e-07,
+      "loss": 0.9607,
+      "step": 1900
+    },
+    {
+      "epoch": 11.88,
+      "eval_loss": 0.7066251635551453,
+      "eval_runtime": 33.4966,
+      "eval_samples_per_second": 8.478,
+      "eval_steps_per_second": 1.075,
+      "eval_wer": 0.33440420560747663,
+      "step": 1900
+    },
+    {
+      "epoch": 12.5,
+      "learning_rate": 1.3333333333333334e-08,
+      "loss": 0.8974,
+      "step": 2000
+    },
+    {
+      "epoch": 12.5,
+      "eval_loss": 0.7064409852027893,
+      "eval_runtime": 33.5058,
+      "eval_samples_per_second": 8.476,
+      "eval_steps_per_second": 1.074,
+      "eval_wer": 0.33440420560747663,
+      "step": 2000
+    },
+    {
+      "epoch": 12.5,
+      "step": 2000,
+      "total_flos": 9.605345431463184e+18,
+      "train_loss": 1.0512529983520509,
+      "train_runtime": 9036.2889,
+      "train_samples_per_second": 3.541,
+      "train_steps_per_second": 0.221
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 2000,
+  "num_train_epochs": 13,
+  "save_steps": 100,
+  "total_flos": 9.605345431463184e+18,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:95a11b78a655870ee6c340ec1cd68af57d92048f9c0feafdb2c69d2e11a9d913
+oid sha256:a3a6e76ce66457592fc27ec5dca1633c2db4d07fda9f003b4d738557d0a30d6d
 size 4664