Hello Persian SER

config.json

@@ -0,0 +1,95 @@
+{
+  "_name_or_path": "m3hrdadfi/wav2vec2-xlsr-fa",
+  "activation_dropout": 0.09216,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForSpeechClassification"
+  ],
+  "attention_dropout": 0.05316,
+  "bos_token_id": 1,
+  "conv_bias": true,
+  "conv_dim": [
+    512,
+    512,
+    512,
+    512,
+    512,
+    512,
+    512
+  ],
+  "conv_kernel": [
+    10,
+    3,
+    3,
+    3,
+    3,
+    2,
+    2
+  ],
+  "conv_stride": [
+    5,
+    2,
+    2,
+    2,
+    2,
+    2,
+    2
+  ],
+  "ctc_loss_reduction": "mean",
+  "ctc_zero_infinity": true,
+  "do_stable_layer_norm": true,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_dropout": 0.0,
+  "feat_extract_norm": "layer",
+  "feat_proj_dropout": 0.01249,
+  "final_dropout": 0.0,
+  "finetuning_task": "wav2vec2_clf",
+  "gradient_checkpointing": true,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.01941,
+  "hidden_size": 1024,
+  "id2label": {
+    "0": "Anger",
+    "1": "Fear",
+    "2": "Happiness",
+    "3": "Neutral",
+    "4": "Sadness",
+    "5": "Surprise"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "label2id": {
+    "Anger": 0,
+    "Fear": 1,
+    "Happiness": 2,
+    "Neutral": 3,
+    "Sadness": 4,
+    "Surprise": 5
+  },
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.01377,
+  "mask_channel_length": 10,
+  "mask_channel_min_space": 1,
+  "mask_channel_other": 0.0,
+  "mask_channel_prob": 0.0,
+  "mask_channel_selection": "static",
+  "mask_feature_length": 10,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_space": 1,
+  "mask_time_other": 0.0,
+  "mask_time_prob": 0.04529,
+  "mask_time_selection": "static",
+  "model_type": "wav2vec2",
+  "num_attention_heads": 16,
+  "num_conv_pos_embedding_groups": 16,
+  "num_conv_pos_embeddings": 128,
+  "num_feat_extract_layers": 7,
+  "num_hidden_layers": 24,
+  "pad_token_id": 0,
+  "pooling_mode": "mean",
+  "problem_type": "single_label_classification",
+  "transformers_version": "4.7.0.dev0",
+  "vocab_size": 40
+}

preprocessor_config.json

@@ -0,0 +1,9 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69a746cbe4d4f7c4402475de537f377c560349905c69f8e503cfbb27dcdf8467
+size 1266157609

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "<pad>", "do_lower_case": false, "word_delimiter_token": "|", "special_tokens_map_file": "/workspace/output_models/fa/xlsr-fa/special_tokens_map.json", "tokenizer_file": null, "name_or_path": "m3hrdadfi/wav2vec2-xlsr-fa"}

trainer_state.json

@@ -0,0 +1,121 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.0,
+  "global_step": 1420,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.7,
+      "learning_rate": 8.591549295774647e-05,
+      "loss": 0.9777,
+      "step": 200
+    },
+    {
+      "epoch": 0.7,
+      "eval_accuracy": 0.8169013857841492,
+      "eval_loss": 0.6019795536994934,
+      "eval_runtime": 117.4868,
+      "eval_samples_per_second": 4.835,
+      "step": 200
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 7.183098591549297e-05,
+      "loss": 0.5253,
+      "step": 400
+    },
+    {
+      "epoch": 1.41,
+      "eval_accuracy": 0.8573943376541138,
+      "eval_loss": 0.44053006172180176,
+      "eval_runtime": 130.6116,
+      "eval_samples_per_second": 4.349,
+      "step": 400
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 5.774647887323944e-05,
+      "loss": 0.3972,
+      "step": 600
+    },
+    {
+      "epoch": 2.11,
+      "eval_accuracy": 0.8802816867828369,
+      "eval_loss": 0.3814992308616638,
+      "eval_runtime": 131.0373,
+      "eval_samples_per_second": 4.335,
+      "step": 600
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 4.366197183098591e-05,
+      "loss": 0.2095,
+      "step": 800
+    },
+    {
+      "epoch": 2.82,
+      "eval_accuracy": 0.8661971688270569,
+      "eval_loss": 0.4881167411804199,
+      "eval_runtime": 116.8147,
+      "eval_samples_per_second": 4.862,
+      "step": 800
+    },
+    {
+      "epoch": 3.52,
+      "learning_rate": 2.9577464788732395e-05,
+      "loss": 0.126,
+      "step": 1000
+    },
+    {
+      "epoch": 3.52,
+      "eval_accuracy": 0.9014084339141846,
+      "eval_loss": 0.37247011065483093,
+      "eval_runtime": 130.6973,
+      "eval_samples_per_second": 4.346,
+      "step": 1000
+    },
+    {
+      "epoch": 4.23,
+      "learning_rate": 1.556338028169014e-05,
+      "loss": 0.0865,
+      "step": 1200
+    },
+    {
+      "epoch": 4.23,
+      "eval_accuracy": 0.8978873491287231,
+      "eval_loss": 0.40177592635154724,
+      "eval_runtime": 131.5193,
+      "eval_samples_per_second": 4.319,
+      "step": 1200
+    },
+    {
+      "epoch": 4.93,
+      "learning_rate": 1.4788732394366198e-06,
+      "loss": 0.0579,
+      "step": 1400
+    },
+    {
+      "epoch": 4.93,
+      "eval_accuracy": 0.9031690359115601,
+      "eval_loss": 0.3632720112800598,
+      "eval_runtime": 116.9733,
+      "eval_samples_per_second": 4.856,
+      "step": 1400
+    },
+    {
+      "epoch": 5.0,
+      "step": 1420,
+      "total_flos": 3.369279479928036e+16,
+      "train_runtime": 6152.5452,
+      "train_samples_per_second": 0.231
+    }
+  ],
+  "max_steps": 1420,
+  "num_train_epochs": 5,
+  "total_flos": 3.369279479928036e+16,
+  "trial_name": null,
+  "trial_params": null
+}

vocab.json

@@ -0,0 +1 @@
+{"<pad>": 0, "<s>": 1, "</s>": 2, "<unk>": 3, "|": 4, "آ": 5, "ئ": 6, "ا": 7, "ب": 8, "ت": 9, "ث": 10, "ج": 11, "ح": 12, "خ": 13, "د": 14, "ذ": 15, "ر": 16, "ز": 17, "س": 18, "ش": 19, "ص": 20, "ض": 21, "ط": 22, "ظ": 23, "ع": 24, "غ": 25, "ف": 26, "ق": 27, "ل": 28, "م": 29, "ن": 30, "ه": 31, "و": 32, "پ": 33, "چ": 34, "ژ": 35, "ک": 36, "گ": 37, "ی": 38, "‌": 39}