From 27503d5d381598423ca89d4516bf7c59d5edbd8f Mon Sep 17 00:00:00 2001
From: Ceyda Cinarel <15624271+cceyda@users.noreply.github.com>
Date: Thu, 1 Apr 2021 01:05:49 +0000
Subject: [PATCH] first commit

---
 README.md                |  8 ++---
 config.json              | 67 ++++++++++++++++++++++++++++++++++++++++
 preprocessor_config.json |  8 +++++
 pytorch_model.bin        |  3 ++
 special_tokens_map.json  |  1 +
 tokenizer_config.json    |  1 +
 vocab.json               |  1 +
 7 files changed, 85 insertions(+), 4 deletions(-)
 create mode 100644 config.json
 create mode 100644 preprocessor_config.json
 create mode 100644 pytorch_model.bin
 create mode 100644 special_tokens_map.json
 create mode 100644 tokenizer_config.json
 create mode 100644 vocab.json

diff --git a/README.md b/README.md
index ff1f6f6..63c5f7e 100644
--- a/README.md
+++ b/README.md
@@ -23,10 +23,10 @@ model-index:
     metrics:
        - name: Test WER
          type: wer
-         value: 29.30
+         value: 27.08
 ---
 
-# Wav2Vec2-Base-960-Turkish
+# Wav2Vec2-Base-760-Turkish
 
 # TBA
 Pretrained Turkish model [ceyda/wav2vec2-base-760](https://huggingface.co/ceyda/wav2vec2-base-760). Fine-tuned on Turkish using the [Common Voice](https://huggingface.co/datasets/common_voice)
@@ -87,7 +87,7 @@ processor = Wav2Vec2Processor.from_pretrained("ceyda/wav2vec2-base-960-turkish")
 model = Wav2Vec2ForCTC.from_pretrained("ceyda/wav2vec2-base-960-turkish")
 model.to("cuda")
 
-chars_to_ignore_regex = '[\\,\\?\\.\\!\\-\\;\\:\\"\\“\\‘\\”\\'\\`…\\’»«]'
+chars_to_ignore_regex = '[\,\?\.\!\-\;\:\"\“\‘\”\'\`…\’»«]'
 resampler = torchaudio.transforms.Resample(48_000, 16_000)
 
 # Preprocessing the datasets.
@@ -117,7 +117,7 @@ result = test_dataset.map(evaluate, batched=True, batch_size=8)
 print("WER: {:2f}".format(100 * wer.compute(predictions=result["pred_strings"], references=result["sentence"])))
 ```
 
-**Test Result**: XX.XX % (TBA)
+**Test Result**: 27.08 % (in progress)
 
 
 ## Training
diff --git a/config.json b/config.json
new file mode 100644
index 0000000..cf53f72
--- /dev/null
+++ b/config.json
@@ -0,0 +1,67 @@
+{
+  "_name_or_path": "./pretrained/checkpoint_0.60",
+  "activation_dropout": 0.1,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForCTC"
+  ],
+  "attention_dropout": 0.1,
+  "bos_token_id": 1,
+  "conv_bias": false,
+  "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": false,
+  "eos_token_id": 2,
+  "feat_extract_activation": "gelu",
+  "feat_extract_norm": "group",
+  "feat_proj_dropout": 0.1,
+  "final_dropout": 0.1,
+  "gradient_checkpointing": true,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.05,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.1,
+  "mask_feature_length": 10,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_prob": 0.5,
+  "model_type": "wav2vec2",
+  "num_attention_heads": 12,
+  "num_conv_pos_embedding_groups": 16,
+  "num_conv_pos_embeddings": 128,
+  "num_feat_extract_layers": 7,
+  "num_hidden_layers": 12,
+  "pad_token_id": 36,
+  "transformers_version": "4.5.0.dev0",
+  "vocab_size": 37
+}
diff --git a/preprocessor_config.json b/preprocessor_config.json
new file mode 100644
index 0000000..0886a48
--- /dev/null
+++ b/preprocessor_config.json
@@ -0,0 +1,8 @@
+{
+  "do_normalize": true,
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}
diff --git a/pytorch_model.bin b/pytorch_model.bin
new file mode 100644
index 0000000..6b20cde
--- /dev/null
+++ b/pytorch_model.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ecb5a521af5105ca67abd19a841c08e97b071a8b69e186b06b25633c6fc36804
+size 377690860
diff --git a/special_tokens_map.json b/special_tokens_map.json
new file mode 100644
index 0000000..9abf719
--- /dev/null
+++ b/special_tokens_map.json
@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]"}
\ No newline at end of file
diff --git a/tokenizer_config.json b/tokenizer_config.json
new file mode 100644
index 0000000..a2a8340
--- /dev/null
+++ b/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": "|"}
\ No newline at end of file
diff --git a/vocab.json b/vocab.json
new file mode 100644
index 0000000..7f974b7
--- /dev/null
+++ b/vocab.json
@@ -0,0 +1 @@
+{"e": 0, "x": 1, "î": 2, "l": 3, "â": 4, "j": 5, "ç": 6, "ş": 7, "g": 8, "ı": 9, "v": 10, "d": 11, "t": 12, "n": 13, "a": 14, "c": 15, "h": 16, "p": 17, "r": 18, "w": 19, "z": 20, "k": 21, "u": 22, "b": 23, "ü": 24, "y": 26, "o": 27, "q": 28, "m": 29, "f": 30, "s": 31, "ö": 32, "ğ": 33, "i": 34, "|": 25, "[UNK]": 35, "[PAD]": 36}
\ No newline at end of file