Initial model

README.md

@@ -0,0 +1,306 @@
+---
+language: tr
+datasets:
+- common_voice
+tags:
+- audio
+- automatic-speech-recognition
+- speech
+- xlsr-fine-tuning-week
+license: apache-2.0
+widget:
+- label: Common Voice sample 1378
+  src: https://huggingface.co/m3hrdadfi/wav2vec2-large-xlsr-turkish/resolve/main/sample1378.flac
+- label: Common Voice sample 1589
+  src: https://huggingface.co/m3hrdadfi/wav2vec2-large-xlsr-turkish/resolve/main/sample1589.flac
+model-index:
+- name: XLSR Wav2Vec2 Turkish by Mehrdad Farahani
+  results:
+  - task: 
+      name: Speech Recognition
+      type: automatic-speech-recognition
+    dataset:
+      name: Common Voice tr
+      type: common_voice
+      args: tr
+    metrics:
+       - name: Test WER
+         type: wer
+         value: 27.51
+        
+---
+
+# Wav2Vec2-Large-XLSR-53-Turkish
+
+Fine-tuned [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) in Turkish using [Common Voice](https://huggingface.co/datasets/common_voice). When using this model, make sure that your speech input is sampled at 16kHz.
+
+## Usage
+The model can be used directly (without a language model) as follows:
+
+**Requirements**
+```bash
+# requirement packages
+!pip install git+https://github.com/huggingface/datasets.git
+!pip install git+https://github.com/huggingface/transformers.git
+!pip install torchaudio
+!pip install librosa
+!pip install jiwer
+```
+
+
+**Prediction**
+```python
+import librosa
+import torch
+import torchaudio
+from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
+from datasets import load_dataset
+
+import numpy as np
+import re
+import string
+
+import IPython.display as ipd
+
+chars_to_ignore = [
+    ",", "?", ".", "!", "-", ";", ":", '""', "%", "'", '"', "�",
+    "#", "!", "?", "«", "»", "(", ")", "؛", ",", "?", ".", "!", "-", ";", ":", '"', 
+    "“", "%", "‘", "�", "–", "…", "_", "”", '“', '„'
+]
+chars_to_mapping = {
+"\u200c": " ", "\u200d": " ", "\u200e": " ", "\u200f": " ", "\ufeff": " ",
+}
+
+def multiple_replace(text, chars_to_mapping):
+    pattern = "|".join(map(re.escape, chars_to_mapping.keys()))
+    return re.sub(pattern, lambda m: chars_to_mapping[m.group()], str(text))
+
+def remove_special_characters(text, chars_to_ignore_regex):
+    text = re.sub(chars_to_ignore_regex, '', text).lower() + " "
+    return text
+
+def normalizer(batch, chars_to_ignore, chars_to_mapping):
+    chars_to_ignore_regex = f"""[{"".join(chars_to_ignore)}]"""
+    text = batch["sentence"].lower().strip()
+    
+    text = text.replace("\u0307", " ").strip()
+    text = multiple_replace(text, chars_to_mapping)
+    text = remove_special_characters(text, chars_to_ignore_regex)
+
+    batch["sentence"] = text
+    return batch
+
+
+def speech_file_to_array_fn(batch):
+    speech_array, sampling_rate = torchaudio.load(batch["path"])
+    speech_array = speech_array.squeeze().numpy()
+    speech_array = librosa.resample(np.asarray(speech_array), sampling_rate, 16_000)
+
+    batch["speech"] = speech_array
+    return batch
+
+
+def predict(batch):
+    features = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)
+
+    input_values = features.input_values.to(device)
+    attention_mask = features.attention_mask.to(device)
+
+    with torch.no_grad():
+        logits = model(input_values, attention_mask=attention_mask).logits 
+        
+    pred_ids = torch.argmax(logits, dim=-1)
+
+    batch["predicted"] = processor.batch_decode(pred_ids)[0]
+    return batch
+
+
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+processor = Wav2Vec2Processor.from_pretrained("m3hrdadfi/wav2vec2-large-xlsr-turkish")
+model = Wav2Vec2ForCTC.from_pretrained("m3hrdadfi/wav2vec2-large-xlsr-turkish").to(device)
+
+dataset = load_dataset("common_voice", "et", split="test[:1%]")
+dataset = dataset.map(
+    normalizer, 
+    fn_kwargs={"chars_to_ignore": chars_to_ignore, "chars_to_mapping": chars_to_mapping},
+    remove_columns=list(set(dataset.column_names) - set(['sentence', 'path']))
+)
+
+dataset = dataset.map(speech_file_to_array_fn)
+result = dataset.map(predict)
+
+max_items = np.random.randint(0, len(result), 10).tolist()
+for i in max_items:
+    reference, predicted =  result["sentence"][i], result["predicted"][i]
+    print("reference:", reference)
+    print("predicted:", predicted)
+    print('---')
+```
+
+**Output:**
+```text
+reference: ülke şu anda iki federasyona üye 
+predicted: ülke şu anda iki federasyona üye
+---
+reference: foruma dört yüzde fazla kişi katıldı 
+predicted: soruma dört yüzden fazla kişi katıldı
+---
+reference: mobi altmış üç çalışanları da mutsuz 
+predicted: mobia haltmış üç çalışanları da mutsur
+---
+reference: kentin mali esnekliğinin düşük olduğu bildirildi 
+predicted: kentin mali esnekleğinin düşük olduğu bildirildi
+---
+reference: fouere iki ülkeyi sorunu abartmamaya çağırdı 
+predicted: foor iki ülkeyi soruna abartmamaya çanayordı
+---
+reference: o ülkeden herhangi bir tepki geldi mi 
+predicted: o ülkeden herhayın bir tepki geldi mi
+---
+reference: bunlara asla sırtımızı dönmeyeceğiz 
+predicted: bunlara asla sırtımızı dönmeyeceğiz
+---
+reference: sizi ayakta tutan nedir 
+predicted: sizi ayakta tutan nedir
+---
+reference: artık insanlar daha bireysel yaşıyor 
+predicted: artık insanlar daha bir eyselli yaşıyor
+---
+reference: her ikisi de diyaloga hazır olduğunu söylüyor 
+predicted: her ikisi de diyaloğa hazır olduğunu söylüyor
+---
+reference: merkez bankasının başlıca amacı düşük enflasyon 
+predicted: merkez bankasının başlrıca anatı güşükyen flasyon
+---
+reference: firefox 
+predicted: fair foks
+---
+reference: ülke halkı çok misafirsever ve dışa dönük 
+predicted: ülke halktı çok isatirtever ve dışa dönük
+---
+reference: ancak kamuoyu bu durumu pek de affetmiyor 
+predicted: ancak kamuonyulgukirmu pek deafıf etmiyor
+---
+reference: i ki madende iki bin beş yüzden fazla kişi çalışıyor 
+predicted: i ki madende iki bin beş yüzden fazla kişi çalışıyor
+---
+reference: sunnyside park dışarıdan oldukça iyi görünüyor 
+predicted: sani sahip park dışarıdan oldukça iyi görünüyor
+---
+reference: büyük ödül on beş bin avro 
+predicted: büyük ödül on beş bin avro
+---
+reference: köyümdeki camiler depoya dönüştürüldü 
+predicted: küyümdeki camiler depoya dönüştürüldü
+---
+reference: maç oldukça diplomatik bir sonuçla birbir bitti 
+predicted: maç oldukça diplomatik bir sonuçla bir birbitti
+---
+reference: kuşların ikisi de karantinada öldüler 
+predicted: kuşların ikiste karantinada özdüler
+---
+```
+
+
+## Evaluation
+
+The model can be evaluated as follows on the Turkish test data of Common Voice.
+
+```python
+import librosa
+import torch
+import torchaudio
+from transformers import Wav2Vec2ForCTC, Wav2Vec2Processor
+from datasets import load_dataset, load_metric
+
+import numpy as np
+import re
+import string
+
+
+chars_to_ignore = [
+    ",", "?", ".", "!", "-", ";", ":", '""', "%", "'", '"', "�",
+    "#", "!", "?", "«", "»", "(", ")", "؛", ",", "?", ".", "!", "-", ";", ":", '"', 
+    "“", "%", "‘", "�", "–", "…", "_", "”", '“', '„'
+]
+chars_to_mapping = {
+    "\u200c": " ", "\u200d": " ", "\u200e": " ", "\u200f": " ", "\ufeff": " ",
+    "\u0307": " "
+}
+
+def multiple_replace(text, chars_to_mapping):
+    pattern = "|".join(map(re.escape, chars_to_mapping.keys()))
+    return re.sub(pattern, lambda m: chars_to_mapping[m.group()], str(text))
+
+def remove_special_characters(text, chars_to_ignore_regex):
+    text = re.sub(chars_to_ignore_regex, '', text).lower() + " "
+    return text
+
+def normalizer(batch, chars_to_ignore, chars_to_mapping):
+    chars_to_ignore_regex = f"""[{"".join(chars_to_ignore)}]"""
+    text = batch["sentence"].lower().strip()
+    
+    text = text.replace("\u0307", " ").strip()
+    text = multiple_replace(text, chars_to_mapping)
+    text = remove_special_characters(text, chars_to_ignore_regex)
+    text = re.sub(" +", " ", text)
+    text = text.strip() + " "
+
+    batch["sentence"] = text
+    return batch
+
+
+def speech_file_to_array_fn(batch):
+    speech_array, sampling_rate = torchaudio.load(batch["path"])
+    speech_array = speech_array.squeeze().numpy()
+    speech_array = librosa.resample(np.asarray(speech_array), sampling_rate, 16_000)
+
+    batch["speech"] = speech_array
+    return batch
+
+
+def predict(batch):
+    features = processor(batch["speech"], sampling_rate=16_000, return_tensors="pt", padding=True)
+
+    input_values = features.input_values.to(device)
+    attention_mask = features.attention_mask.to(device)
+
+    with torch.no_grad():
+        logits = model(input_values, attention_mask=attention_mask).logits 
+        
+    pred_ids = torch.argmax(logits, dim=-1)
+
+    batch["predicted"] = processor.batch_decode(pred_ids)[0]
+    return batch
+
+
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+processor = Wav2Vec2Processor.from_pretrained("m3hrdadfi/wav2vec2-large-xlsr-turkish")
+model = Wav2Vec2ForCTC.from_pretrained("m3hrdadfi/wav2vec2-large-xlsr-turkish").to(device)
+
+dataset = load_dataset("common_voice", "tr", split="test")
+dataset = dataset.map(
+    normalizer, 
+    fn_kwargs={"chars_to_ignore": chars_to_ignore, "chars_to_mapping": chars_to_mapping},
+    remove_columns=list(set(dataset.column_names) - set(['sentence', 'path']))
+)
+
+dataset = dataset.map(speech_file_to_array_fn)
+result = dataset.map(predict)
+
+wer = load_metric("wer")
+
+print("WER: {:.2f}".format(100 * wer.compute(predictions=result["predicted"], references=result["sentence"])))
+```
+]
+
+**Test Result**: 
+- WER: 27.51%
+
+
+## Training & Report
+The Common Voice `train`, `validation` datasets were used for training.
+
+You can see the training states [here](https://wandb.ai/m3hrdadfi/finetuned_wav2vec_xlsr_turkish/reports/Fine-Tuning-for-Wav2Vec2-Large-XLSR-53-Turkish--Vmlldzo1Njc1MDc?accessToken=02vm5cwbi7d342vyt7h9w9859zex0enltdmjoreyjt3bd5qwv0vs0g3u93iv92q0)
+
+The script used for training can be found [here](https://colab.research.google.com/github/m3hrdadfi/notebooks/blob/main/Fine_Tune_XLSR_Wav2Vec2_on_Turkish_ASR_with_%F0%9F%A4%97_Transformers_ipynb.ipynb)

all_results.json

@@ -0,0 +1,24 @@
+{
+    "epoch": 79.99,
+    "eval_loss": 0.43752962350845337,
+    "eval_mem_cpu_alloc_delta": 135991777,
+    "eval_mem_cpu_peaked_delta": 12055738,
+    "eval_mem_gpu_alloc_delta": 0,
+    "eval_mem_gpu_peaked_delta": 5735247360,
+    "eval_runtime": 168.8086,
+    "eval_samples": 1647,
+    "eval_samples_per_second": 9.757,
+    "eval_wer": 0.29045685279187816,
+    "init_mem_cpu_alloc_delta": 51353,
+    "init_mem_cpu_peaked_delta": 18306,
+    "init_mem_gpu_alloc_delta": 0,
+    "init_mem_gpu_peaked_delta": 0,
+    "total_flos": 3.4759190857567523e+19,
+    "train_mem_cpu_alloc_delta": 34575838,
+    "train_mem_cpu_peaked_delta": 171805020,
+    "train_mem_gpu_alloc_delta": 3790510080,
+    "train_mem_gpu_peaked_delta": 6257791488,
+    "train_runtime": 32940.4351,
+    "train_samples": 3478,
+    "train_samples_per_second": 0.151
+}

config.json

@@ -0,0 +1,76 @@
+{
+  "_name_or_path": "facebook/wav2vec2-large-xlsr-53",
+  "activation_dropout": 0.0,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForCTC"
+  ],
+  "attention_dropout": 0.1,
+  "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": 0.0,
+  "final_dropout": 0.0,
+  "gradient_checkpointing": true,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.1,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-05,
+  "layerdrop": 0.1,
+  "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": 0.05,
+  "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,
+  "transformers_version": "4.5.0.dev0",
+  "vocab_size": 40
+}

eval_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 79.99,
+    "eval_loss": 0.43752962350845337,
+    "eval_mem_cpu_alloc_delta": 135991777,
+    "eval_mem_cpu_peaked_delta": 12055738,
+    "eval_mem_gpu_alloc_delta": 0,
+    "eval_mem_gpu_peaked_delta": 5735247360,
+    "eval_runtime": 168.8086,
+    "eval_samples": 1647,
+    "eval_samples_per_second": 9.757,
+    "eval_wer": 0.29045685279187816
+}

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
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7231dcadd7d585fe2d05a7dc13addf8d6c70d29a6b781f2a4a7c0d42e674225
+size 1262097815

result.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1e91eac457062cde32dd686155804964e2a496d34a50d274ff5b92478d287e9
+size 3183

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": "|"}

train_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 79.99,
+    "init_mem_cpu_alloc_delta": 51353,
+    "init_mem_cpu_peaked_delta": 18306,
+    "init_mem_gpu_alloc_delta": 0,
+    "init_mem_gpu_peaked_delta": 0,
+    "total_flos": 3.4759190857567523e+19,
+    "train_mem_cpu_alloc_delta": 34575838,
+    "train_mem_cpu_peaked_delta": 171805020,
+    "train_mem_gpu_alloc_delta": 3790510080,
+    "train_mem_gpu_peaked_delta": 6257791488,
+    "train_runtime": 32940.4351,
+    "train_samples": 3478,
+    "train_samples_per_second": 0.151
+}

trainer_state.json

@@ -0,0 +1,247 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 79.992,
+  "global_step": 4960,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 4.83,
+      "learning_rate": 0.00017999999999999998,
+      "loss": 5.8789,
+      "step": 300
+    },
+    {
+      "epoch": 4.83,
+      "eval_loss": 3.1512646675109863,
+      "eval_runtime": 167.7164,
+      "eval_samples_per_second": 9.82,
+      "eval_wer": 1.0,
+      "step": 300
+    },
+    {
+      "epoch": 9.67,
+      "learning_rate": 0.00029327354260089687,
+      "loss": 1.8544,
+      "step": 600
+    },
+    {
+      "epoch": 9.67,
+      "eval_loss": 0.4943191707134247,
+      "eval_runtime": 170.9272,
+      "eval_samples_per_second": 9.636,
+      "eval_wer": 0.5103553299492386,
+      "step": 600
+    },
+    {
+      "epoch": 14.51,
+      "learning_rate": 0.0002730941704035874,
+      "loss": 0.2628,
+      "step": 900
+    },
+    {
+      "epoch": 14.51,
+      "eval_loss": 0.46906277537345886,
+      "eval_runtime": 172.4339,
+      "eval_samples_per_second": 9.551,
+      "eval_wer": 0.43147208121827413,
+      "step": 900
+    },
+    {
+      "epoch": 19.35,
+      "learning_rate": 0.000252914798206278,
+      "loss": 0.124,
+      "step": 1200
+    },
+    {
+      "epoch": 19.35,
+      "eval_loss": 0.4130130708217621,
+      "eval_runtime": 172.8773,
+      "eval_samples_per_second": 9.527,
+      "eval_wer": 0.36,
+      "step": 1200
+    },
+    {
+      "epoch": 24.19,
+      "learning_rate": 0.0002327354260089686,
+      "loss": 0.0857,
+      "step": 1500
+    },
+    {
+      "epoch": 24.19,
+      "eval_loss": 0.43019285798072815,
+      "eval_runtime": 176.7931,
+      "eval_samples_per_second": 9.316,
+      "eval_wer": 0.3513705583756345,
+      "step": 1500
+    },
+    {
+      "epoch": 29.03,
+      "learning_rate": 0.00021255605381165918,
+      "loss": 0.0653,
+      "step": 1800
+    },
+    {
+      "epoch": 29.03,
+      "eval_loss": 0.4362075626850128,
+      "eval_runtime": 190.6223,
+      "eval_samples_per_second": 8.64,
+      "eval_wer": 0.3595939086294416,
+      "step": 1800
+    },
+    {
+      "epoch": 33.86,
+      "learning_rate": 0.00019237668161434975,
+      "loss": 0.0632,
+      "step": 2100
+    },
+    {
+      "epoch": 33.86,
+      "eval_loss": 0.4273272156715393,
+      "eval_runtime": 177.8151,
+      "eval_samples_per_second": 9.262,
+      "eval_wer": 0.33390862944162436,
+      "step": 2100
+    },
+    {
+      "epoch": 38.7,
+      "learning_rate": 0.00017219730941704035,
+      "loss": 0.0499,
+      "step": 2400
+    },
+    {
+      "epoch": 38.7,
+      "eval_loss": 0.4455905258655548,
+      "eval_runtime": 177.4805,
+      "eval_samples_per_second": 9.28,
+      "eval_wer": 0.32558375634517767,
+      "step": 2400
+    },
+    {
+      "epoch": 43.54,
+      "learning_rate": 0.00015201793721973095,
+      "loss": 0.0412,
+      "step": 2700
+    },
+    {
+      "epoch": 43.54,
+      "eval_loss": 0.4279979467391968,
+      "eval_runtime": 179.3129,
+      "eval_samples_per_second": 9.185,
+      "eval_wer": 0.3316751269035533,
+      "step": 2700
+    },
+    {
+      "epoch": 48.38,
+      "learning_rate": 0.00013183856502242152,
+      "loss": 0.0428,
+      "step": 3000
+    },
+    {
+      "epoch": 48.38,
+      "eval_loss": 0.42648839950561523,
+      "eval_runtime": 179.9053,
+      "eval_samples_per_second": 9.155,
+      "eval_wer": 0.3197969543147208,
+      "step": 3000
+    },
+    {
+      "epoch": 53.22,
+      "learning_rate": 0.0001116591928251121,
+      "loss": 0.0345,
+      "step": 3300
+    },
+    {
+      "epoch": 53.22,
+      "eval_loss": 0.46762773394584656,
+      "eval_runtime": 179.9553,
+      "eval_samples_per_second": 9.152,
+      "eval_wer": 0.31228426395939085,
+      "step": 3300
+    },
+    {
+      "epoch": 58.06,
+      "learning_rate": 9.147982062780269e-05,
+      "loss": 0.0335,
+      "step": 3600
+    },
+    {
+      "epoch": 58.06,
+      "eval_loss": 0.4403984248638153,
+      "eval_runtime": 181.1578,
+      "eval_samples_per_second": 9.092,
+      "eval_wer": 0.3096446700507614,
+      "step": 3600
+    },
+    {
+      "epoch": 62.9,
+      "learning_rate": 7.130044843049327e-05,
+      "loss": 0.0308,
+      "step": 3900
+    },
+    {
+      "epoch": 62.9,
+      "eval_loss": 0.4584444463253021,
+      "eval_runtime": 182.2608,
+      "eval_samples_per_second": 9.036,
+      "eval_wer": 0.3082233502538071,
+      "step": 3900
+    },
+    {
+      "epoch": 67.74,
+      "learning_rate": 5.112107623318385e-05,
+      "loss": 0.0253,
+      "step": 4200
+    },
+    {
+      "epoch": 67.74,
+      "eval_loss": 0.4203069806098938,
+      "eval_runtime": 182.5783,
+      "eval_samples_per_second": 9.021,
+      "eval_wer": 0.3001015228426396,
+      "step": 4200
+    },
+    {
+      "epoch": 72.58,
+      "learning_rate": 3.094170403587444e-05,
+      "loss": 0.0243,
+      "step": 4500
+    },
+    {
+      "epoch": 72.58,
+      "eval_loss": 0.43954339623451233,
+      "eval_runtime": 183.1424,
+      "eval_samples_per_second": 8.993,
+      "eval_wer": 0.29736040609137054,
+      "step": 4500
+    },
+    {
+      "epoch": 77.42,
+      "learning_rate": 1.0762331838565022e-05,
+      "loss": 0.0229,
+      "step": 4800
+    },
+    {
+      "epoch": 77.42,
+      "eval_loss": 0.44123971462249756,
+      "eval_runtime": 183.8976,
+      "eval_samples_per_second": 8.956,
+      "eval_wer": 0.29218274111675124,
+      "step": 4800
+    },
+    {
+      "epoch": 79.99,
+      "step": 4960,
+      "total_flos": 3.4759190857567523e+19,
+      "train_runtime": 32940.4351,
+      "train_samples_per_second": 0.151
+    }
+  ],
+  "max_steps": 4960,
+  "num_train_epochs": 80,
+  "total_flos": 3.4759190857567523e+19,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7743363a3e402661db5a2194639a678362707f82d18b359e7fe33276471e0169
+size 2351

vocab.json

@@ -0,0 +1 @@
+{"<pad>": 0, "<s>": 1, "</s>": 2, "<unk>": 3, "|": 4, "a": 5, "b": 6, "c": 7, "d": 8, "e": 9, "f": 10, "g": 11, "h": 12, "i": 13, "j": 14, "k": 15, "l": 16, "m": 17, "n": 18, "o": 19, "p": 20, "q": 21, "r": 22, "s": 23, "t": 24, "u": 25, "v": 26, "w": 27, "x": 28, "y": 29, "z": 30, "â": 31, "ç": 32, "ë": 33, "î": 34, "ö": 35, "ü": 36, "ğ": 37, "ı": 38, "ş": 39}