Model save

.ipynb_checkpoints/README-checkpoint.md

@@ -0,0 +1,70 @@
+---
+language:
+- fr
+license: apache-2.0
+tags:
+- automatic-speech-recognition
+- mozilla-foundation/common_voice_7_0
+- generated_from_trainer
+datasets:
+- common_voice
+model-index:
+- name: ''
+  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. -->
+
+# 
+
+This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the MOZILLA-FOUNDATION/COMMON_VOICE_7_0 - FR dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.5417
+- Wer: 0.4479
+
+## 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: 7.5e-05
+- train_batch_size: 8
+- eval_batch_size: 8
+- seed: 42
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 32
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 2000
+- num_epochs: 0.2
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Wer    |
+|:-------------:|:-----:|:----:|:---------------:|:------:|
+| 6.9106        | 0.04  | 500  | 6.7171          | 1.0    |
+| 3.0034        | 0.08  | 1000 | 3.0126          | 1.0    |
+| 2.8699        | 0.12  | 1500 | 2.8509          | 0.9817 |
+| 1.629         | 0.16  | 2000 | 0.7787          | 0.5861 |
+
+
+### Framework versions
+
+- Transformers 4.17.0.dev0
+- Pytorch 1.10.2+cu102
+- Datasets 1.18.2.dev0
+- Tokenizers 0.11.0

.ipynb_checkpoints/config-checkpoint.json

@@ -1,107 +0,0 @@
-{
-  "_name_or_path": "facebook/wav2vec2-xls-r-300m",
-  "activation_dropout": 0.1,
-  "adapter_kernel_size": 3,
-  "adapter_stride": 2,
-  "add_adapter": false,
-  "apply_spec_augment": true,
-  "architectures": [
-    "Wav2Vec2ForPreTraining"
-  ],
-  "attention_dropout": 0.0,
-  "bos_token_id": 1,
-  "classifier_proj_size": 256,
-  "codevector_dim": 768,
-  "contrastive_logits_temperature": 0.1,
-  "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": false,
-  "diversity_loss_weight": 0.1,
-  "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,
-  "feat_quantizer_dropout": 0.0,
-  "final_dropout": 0.0,
-  "hidden_act": "gelu",
-  "hidden_dropout": 0.0,
-  "hidden_size": 1024,
-  "initializer_range": 0.02,
-  "intermediate_size": 4096,
-  "layer_norm_eps": 1e-05,
-  "layerdrop": 0.0,
-  "mask_feature_length": 64,
-  "mask_feature_min_masks": 0,
-  "mask_feature_prob": 0.25,
-  "mask_time_length": 10,
-  "mask_time_min_masks": 2,
-  "mask_time_prob": 0.75,
-  "model_type": "wav2vec2",
-  "num_adapter_layers": 3,
-  "num_attention_heads": 16,
-  "num_codevector_groups": 2,
-  "num_codevectors_per_group": 320,
-  "num_conv_pos_embedding_groups": 16,
-  "num_conv_pos_embeddings": 128,
-  "num_feat_extract_layers": 7,
-  "num_hidden_layers": 24,
-  "num_negatives": 100,
-  "output_hidden_size": 1024,
-  "pad_token_id": 283,
-  "proj_codevector_dim": 768,
-  "tdnn_dilation": [
-    1,
-    2,
-    3,
-    1,
-    1
-  ],
-  "tdnn_dim": [
-    512,
-    512,
-    512,
-    512,
-    1500
-  ],
-  "tdnn_kernel": [
-    5,
-    3,
-    3,
-    1,
-    1
-  ],
-  "torch_dtype": "float32",
-  "transformers_version": "4.17.0.dev0",
-  "use_weighted_layer_sum": false,
-  "vocab_size": 284,
-  "xvector_output_dim": 512
-}

.ipynb_checkpoints/eval-checkpoint.py

@@ -49,9 +49,9 @@ def log_results(result: Dataset, args: Dict[str, str]):
 def normalize_text(text: str) -> str:
     """DO ADAPT FOR YOUR USE CASE. this function normalizes the target text."""
 
-    chars_to_ignore_regex = '[,?.!-;:"“%‘”�—…–=^_`{|}~£§«®°±´µ·º»½×ßáãäìíðñòóõöøýþÿāăąćċčďđēėęěğġħĩī생집]'  # noqa: W605 IMPORTANT: this should correspond to the chars that were ignored during training
+    chars_to_ignore_regex = '[^a-zàâäçéèêëîïôöùûüÿ\'’ ]'  # noqa: W605 IMPORTANT: this should correspond to the chars that were ignored during training
 
-    text = re.sub(chars_to_ignore_regex, "", text.lower())
+    text = re.sub(chars_to_ignore_regex, "", text.lower()).replace('’', "'")
 
     # In addition, we can normalize the target text, e.g. removing new lines characters etc...
     # note that order is important here!
@@ -68,7 +68,7 @@ def main(args):
     dataset = load_dataset(args.dataset, args.config, split=args.split, use_auth_token=True)
 
     # for testing: only process the first two examples as a test
-    dataset = dataset.select(range(10))
+    # dataset = dataset.select(range(10))
 
     # load processor
     feature_extractor = AutoFeatureExtractor.from_pretrained(args.model_id)
@@ -86,7 +86,7 @@ def main(args):
             batch["audio"]["array"], chunk_length_s=args.chunk_length_s, stride_length_s=args.stride_length_s
         )
 
-        batch["prediction"] = prediction["text"].replace("<s>", "")
+        batch["prediction"] = prediction["text"]
         batch["target"] = normalize_text(batch["sentence"])
         return batch
 

.ipynb_checkpoints/preprocessor_config-checkpoint.json

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

.ipynb_checkpoints/run-checkpoint.sh

@@ -1,33 +1,39 @@
+WANDB_PROJECT=xls-r-300-fr
 python run_speech_recognition_ctc.py \
-	--dataset_name="mozilla-foundation/common_voice_7_0" \
-	--model_name_or_path="facebook/wav2vec2-xls-r-300m" \
-	--dataset_config_name="fr" \
-	--output_dir="./" \
-	--overwrite_output_dir \
-	--num_train_epochs="0.2" \
-	--per_device_train_batch_size="8" \
-	--per_device_eval_batch_size="8" \
-	--gradient_accumulation_steps="4" \
-	--learning_rate="7.5e-5" \
-	--warmup_steps="2000" \
-	--length_column_name="input_length" \
-	--evaluation_strategy="steps" \
-	--text_column_name="sentence" \
-	--save_steps="500" \
-	--eval_steps="500" \
-	--logging_steps="100" \
-	--layerdrop="0.0" \
-	--activation_dropout="0.1" \
-	--save_total_limit="3" \
-	--freeze_feature_encoder \
-	--feat_proj_dropout="0.0" \
-	--mask_time_prob="0.75" \
-	--mask_time_length="10" \
-	--mask_feature_prob="0.25" \
-	--mask_feature_length="64" \
-	--gradient_checkpointing \
-	--use_auth_token \
-	--fp16 \
-	--group_by_length \
-	--do_train --do_eval \
-	--push_to_hub
+    --activation_dropout="0.1" \
+    --dataset_name="mozilla-foundation/common_voice_7_0" \
+    --dataset_config_name="fr" \
+    --eval_steps="500" \
+    --evaluation_strategy="steps" \
+    --feat_proj_dropout="0.0" \
+    --freeze_feature_encoder \
+    --fp16 \
+    --gradient_accumulation_steps="8" \
+    --gradient_checkpointing \
+    --group_by_length \
+    --layerdrop="0.0" \
+    --learning_rate="7.5e-5" \
+    --length_column_name="input_length" \
+    --load_best_model_at_end \
+    --logging_steps="100" \
+    --mask_feature_length="64" \
+    --mask_feature_prob="0.25" \
+    --mask_time_length="10" \
+    --mask_time_prob="0.75" \
+    --max_train_samples="1000" \
+    --max_eval_samples="200" \
+    --model_name_or_path="facebook/wav2vec2-xls-r-300m" \
+    --num_train_epochs="0.2" \
+    --output_dir="./" \
+    --overwrite_output_dir \
+    --per_device_train_batch_size="8" \
+    --per_device_eval_batch_size="8" \
+    --preprocessing_num_workers="4" \
+    --push_to_hub \
+    --report_to="wandb" \
+    --save_steps="500" \
+    --save_total_limit="3" \
+    --text_column_name="sentence" \
+    --use_auth_token \
+    --warmup_steps="2000" \
+    --do_train --do_eval

.ipynb_checkpoints/special_tokens_map-checkpoint.json

@@ -1 +0,0 @@
-{"unk_token": "[UNK]", "pad_token": "[PAD]"}

.ipynb_checkpoints/tokenizer_config-checkpoint.json

@@ -1 +0,0 @@
-{"unk_token": "[UNK]", "bos_token": null, "eos_token": null, "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|", "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "./", "tokenizer_class": "Wav2Vec2CTCTokenizer"}

.ipynb_checkpoints/vocab-checkpoint.json

@@ -1 +0,0 @@
-{"'": 1, "a": 2, "b": 3, "c": 4, "d": 5, "e": 6, "f": 7, "g": 8, "h": 9, "i": 10, "j": 11, "k": 12, "l": 13, "m": 14, "n": 15, "o": 16, "p": 17, "q": 18, "r": 19, "s": 20, "t": 21, "u": 22, "v": 23, "w": 24, "x": 25, "y": 26, "z": 27, "\u00e0": 28, "\u00e1": 29, "\u00e2": 30, "\u00e4": 31, "\u00e5": 32, "\u00e6": 33, "\u00e7": 34, "\u00e8": 35, "\u00e9": 36, "\u00ea": 37, "\u00eb": 38, "\u00ec": 39, "\u00ed": 40, "\u00ee": 41, "\u00ef": 42, "\u00f1": 43, "\u00f2": 44, "\u00f3": 45, "\u00f4": 46, "\u00f5": 47, "\u00f6": 48, "\u00f8": 49, "\u00f9": 50, "\u00fa": 51, "\u00fb": 52, "\u00fc": 53, "\u00fe": 54, "\u00ff": 55, "\u0101": 56, "\u0107": 57, "\u010b": 58, "\u010d": 59, "\u0111": 60, "\u0113": 61, "\u0121": 62, "\u012b": 63, "\u0131": 64, "\u0137": 65, "\u013a": 66, "\u013c": 67, "\u013e": 68, "\u0142": 69, "\u0144": 70, "\u0146": 71, "\u0148": 72, "\u014d": 73, "\u014f": 74, "\u0151": 75, "\u0153": 76, "\u0159": 77, "\u015b": 78, "\u015f": 79, "\u0161": 80, "\u0163": 81, "\u0165": 82, "\u0169": 83, "\u016b": 84, "\u016f": 85, "\u0171": 86, "\u0173": 87, "\u017a": 88, "\u017c": 89, "\u017e": 90, "\u01a1": 91, "\u01b0": 92, "\u01c0": 93, "\u01c3": 94, "\u01ce": 95, "\u01d4": 96, "\u01eb": 97, "\u01f9": 98, "\u0219": 99, "\u021b": 100, "\u0259": 101, "\u0268": 102, "\u0289": 103, "\u0294": 104, "\u02bb": 105, "\u02bc": 106, "\u02bd": 107, "\u02be": 108, "\u02bf": 109, "\u02d0": 110, "\u02e2": 111, "\u0300": 112, "\u0301": 113, "\u0302": 114, "\u0303": 115, "\u0307": 116, "\u0308": 117, "\u0310": 118, "\u0320": 119, "\u0327": 120, "\u0331": 121, "\u0332": 122, "\u03b1": 123, "\u03b2": 124, "\u03b3": 125, "\u03b4": 126, "\u03b5": 127, "\u03b6": 128, "\u03b7": 129, "\u03b8": 130, "\u03b9": 131, "\u03ba": 132, "\u03bb": 133, "\u03bc": 134, "\u03bd": 135, "\u03bf": 136, "\u03c0": 137, "\u03c1": 138, "\u03c2": 139, "\u03c3": 140, "\u03c4": 141, "\u03c5": 142, "\u03c6": 143, "\u03c7": 144, "\u03c8": 145, "\u03c9": 146, "\u03cc": 147, "\u0430": 148, "\u0433": 149, "\u0435": 150, "\u0437": 151, "\u0438": 152, "\u043a": 153, "\u043c": 154, "\u043d": 155, "\u043f": 156, "\u0440": 157, "\u0446": 158, "\u0447": 159, "\u044d": 160, "\u044f": 161, "\u0454": 162, "\u0456": 163, "\u0458": 164, "\u04ab": 165, "\u04cc": 166, "\u0563": 167, "\u0566": 168, "\u0627": 169, "\u0628": 170, "\u0629": 171, "\u062f": 172, "\u0631": 173, "\u0644": 174, "\u0645": 175, "\u0646": 176, "\u0648": 177, "\u064a": 178, "\u1240": 179, "\u12a8": 180, "\u12c8": 181, "\u12f0": 182, "\u1300": 183, "\u1320": 184, "\u1e0d": 185, "\u1e25": 186, "\u1e45": 187, "\u1e47": 188, "\u1e63": 189, "\u1e6d": 190, "\u1e6f": 191, "\u1e93": 192, "\u1ea1": 193, "\u1ea3": 194, "\u1ea7": 195, "\u1ead": 196, "\u1eaf": 197, "\u1eb5": 198, "\u1ebf": 199, "\u1ec1": 200, "\u1ec5": 201, "\u1ec7": 202, "\u1ecb": 203, "\u1ed1": 204, "\u1ed3": 205, "\u1ed5": 206, "\u1ed9": 207, "\u1edb": 208, "\u1edd": 209, "\u1ee3": 210, "\u1ee5": 211, "\u1ee7": 212, "\u1ee9": 213, "\u1eed": 214, "\u1ef3": 215, "\u2010": 216, "\u2015": 217, "\u201e": 218, "\u2020": 219, "\u2032": 220, "\u2039": 221, "\u203a": 222, "\u20ac": 223, "\u20bd": 224, "\u2102": 225, "\u2115": 226, "\u211a": 227, "\u211d": 228, "\u2124": 229, "\u2130": 230, "\u2135": 231, "\u2192": 232, "\u2194": 233, "\u2205": 234, "\u2206": 235, "\u2208": 236, "\u2212": 237, "\u221e": 238, "\u2228": 239, "\u223c": 240, "\u2265": 241, "\u22c5": 242, "\u2500": 243, "\u2609": 244, "\u2c45": 245, "\u2c4e": 246, "\u3044": 247, "\u3046": 248, "\u305f": 249, "\u3064": 250, "\u306c": 251, "\u306e": 252, "\u3072": 253, "\u3078": 254, "\u307e": 255, "\u3080": 256, "\u3081": 257, "\u3082": 258, "\u3084": 259, "\u3094": 260, "\u4e39": 261, "\u4e43": 262, "\u4eac": 263, "\u5317": 264, "\u626c": 265, "\u6587": 266, "\u661f": 267, "\u672f": 268, "\u675c": 269, "\u7261": 270, "\u750c": 271, "\u7f8e": 272, "\u897f": 273, "\u8cb4": 274, "\u9752": 275, "\u9986": 276, "\ua751": 277, "\uace0": 278, "\uae30": 279, "\uba39": 280, "\uc0bc": 281, "|": 0, "[UNK]": 282, "[PAD]": 283}

all_results.json

@@ -1,14 +0,0 @@
-{
-    "epoch": 0.2,
-    "eval_loss": 0.5416641235351562,
-    "eval_runtime": 816.3514,
-    "eval_samples": 15941,
-    "eval_samples_per_second": 19.527,
-    "eval_steps_per_second": 2.441,
-    "eval_wer": 0.4478597919027506,
-    "train_loss": 4.42930416077822,
-    "train_runtime": 8599.8864,
-    "train_samples": 395042,
-    "train_samples_per_second": 9.187,
-    "train_steps_per_second": 0.287
-}

config.json

@@ -76,7 +76,7 @@
   "num_hidden_layers": 24,
   "num_negatives": 100,
   "output_hidden_size": 1024,
-  "pad_token_id": 283,
+  "pad_token_id": 40,
   "proj_codevector_dim": 768,
   "tdnn_dilation": [
     1,
@@ -102,6 +102,6 @@
   "torch_dtype": "float32",
   "transformers_version": "4.17.0.dev0",
   "use_weighted_layer_sum": false,
-  "vocab_size": 284,
+  "vocab_size": 41,
   "xvector_output_dim": 512
 }

eval.py

@@ -49,9 +49,9 @@ def log_results(result: Dataset, args: Dict[str, str]):
 def normalize_text(text: str) -> str:
     """DO ADAPT FOR YOUR USE CASE. this function normalizes the target text."""
 
-    chars_to_ignore_regex = '[,?.!-;:"“%‘”�—…–=^_`{|}~£§«®°±´µ·º»½×ßáãäìíðñòóõöøýþÿāăąćċčďđēėęěğġħĩī생집]'  # noqa: W605 IMPORTANT: this should correspond to the chars that were ignored during training
+    chars_to_ignore_regex = '[^a-zàâäçéèêëîïôöùûüÿ\'’ ]'  # noqa: W605 IMPORTANT: this should correspond to the chars that were ignored during training
 
-    text = re.sub(chars_to_ignore_regex, "", text.lower())
+    text = re.sub(chars_to_ignore_regex, "", text.lower()).replace('’', "'")
 
     # In addition, we can normalize the target text, e.g. removing new lines characters etc...
     # note that order is important here!
@@ -68,7 +68,7 @@ def main(args):
     dataset = load_dataset(args.dataset, args.config, split=args.split, use_auth_token=True)
 
     # for testing: only process the first two examples as a test
-    dataset = dataset.select(range(10))
+    # dataset = dataset.select(range(10))
 
     # load processor
     feature_extractor = AutoFeatureExtractor.from_pretrained(args.model_id)
@@ -86,7 +86,7 @@ def main(args):
             batch["audio"]["array"], chunk_length_s=args.chunk_length_s, stride_length_s=args.stride_length_s
         )
 
-        batch["prediction"] = prediction["text"].replace("<s>", "")
+        batch["prediction"] = prediction["text"]
         batch["target"] = normalize_text(batch["sentence"])
         return batch
 

eval_results.json

@@ -1,9 +0,0 @@
-{
-    "epoch": 0.2,
-    "eval_loss": 0.5416641235351562,
-    "eval_runtime": 816.3514,
-    "eval_samples": 15941,
-    "eval_samples_per_second": 19.527,
-    "eval_steps_per_second": 2.441,
-    "eval_wer": 0.4478597919027506
-}

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3ec1c7675b56877de46f623ac51149e73d4f88cac691dc72595621d35344ce9b
-size 1263088113
+oid sha256:b042f9e3e1a3adceae4bf7dbadeee3279b0dc94f7582075e5635e30c922536ed
+size 1262091761

run.sh

@@ -1,33 +1,39 @@
+WANDB_PROJECT=xls-r-300-fr
 python run_speech_recognition_ctc.py \
-	--dataset_name="mozilla-foundation/common_voice_7_0" \
-	--model_name_or_path="facebook/wav2vec2-xls-r-300m" \
-	--dataset_config_name="fr" \
-	--output_dir="./" \
-	--overwrite_output_dir \
-	--num_train_epochs="0.2" \
-	--per_device_train_batch_size="8" \
-	--per_device_eval_batch_size="8" \
-	--gradient_accumulation_steps="4" \
-	--learning_rate="7.5e-5" \
-	--warmup_steps="2000" \
-	--length_column_name="input_length" \
-	--evaluation_strategy="steps" \
-	--text_column_name="sentence" \
-	--save_steps="500" \
-	--eval_steps="500" \
-	--logging_steps="100" \
-	--layerdrop="0.0" \
-	--activation_dropout="0.1" \
-	--save_total_limit="3" \
-	--freeze_feature_encoder \
-	--feat_proj_dropout="0.0" \
-	--mask_time_prob="0.75" \
-	--mask_time_length="10" \
-	--mask_feature_prob="0.25" \
-	--mask_feature_length="64" \
-	--gradient_checkpointing \
-	--use_auth_token \
-	--fp16 \
-	--group_by_length \
-	--do_train --do_eval \
-	--push_to_hub
+    --activation_dropout="0.1" \
+    --dataset_name="mozilla-foundation/common_voice_7_0" \
+    --dataset_config_name="fr" \
+    --eval_steps="500" \
+    --evaluation_strategy="steps" \
+    --feat_proj_dropout="0.0" \
+    --freeze_feature_encoder \
+    --fp16 \
+    --gradient_accumulation_steps="8" \
+    --gradient_checkpointing \
+    --group_by_length \
+    --layerdrop="0.0" \
+    --learning_rate="7.5e-5" \
+    --length_column_name="input_length" \
+    --load_best_model_at_end \
+    --logging_steps="100" \
+    --mask_feature_length="64" \
+    --mask_feature_prob="0.25" \
+    --mask_time_length="10" \
+    --mask_time_prob="0.75" \
+    --max_train_samples="1000" \
+    --max_eval_samples="200" \
+    --model_name_or_path="facebook/wav2vec2-xls-r-300m" \
+    --num_train_epochs="0.2" \
+    --output_dir="./" \
+    --overwrite_output_dir \
+    --per_device_train_batch_size="8" \
+    --per_device_eval_batch_size="8" \
+    --preprocessing_num_workers="4" \
+    --push_to_hub \
+    --report_to="wandb" \
+    --save_steps="500" \
+    --save_total_limit="3" \
+    --text_column_name="sentence" \
+    --use_auth_token \
+    --warmup_steps="2000" \
+    --do_train --do_eval

train_results.json

@@ -1,8 +0,0 @@
-{
-    "epoch": 0.2,
-    "train_loss": 4.42930416077822,
-    "train_runtime": 8599.8864,
-    "train_samples": 395042,
-    "train_samples_per_second": 9.187,
-    "train_steps_per_second": 0.287
-}

trainer_state.json

@@ -1,205 +0,0 @@
-{
-  "best_metric": null,
-  "best_model_checkpoint": null,
-  "epoch": 0.19999594985925762,
-  "global_step": 2469,
-  "is_hyper_param_search": false,
-  "is_local_process_zero": true,
-  "is_world_process_zero": true,
-  "log_history": [
-    {
-      "epoch": 0.01,
-      "learning_rate": 3.7125e-06,
-      "loss": 19.7984,
-      "step": 100
-    },
-    {
-      "epoch": 0.02,
-      "learning_rate": 7.425e-06,
-      "loss": 13.3485,
-      "step": 200
-    },
-    {
-      "epoch": 0.02,
-      "learning_rate": 1.1174999999999999e-05,
-      "loss": 10.0074,
-      "step": 300
-    },
-    {
-      "epoch": 0.03,
-      "learning_rate": 1.4925e-05,
-      "loss": 8.4765,
-      "step": 400
-    },
-    {
-      "epoch": 0.04,
-      "learning_rate": 1.8675e-05,
-      "loss": 6.9106,
-      "step": 500
-    },
-    {
-      "epoch": 0.04,
-      "eval_loss": 6.717121124267578,
-      "eval_runtime": 908.8081,
-      "eval_samples_per_second": 17.541,
-      "eval_steps_per_second": 2.193,
-      "eval_wer": 1.0,
-      "step": 500
-    },
-    {
-      "epoch": 0.05,
-      "learning_rate": 2.2424999999999996e-05,
-      "loss": 5.4175,
-      "step": 600
-    },
-    {
-      "epoch": 0.06,
-      "learning_rate": 2.6174999999999996e-05,
-      "loss": 4.1471,
-      "step": 700
-    },
-    {
-      "epoch": 0.06,
-      "learning_rate": 2.9925e-05,
-      "loss": 3.3536,
-      "step": 800
-    },
-    {
-      "epoch": 0.07,
-      "learning_rate": 3.3675e-05,
-      "loss": 3.0742,
-      "step": 900
-    },
-    {
-      "epoch": 0.08,
-      "learning_rate": 3.7424999999999995e-05,
-      "loss": 3.0034,
-      "step": 1000
-    },
-    {
-      "epoch": 0.08,
-      "eval_loss": 3.01257061958313,
-      "eval_runtime": 815.2192,
-      "eval_samples_per_second": 19.554,
-      "eval_steps_per_second": 2.445,
-      "eval_wer": 1.0,
-      "step": 1000
-    },
-    {
-      "epoch": 0.09,
-      "learning_rate": 4.1175e-05,
-      "loss": 2.9695,
-      "step": 1100
-    },
-    {
-      "epoch": 0.1,
-      "learning_rate": 4.4924999999999994e-05,
-      "loss": 2.9321,
-      "step": 1200
-    },
-    {
-      "epoch": 0.11,
-      "learning_rate": 4.8675e-05,
-      "loss": 2.916,
-      "step": 1300
-    },
-    {
-      "epoch": 0.11,
-      "learning_rate": 5.2424999999999994e-05,
-      "loss": 2.8952,
-      "step": 1400
-    },
-    {
-      "epoch": 0.12,
-      "learning_rate": 5.6175e-05,
-      "loss": 2.8699,
-      "step": 1500
-    },
-    {
-      "epoch": 0.12,
-      "eval_loss": 2.850850820541382,
-      "eval_runtime": 814.3164,
-      "eval_samples_per_second": 19.576,
-      "eval_steps_per_second": 2.447,
-      "eval_wer": 0.9817464201091995,
-      "step": 1500
-    },
-    {
-      "epoch": 0.13,
-      "learning_rate": 5.9925e-05,
-      "loss": 2.7752,
-      "step": 1600
-    },
-    {
-      "epoch": 0.14,
-      "learning_rate": 6.367499999999999e-05,
-      "loss": 2.4403,
-      "step": 1700
-    },
-    {
-      "epoch": 0.15,
-      "learning_rate": 6.7425e-05,
-      "loss": 1.9873,
-      "step": 1800
-    },
-    {
-      "epoch": 0.15,
-      "learning_rate": 7.1175e-05,
-      "loss": 1.7549,
-      "step": 1900
-    },
-    {
-      "epoch": 0.16,
-      "learning_rate": 7.48875e-05,
-      "loss": 1.629,
-      "step": 2000
-    },
-    {
-      "epoch": 0.16,
-      "eval_loss": 0.7786943316459656,
-      "eval_runtime": 820.0379,
-      "eval_samples_per_second": 19.439,
-      "eval_steps_per_second": 2.43,
-      "eval_wer": 0.586136293396518,
-      "step": 2000
-    },
-    {
-      "epoch": 0.17,
-      "learning_rate": 5.948827292110874e-05,
-      "loss": 1.5154,
-      "step": 2100
-    },
-    {
-      "epoch": 0.18,
-      "learning_rate": 4.349680170575692e-05,
-      "loss": 1.4719,
-      "step": 2200
-    },
-    {
-      "epoch": 0.19,
-      "learning_rate": 2.7505330490405118e-05,
-      "loss": 1.4058,
-      "step": 2300
-    },
-    {
-      "epoch": 0.19,
-      "learning_rate": 1.1513859275053305e-05,
-      "loss": 1.3579,
-      "step": 2400
-    },
-    {
-      "epoch": 0.2,
-      "step": 2469,
-      "total_flos": 1.1608261301380424e+19,
-      "train_loss": 4.42930416077822,
-      "train_runtime": 8599.8864,
-      "train_samples_per_second": 9.187,
-      "train_steps_per_second": 0.287
-    }
-  ],
-  "max_steps": 2469,
-  "num_train_epochs": 1,
-  "total_flos": 1.1608261301380424e+19,
-  "trial_name": null,
-  "trial_params": null
-}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:839f65dcbffe149e64e74ecde13830fc1e30c874f884e617aadad73ff74a039e
+oid sha256:6e91b39976977c0a605521323547644a3699b13dbfd044f77607812c3c5cc2b4
 size 2991

vocab.json

@@ -1 +1 @@
-{"'": 1, "a": 2, "b": 3, "c": 4, "d": 5, "e": 6, "f": 7, "g": 8, "h": 9, "i": 10, "j": 11, "k": 12, "l": 13, "m": 14, "n": 15, "o": 16, "p": 17, "q": 18, "r": 19, "s": 20, "t": 21, "u": 22, "v": 23, "w": 24, "x": 25, "y": 26, "z": 27, "à": 28, "á": 29, "â": 30, "ä": 31, "å": 32, "æ": 33, "ç": 34, "è": 35, "é": 36, "ê": 37, "ë": 38, "ì": 39, "í": 40, "î": 41, "ï": 42, "ñ": 43, "ò": 44, "ó": 45, "ô": 46, "õ": 47, "ö": 48, "ø": 49, "ù": 50, "ú": 51, "û": 52, "ü": 53, "þ": 54, "ÿ": 55, "ā": 56, "ć": 57, "ċ": 58, "č": 59, "đ": 60, "ē": 61, "ġ": 62, "ī": 63, "ı": 64, "ķ": 65, "ĺ": 66, "ļ": 67, "ľ": 68, "ł": 69, "ń": 70, "ņ": 71, "ň": 72, "ō": 73, "ŏ": 74, "ő": 75, "œ": 76, "ř": 77, "ś": 78, "ş": 79, "š": 80, "ţ": 81, "ť": 82, "ũ": 83, "ū": 84, "ů": 85, "ű": 86, "ų": 87, "ź": 88, "ż": 89, "ž": 90, "ơ": 91, "ư": 92, "ǀ": 93, "ǃ": 94, "ǎ": 95, "ǔ": 96, "ǫ": 97, "ǹ": 98, "ș": 99, "ț": 100, "ə": 101, "ɨ": 102, "ʉ": 103, "ʔ": 104, "ʻ": 105, "ʼ": 106, "ʽ": 107, "ʾ": 108, "ʿ": 109, "ː": 110, "ˢ": 111, "̀": 112, "́": 113, "̂": 114, "̃": 115, "̇": 116, "̈": 117, "̐": 118, "̠": 119, "̧": 120, "̱": 121, "̲": 122, "α": 123, "β": 124, "γ": 125, "δ": 126, "ε": 127, "ζ": 128, "η": 129, "θ": 130, "ι": 131, "κ": 132, "λ": 133, "μ": 134, "ν": 135, "ο": 136, "π": 137, "ρ": 138, "ς": 139, "σ": 140, "τ": 141, "υ": 142, "φ": 143, "χ": 144, "ψ": 145, "ω": 146, "ό": 147, "а": 148, "г": 149, "е": 150, "з": 151, "и": 152, "к": 153, "м": 154, "н": 155, "п": 156, "р": 157, "ц": 158, "ч": 159, "э": 160, "я": 161, "є": 162, "і": 163, "ј": 164, "ҫ": 165, "ӌ": 166, "գ": 167, "զ": 168, "ا": 169, "ب": 170, "ة": 171, "د": 172, "ر": 173, "ل": 174, "م": 175, "ن": 176, "و": 177, "ي": 178, "ቀ": 179, "ከ": 180, "ወ": 181, "ደ": 182, "ጀ": 183, "ጠ": 184, "ḍ": 185, "ḥ": 186, "ṅ": 187, "ṇ": 188, "ṣ": 189, "ṭ": 190, "ṯ": 191, "ẓ": 192, "ạ": 193, "ả": 194, "ầ": 195, "ậ": 196, "ắ": 197, "ẵ": 198, "ế": 199, "ề": 200, "ễ": 201, "ệ": 202, "ị": 203, "ố": 204, "ồ": 205, "ổ": 206, "ộ": 207, "ớ": 208, "ờ": 209, "ợ": 210, "ụ": 211, "ủ": 212, "ứ": 213, "ử": 214, "ỳ": 215, "‐": 216, "―": 217, "„": 218, "†": 219, "′": 220, "‹": 221, "›": 222, "€": 223, "₽": 224, "ℂ": 225, "ℕ": 226, "ℚ": 227, "ℝ": 228, "ℤ": 229, "ℰ": 230, "ℵ": 231, "→": 232, "↔": 233, "∅": 234, "∆": 235, "∈": 236, "−": 237, "∞": 238, "∨": 239, "∼": 240, "≥": 241, "⋅": 242, "─": 243, "☉": 244, "ⱅ": 245, "ⱎ": 246, "い": 247, "う": 248, "た": 249, "つ": 250, "ぬ": 251, "の": 252, "ひ": 253, "へ": 254, "ま": 255, "む": 256, "め": 257, "も": 258, "や": 259, "ゔ": 260, "丹": 261, "乃": 262, "京": 263, "北": 264, "扬": 265, "文": 266, "星": 267, "术": 268, "杜": 269, "牡": 270, "甌": 271, "美": 272, "西": 273, "貴": 274, "青": 275, "馆": 276, "ꝑ": 277, "고": 278, "기": 279, "먹": 280, "삼": 281, "|": 0, "[UNK]": 282, "[PAD]": 283}
+{"'": 1, "a": 2, "b": 3, "c": 4, "d": 5, "e": 6, "f": 7, "g": 8, "h": 9, "i": 10, "j": 11, "k": 12, "l": 13, "m": 14, "n": 15, "o": 16, "p": 17, "q": 18, "r": 19, "s": 20, "t": 21, "u": 22, "v": 23, "w": 24, "x": 25, "y": 26, "z": 27, "à": 28, "â": 29, "ç": 30, "è": 31, "é": 32, "ê": 33, "î": 34, "ï": 35, "ô": 36, "ù": 37, "û": 38, "|": 0, "[UNK]": 39, "[PAD]": 40}

wandb/debug-internal.log

@@ -0,0 +1 @@
+run-20220130_224738-2uzt3kt1/logs/debug-internal.log

wandb/debug.log

@@ -0,0 +1 @@
+run-20220130_224738-2uzt3kt1/logs/debug.log

wandb/latest-run

@@ -0,0 +1 @@
+run-20220130_224738-2uzt3kt1

wandb/run-20220130_224738-2uzt3kt1/files/config.yaml

@@ -0,0 +1,672 @@
+wandb_version: 1
+
+_n_gpu:
+  desc: null
+  value: 1
+_name_or_path:
+  desc: null
+  value: facebook/wav2vec2-xls-r-300m
+_wandb:
+  desc: null
+  value:
+    cli_version: 0.12.9
+    framework: huggingface
+    huggingface_version: 4.17.0.dev0
+    is_jupyter_run: false
+    is_kaggle_kernel: false
+    m:
+    - 1: train/global_step
+      6:
+      - 3
+    - 1: train/train_runtime
+      5: 1
+      6:
+      - 1
+    - 1: train/train_samples_per_second
+      5: 1
+      6:
+      - 1
+    - 1: train/train_steps_per_second
+      5: 1
+      6:
+      - 1
+    - 1: train/total_flos
+      5: 1
+      6:
+      - 1
+    - 1: train/train_loss
+      5: 1
+      6:
+      - 1
+    - 1: train/epoch
+      5: 1
+      6:
+      - 1
+    python_version: 3.8.8
+    start_time: 1643582858
+    t:
+      1:
+      - 1
+      - 5
+      - 11
+      3:
+      - 13
+      4: 3.8.8
+      5: 0.12.9
+      6: 4.17.0.dev0
+      8:
+      - 5
+activation_dropout:
+  desc: null
+  value: 0.1
+adafactor:
+  desc: null
+  value: false
+adam_beta1:
+  desc: null
+  value: 0.9
+adam_beta2:
+  desc: null
+  value: 0.999
+adam_epsilon:
+  desc: null
+  value: 1.0e-08
+adapter_kernel_size:
+  desc: null
+  value: 3
+adapter_stride:
+  desc: null
+  value: 2
+add_adapter:
+  desc: null
+  value: false
+add_cross_attention:
+  desc: null
+  value: false
+apply_spec_augment:
+  desc: null
+  value: true
+architectures:
+  desc: null
+  value:
+  - Wav2Vec2ForPreTraining
+attention_dropout:
+  desc: null
+  value: 0.0
+bad_words_ids:
+  desc: null
+  value: null
+bf16:
+  desc: null
+  value: false
+bf16_full_eval:
+  desc: null
+  value: false
+bos_token_id:
+  desc: null
+  value: 1
+chunk_size_feed_forward:
+  desc: null
+  value: 0
+classifier_proj_size:
+  desc: null
+  value: 256
+codevector_dim:
+  desc: null
+  value: 768
+contrastive_logits_temperature:
+  desc: null
+  value: 0.1
+conv_bias:
+  desc: null
+  value: true
+conv_dim:
+  desc: null
+  value:
+  - 512
+  - 512
+  - 512
+  - 512
+  - 512
+  - 512
+  - 512
+conv_kernel:
+  desc: null
+  value:
+  - 10
+  - 3
+  - 3
+  - 3
+  - 3
+  - 2
+  - 2
+conv_stride:
+  desc: null
+  value:
+  - 5
+  - 2
+  - 2
+  - 2
+  - 2
+  - 2
+  - 2
+cross_attention_hidden_size:
+  desc: null
+  value: null
+ctc_loss_reduction:
+  desc: null
+  value: mean
+ctc_zero_infinity:
+  desc: null
+  value: false
+dataloader_drop_last:
+  desc: null
+  value: false
+dataloader_num_workers:
+  desc: null
+  value: 0
+dataloader_pin_memory:
+  desc: null
+  value: true
+ddp_bucket_cap_mb:
+  desc: null
+  value: None
+ddp_find_unused_parameters:
+  desc: null
+  value: None
+debug:
+  desc: null
+  value: '[]'
+decoder_start_token_id:
+  desc: null
+  value: null
+deepspeed:
+  desc: null
+  value: None
+disable_tqdm:
+  desc: null
+  value: false
+diversity_loss_weight:
+  desc: null
+  value: 0.1
+diversity_penalty:
+  desc: null
+  value: 0.0
+do_eval:
+  desc: null
+  value: true
+do_predict:
+  desc: null
+  value: false
+do_sample:
+  desc: null
+  value: false
+do_stable_layer_norm:
+  desc: null
+  value: true
+do_train:
+  desc: null
+  value: true
+early_stopping:
+  desc: null
+  value: false
+encoder_no_repeat_ngram_size:
+  desc: null
+  value: 0
+eos_token_id:
+  desc: null
+  value: 2
+eval_accumulation_steps:
+  desc: null
+  value: None
+eval_batch_size:
+  desc: null
+  value: 8
+eval_steps:
+  desc: null
+  value: 500
+evaluation_strategy:
+  desc: null
+  value: steps
+feat_extract_activation:
+  desc: null
+  value: gelu
+feat_extract_dropout:
+  desc: null
+  value: 0.0
+feat_extract_norm:
+  desc: null
+  value: layer
+feat_proj_dropout:
+  desc: null
+  value: 0.0
+feat_quantizer_dropout:
+  desc: null
+  value: 0.0
+final_dropout:
+  desc: null
+  value: 0.0
+finetuning_task:
+  desc: null
+  value: null
+forced_bos_token_id:
+  desc: null
+  value: null
+forced_eos_token_id:
+  desc: null
+  value: null
+fp16:
+  desc: null
+  value: true
+fp16_backend:
+  desc: null
+  value: auto
+fp16_full_eval:
+  desc: null
+  value: false
+fp16_opt_level:
+  desc: null
+  value: O1
+gradient_accumulation_steps:
+  desc: null
+  value: 8
+gradient_checkpointing:
+  desc: null
+  value: true
+greater_is_better:
+  desc: null
+  value: false
+group_by_length:
+  desc: null
+  value: true
+half_precision_backend:
+  desc: null
+  value: amp
+hidden_act:
+  desc: null
+  value: gelu
+hidden_dropout:
+  desc: null
+  value: 0.0
+hidden_size:
+  desc: null
+  value: 1024
+hub_model_id:
+  desc: null
+  value: None
+hub_strategy:
+  desc: null
+  value: every_save
+hub_token:
+  desc: null
+  value: <HUB_TOKEN>
+id2label:
+  desc: null
+  value:
+    '0': LABEL_0
+    '1': LABEL_1
+ignore_data_skip:
+  desc: null
+  value: false
+initializer_range:
+  desc: null
+  value: 0.02
+intermediate_size:
+  desc: null
+  value: 4096
+is_decoder:
+  desc: null
+  value: false
+is_encoder_decoder:
+  desc: null
+  value: false
+label2id:
+  desc: null
+  value:
+    LABEL_0: 0
+    LABEL_1: 1
+label_names:
+  desc: null
+  value: None
+label_smoothing_factor:
+  desc: null
+  value: 0.0
+layer_norm_eps:
+  desc: null
+  value: 1.0e-05
+layerdrop:
+  desc: null
+  value: 0.0
+learning_rate:
+  desc: null
+  value: 7.5e-05
+length_column_name:
+  desc: null
+  value: input_length
+length_penalty:
+  desc: null
+  value: 1.0
+load_best_model_at_end:
+  desc: null
+  value: true
+local_rank:
+  desc: null
+  value: -1
+log_level:
+  desc: null
+  value: -1
+log_level_replica:
+  desc: null
+  value: -1
+log_on_each_node:
+  desc: null
+  value: true
+logging_dir:
+  desc: null
+  value: ./runs/Jan30_22-46-41_job-3261699b-76eb-4c28-8419-66a66c5c9199
+logging_first_step:
+  desc: null
+  value: false
+logging_nan_inf_filter:
+  desc: null
+  value: true
+logging_steps:
+  desc: null
+  value: 100
+logging_strategy:
+  desc: null
+  value: steps
+lr_scheduler_type:
+  desc: null
+  value: linear
+mask_feature_length:
+  desc: null
+  value: 64
+mask_feature_min_masks:
+  desc: null
+  value: 0
+mask_feature_prob:
+  desc: null
+  value: 0.25
+mask_time_length:
+  desc: null
+  value: 10
+mask_time_min_masks:
+  desc: null
+  value: 2
+mask_time_prob:
+  desc: null
+  value: 0.75
+max_grad_norm:
+  desc: null
+  value: 1.0
+max_length:
+  desc: null
+  value: 20
+max_steps:
+  desc: null
+  value: -1
+metric_for_best_model:
+  desc: null
+  value: loss
+min_length:
+  desc: null
+  value: 0
+model_type:
+  desc: null
+  value: wav2vec2
+mp_parameters:
+  desc: null
+  value: ''
+no_cuda:
+  desc: null
+  value: false
+no_repeat_ngram_size:
+  desc: null
+  value: 0
+num_adapter_layers:
+  desc: null
+  value: 3
+num_attention_heads:
+  desc: null
+  value: 16
+num_beam_groups:
+  desc: null
+  value: 1
+num_beams:
+  desc: null
+  value: 1
+num_codevector_groups:
+  desc: null
+  value: 2
+num_codevectors_per_group:
+  desc: null
+  value: 320
+num_conv_pos_embedding_groups:
+  desc: null
+  value: 16
+num_conv_pos_embeddings:
+  desc: null
+  value: 128
+num_feat_extract_layers:
+  desc: null
+  value: 7
+num_hidden_layers:
+  desc: null
+  value: 24
+num_negatives:
+  desc: null
+  value: 100
+num_return_sequences:
+  desc: null
+  value: 1
+num_train_epochs:
+  desc: null
+  value: 0.2
+optim:
+  desc: null
+  value: adamw_hf
+output_attentions:
+  desc: null
+  value: false
+output_dir:
+  desc: null
+  value: ./
+output_hidden_size:
+  desc: null
+  value: 1024
+output_hidden_states:
+  desc: null
+  value: false
+output_scores:
+  desc: null
+  value: false
+overwrite_output_dir:
+  desc: null
+  value: true
+pad_token_id:
+  desc: null
+  value: 40
+past_index:
+  desc: null
+  value: -1
+per_device_eval_batch_size:
+  desc: null
+  value: 8
+per_device_train_batch_size:
+  desc: null
+  value: 8
+per_gpu_eval_batch_size:
+  desc: null
+  value: None
+per_gpu_train_batch_size:
+  desc: null
+  value: None
+prediction_loss_only:
+  desc: null
+  value: false
+prefix:
+  desc: null
+  value: null
+problem_type:
+  desc: null
+  value: null
+proj_codevector_dim:
+  desc: null
+  value: 768
+pruned_heads:
+  desc: null
+  value: {}
+push_to_hub:
+  desc: null
+  value: true
+push_to_hub_model_id:
+  desc: null
+  value: None
+push_to_hub_organization:
+  desc: null
+  value: None
+push_to_hub_token:
+  desc: null
+  value: <PUSH_TO_HUB_TOKEN>
+remove_invalid_values:
+  desc: null
+  value: false
+remove_unused_columns:
+  desc: null
+  value: true
+repetition_penalty:
+  desc: null
+  value: 1.0
+report_to:
+  desc: null
+  value: '[''wandb'']'
+resume_from_checkpoint:
+  desc: null
+  value: None
+return_dict:
+  desc: null
+  value: true
+return_dict_in_generate:
+  desc: null
+  value: false
+run_name:
+  desc: null
+  value: ./
+save_on_each_node:
+  desc: null
+  value: false
+save_steps:
+  desc: null
+  value: 500
+save_strategy:
+  desc: null
+  value: steps
+save_total_limit:
+  desc: null
+  value: 3
+seed:
+  desc: null
+  value: 42
+sep_token_id:
+  desc: null
+  value: null
+sharded_ddp:
+  desc: null
+  value: '[]'
+skip_memory_metrics:
+  desc: null
+  value: true
+task_specific_params:
+  desc: null
+  value: null
+tdnn_dilation:
+  desc: null
+  value:
+  - 1
+  - 2
+  - 3
+  - 1
+  - 1
+tdnn_dim:
+  desc: null
+  value:
+  - 512
+  - 512
+  - 512
+  - 512
+  - 1500
+tdnn_kernel:
+  desc: null
+  value:
+  - 5
+  - 3
+  - 3
+  - 1
+  - 1
+temperature:
+  desc: null
+  value: 1.0
+tf32:
+  desc: null
+  value: None
+tie_encoder_decoder:
+  desc: null
+  value: false
+tie_word_embeddings:
+  desc: null
+  value: true
+tokenizer_class:
+  desc: null
+  value: null
+top_k:
+  desc: null
+  value: 50
+top_p:
+  desc: null
+  value: 1.0
+torch_dtype:
+  desc: null
+  value: float32
+torchscript:
+  desc: null
+  value: false
+tpu_metrics_debug:
+  desc: null
+  value: false
+tpu_num_cores:
+  desc: null
+  value: None
+train_batch_size:
+  desc: null
+  value: 8
+transformers_version:
+  desc: null
+  value: 4.17.0.dev0
+use_bfloat16:
+  desc: null
+  value: false
+use_legacy_prediction_loop:
+  desc: null
+  value: false
+use_weighted_layer_sum:
+  desc: null
+  value: false
+vocab_size:
+  desc: null
+  value: 41
+warmup_ratio:
+  desc: null
+  value: 0.0
+warmup_steps:
+  desc: null
+  value: 2000
+weight_decay:
+  desc: null
+  value: 0.0
+xpu_backend:
+  desc: null
+  value: None
+xvector_output_dim:
+  desc: null
+  value: 512

wandb/run-20220130_224738-2uzt3kt1/files/output.log

@@ -0,0 +1,14 @@
+
+
+
+ 67%|██████████████████████████████████████████████████████████████████████████████████████████                                             | 2/3 [00:07<00:03,  3.88s/it]
+100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [00:10<00:00,  3.23s/it]
+Training completed. Do not forget to share your model on huggingface.co/models =)
+100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [00:10<00:00,  3.48s/it]
+Saving model checkpoint to ./
+Configuration saved in ./config.json
+Model weights saved in ./pytorch_model.bin
+Configuration saved in ./preprocessor_config.json
+Saving model checkpoint to ./
+Configuration saved in ./config.json
+Model weights saved in ./pytorch_model.bin

wandb/run-20220130_224738-2uzt3kt1/files/requirements.txt

@@ -0,0 +1,180 @@
+aiohttp==3.8.1
+aiosignal==1.2.0
+analytics-python==1.4.0
+anyio==3.5.0
+appdirs==1.4.4
+argon2-cffi-bindings==21.2.0
+argon2-cffi==21.3.0
+asgiref==3.5.0
+asttokens==2.0.5
+async-timeout==4.0.2
+attrs==21.4.0
+audioread==2.1.9
+backcall==0.2.0
+backoff==1.10.0
+bcrypt==3.2.0
+beautifulsoup4==4.9.3
+black==21.12b0
+bleach==4.1.0
+brotlipy==0.7.0
+certifi==2020.12.5
+cffi==1.14.3
+chardet==3.0.4
+charset-normalizer==2.0.10
+click==8.0.3
+conda-build==3.21.4
+conda-package-handling==1.7.2
+conda==4.9.2
+configparser==5.2.0
+cryptography==3.2.1
+cycler==0.11.0
+datasets==1.18.2.dev0
+debugpy==1.5.1
+decorator==4.4.2
+defusedxml==0.7.1
+dill==0.3.4
+dnspython==2.1.0
+docker-pycreds==0.4.0
+entrypoints==0.3
+executing==0.8.2
+fastapi==0.73.0
+ffmpy==0.3.0
+filelock==3.0.12
+fonttools==4.29.0
+frozenlist==1.3.0
+fsspec==2022.1.0
+gitdb==4.0.9
+gitpython==3.1.26
+glob2==0.7
+gradio==2.7.5.2
+h11==0.13.0
+huggingface-hub==0.4.0
+idna==2.10
+importlib-resources==5.4.0
+ipykernel==6.7.0
+ipython-genutils==0.2.0
+ipython==8.0.1
+ipywidgets==7.6.3
+jedi==0.17.0
+jinja2==2.11.3
+jiwer==2.3.0
+joblib==1.1.0
+json5==0.9.6
+jsonschema==4.4.0
+jupyter-client==7.1.2
+jupyter-core==4.9.1
+jupyterlab-pygments==0.1.2
+jupyterlab-server==1.2.0
+jupyterlab-widgets==1.0.2
+jupyterlab==2.2.9
+kiwisolver==1.3.2
+libarchive-c==2.9
+librosa==0.8.1
+llvmlite==0.38.0
+markdown2==2.4.2
+markupsafe==1.1.1
+matplotlib-inline==0.1.3
+matplotlib==3.5.1
+mistune==0.8.4
+mkl-fft==1.3.0
+mkl-random==1.1.1
+mkl-service==2.3.0
+monotonic==1.6
+multidict==6.0.2
+multiprocess==0.70.12.2
+mypy-extensions==0.4.3
+nano==0.10.0
+nbclient==0.5.10
+nbconvert==6.4.1
+nbformat==5.1.3
+nest-asyncio==1.5.4
+notebook==6.4.8
+numba==0.55.1
+numpy==1.19.2
+olefile==0.46
+packaging==21.3
+pandas==1.4.0
+pandocfilters==1.5.0
+paramiko==2.9.2
+parso==0.8.1
+pathspec==0.9.0
+pathtools==0.1.2
+pexpect==4.8.0
+pickleshare==0.7.5
+pillow==8.1.2
+pip==21.3.1
+pkginfo==1.7.0
+platformdirs==2.4.1
+pooch==1.6.0
+prometheus-client==0.13.0
+promise==2.3
+prompt-toolkit==3.0.8
+protobuf==3.19.4
+psutil==5.8.0
+ptyprocess==0.7.0
+pure-eval==0.2.2
+pyarrow==6.0.1
+pycosat==0.6.3
+pycparser==2.20
+pycryptodome==3.13.0
+pydantic==1.9.0
+pydub==0.25.1
+pygments==2.8.0
+pynacl==1.5.0
+pyopenssl==19.1.0
+pyparsing==3.0.7
+pyrsistent==0.18.1
+pysocks==1.7.1
+python-dateutil==2.8.2
+python-etcd==0.4.5
+python-levenshtein==0.12.2
+python-multipart==0.0.5
+pytz==2021.1
+pyyaml==5.4.1
+pyzmq==22.3.0
+regex==2022.1.18
+requests==2.24.0
+resampy==0.2.2
+ruamel-yaml==0.15.87
+sacremoses==0.0.47
+scikit-learn==1.0.2
+scipy==1.7.3
+send2trash==1.8.0
+sentry-sdk==1.5.4
+setuptools==50.3.1.post20201107
+shortuuid==1.0.8
+six==1.15.0
+smmap==5.0.0
+sniffio==1.2.0
+soundfile==0.10.3.post1
+soupsieve==2.2
+stack-data==0.1.4
+starlette==0.17.1
+subprocess32==3.5.4
+termcolor==1.1.0
+terminado==0.13.1
+testpath==0.5.0
+threadpoolctl==3.0.0
+tokenizers==0.11.4
+tomli==1.2.3
+torch==1.10.2
+torchaudio==0.10.2
+torchelastic==0.2.2
+torchtext==0.9.1
+torchvision==0.9.1
+tornado==6.1
+tqdm==4.62.3
+traitlets==5.1.1
+transformers==4.17.0.dev0
+typing-extensions==4.0.1
+urllib3==1.25.11
+uvicorn==0.17.1
+wandb==0.12.9
+wcwidth==0.2.5
+webencodings==0.5.1
+wheel==0.35.1
+widgetsnbextension==3.5.2
+xxhash==2.0.2
+yarl==1.7.2
+yaspin==2.1.0
+zipp==3.7.0

wandb/run-20220130_224738-2uzt3kt1/files/wandb-metadata.json

@@ -0,0 +1,63 @@
+{
+    "os": "Linux-4.15.0-151-generic-x86_64-with-glibc2.10",
+    "python": "3.8.8",
+    "heartbeatAt": "2022-01-30T22:47:39.607019",
+    "startedAt": "2022-01-30T22:47:38.310593",
+    "docker": null,
+    "gpu": "Tesla V100S-PCIE-32GB",
+    "gpu_count": 1,
+    "cpu_count": 60,
+    "cuda": null,
+    "args": [
+        "--activation_dropout=0.1",
+        "--dataset_name=mozilla-foundation/common_voice_7_0",
+        "--dataset_config_name=fr",
+        "--eval_steps=500",
+        "--evaluation_strategy=steps",
+        "--feat_proj_dropout=0.0",
+        "--freeze_feature_encoder",
+        "--fp16",
+        "--gradient_accumulation_steps=8",
+        "--gradient_checkpointing",
+        "--group_by_length",
+        "--layerdrop=0.0",
+        "--learning_rate=7.5e-5",
+        "--length_column_name=input_length",
+        "--load_best_model_at_end",
+        "--logging_steps=100",
+        "--mask_feature_length=64",
+        "--mask_feature_prob=0.25",
+        "--mask_time_length=10",
+        "--mask_time_prob=0.75",
+        "--max_train_samples=1000",
+        "--max_eval_samples=200",
+        "--model_name_or_path=facebook/wav2vec2-xls-r-300m",
+        "--num_train_epochs=0.2",
+        "--output_dir=./",
+        "--overwrite_output_dir",
+        "--per_device_train_batch_size=8",
+        "--per_device_eval_batch_size=8",
+        "--preprocessing_num_workers=4",
+        "--push_to_hub",
+        "--report_to=wandb",
+        "--save_steps=500",
+        "--save_total_limit=3",
+        "--text_column_name=sentence",
+        "--use_auth_token",
+        "--warmup_steps=2000",
+        "--do_train",
+        "--do_eval"
+    ],
+    "state": "running",
+    "program": "run_speech_recognition_ctc.py",
+    "codePath": "run_speech_recognition_ctc.py",
+    "git": {
+        "remote": "https://huggingface.co/Plim/xls-r-300m-fr",
+        "commit": "1d172876193bf100999c8d09d283f8d0894252f2"
+    },
+    "email": "[email protected]",
+    "root": "/workspace/xls-r-300m-fr",
+    "host": "job-3261699b-76eb-4c28-8419-66a66c5c9199",
+    "username": "ovh",
+    "executable": "/opt/conda/bin/python"
+}

wandb/run-20220130_224738-2uzt3kt1/files/wandb-summary.json

@@ -0,0 +1 @@
+{"train/train_runtime": 12.893, "train/train_samples_per_second": 15.512, "train/train_steps_per_second": 0.233, "train/total_flos": 2.67196543170048e+16, "train/train_loss": 12.496875762939453, "train/epoch": 0.19, "train/global_step": 3, "_runtime": 12, "_timestamp": 1643582870, "_step": 0}

wandb/run-20220130_224738-2uzt3kt1/logs/debug-internal.log

@@ -0,0 +1,57 @@
+2022-01-30 22:47:39,297 INFO    MainThread:23196 [internal.py:wandb_internal():87] W&B internal server running at pid: 23196, started at: 2022-01-30 22:47:39.296970
+2022-01-30 22:47:39,300 INFO    WriterThread:23196 [datastore.py:open_for_write():77] open: /workspace/xls-r-300m-fr/wandb/run-20220130_224738-2uzt3kt1/run-2uzt3kt1.wandb
+2022-01-30 22:47:39,301 DEBUG   HandlerThread:23196 [handler.py:handle_request():130] handle_request: check_version
+2022-01-30 22:47:39,304 DEBUG   SenderThread:23196 [sender.py:send():234] send: header
+2022-01-30 22:47:39,304 DEBUG   SenderThread:23196 [sender.py:send_request():248] send_request: check_version
+2022-01-30 22:47:39,377 DEBUG   SenderThread:23196 [sender.py:send():234] send: run
+2022-01-30 22:47:39,597 DEBUG   HandlerThread:23196 [handler.py:handle_request():130] handle_request: run_start
+2022-01-30 22:47:39,599 INFO    SenderThread:23196 [dir_watcher.py:__init__():169] watching files in: /workspace/xls-r-300m-fr/wandb/run-20220130_224738-2uzt3kt1/files
+2022-01-30 22:47:39,600 INFO    SenderThread:23196 [sender.py:_start_run_threads():804] run started: 2uzt3kt1 with start time 1643582858
+2022-01-30 22:47:39,600 DEBUG   SenderThread:23196 [sender.py:send():234] send: summary
+2022-01-30 22:47:39,600 INFO    SenderThread:23196 [sender.py:_save_file():939] saving file wandb-summary.json with policy end
+2022-01-30 22:47:39,606 DEBUG   HandlerThread:23196 [meta.py:__init__():40] meta init
+2022-01-30 22:47:39,606 DEBUG   HandlerThread:23196 [meta.py:__init__():54] meta init done
+2022-01-30 22:47:39,606 DEBUG   HandlerThread:23196 [meta.py:probe():214] probe
+2022-01-30 22:47:39,615 DEBUG   HandlerThread:23196 [meta.py:_setup_git():204] setup git
+2022-01-30 22:47:39,653 DEBUG   HandlerThread:23196 [meta.py:_setup_git():211] setup git done
+2022-01-30 22:47:39,653 DEBUG   HandlerThread:23196 [meta.py:_save_pip():58] save pip
+2022-01-30 22:47:39,654 DEBUG   HandlerThread:23196 [meta.py:_save_pip():72] save pip done
+2022-01-30 22:47:39,655 DEBUG   HandlerThread:23196 [meta.py:_save_conda():79] save conda
+2022-01-30 22:47:40,176 DEBUG   HandlerThread:23196 [meta.py:_save_conda():89] save conda done
+2022-01-30 22:47:40,176 DEBUG   HandlerThread:23196 [meta.py:probe():252] probe done
+2022-01-30 22:47:40,185 DEBUG   SenderThread:23196 [sender.py:send():234] send: files
+2022-01-30 22:47:40,186 INFO    SenderThread:23196 [sender.py:_save_file():939] saving file wandb-metadata.json with policy now
+2022-01-30 22:47:40,197 DEBUG   HandlerThread:23196 [handler.py:handle_request():130] handle_request: stop_status
+2022-01-30 22:47:40,198 DEBUG   SenderThread:23196 [sender.py:send_request():248] send_request: stop_status
+2022-01-30 22:47:40,354 DEBUG   SenderThread:23196 [sender.py:send():234] send: config
+2022-01-30 22:47:40,357 DEBUG   SenderThread:23196 [sender.py:send():234] send: metric
+2022-01-30 22:47:40,357 DEBUG   SenderThread:23196 [sender.py:send():234] send: metric
+2022-01-30 22:47:40,357 WARNING SenderThread:23196 [sender.py:send_metric():897] Seen metric with glob (shouldnt happen)
+2022-01-30 22:47:40,601 INFO    Thread-8  :23196 [dir_watcher.py:_on_file_created():217] file/dir created: /workspace/xls-r-300m-fr/wandb/run-20220130_224738-2uzt3kt1/files/requirements.txt
+2022-01-30 22:47:40,601 INFO    Thread-8  :23196 [dir_watcher.py:_on_file_created():217] file/dir created: /workspace/xls-r-300m-fr/wandb/run-20220130_224738-2uzt3kt1/files/wandb-summary.json
+2022-01-30 22:47:40,601 INFO    Thread-8  :23196 [dir_watcher.py:_on_file_created():217] file/dir created: /workspace/xls-r-300m-fr/wandb/run-20220130_224738-2uzt3kt1/files/output.log
+2022-01-30 22:47:40,601 INFO    Thread-8  :23196 [dir_watcher.py:_on_file_created():217] file/dir created: /workspace/xls-r-300m-fr/wandb/run-20220130_224738-2uzt3kt1/files/wandb-metadata.json
+2022-01-30 22:47:40,601 INFO    Thread-8  :23196 [dir_watcher.py:_on_file_created():217] file/dir created: /workspace/xls-r-300m-fr/wandb/run-20220130_224738-2uzt3kt1/files/conda-environment.yaml
+2022-01-30 22:47:40,709 INFO    Thread-11 :23196 [upload_job.py:push():137] Uploaded file /tmp/tmp51rrl_mrwandb/2urk4c1m-wandb-metadata.json
+2022-01-30 22:47:42,601 INFO    Thread-8  :23196 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220130_224738-2uzt3kt1/files/output.log
+2022-01-30 22:47:46,604 INFO    Thread-8  :23196 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220130_224738-2uzt3kt1/files/output.log
+2022-01-30 22:47:50,606 INFO    Thread-8  :23196 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220130_224738-2uzt3kt1/files/output.log
+2022-01-30 22:47:50,676 DEBUG   SenderThread:23196 [sender.py:send():234] send: metric
+2022-01-30 22:47:50,677 DEBUG   SenderThread:23196 [sender.py:send():234] send: metric
+2022-01-30 22:47:50,677 DEBUG   SenderThread:23196 [sender.py:send():234] send: metric
+2022-01-30 22:47:50,677 DEBUG   SenderThread:23196 [sender.py:send():234] send: metric
+2022-01-30 22:47:50,677 DEBUG   SenderThread:23196 [sender.py:send():234] send: metric
+2022-01-30 22:47:50,677 DEBUG   SenderThread:23196 [sender.py:send():234] send: metric
+2022-01-30 22:47:50,678 DEBUG   SenderThread:23196 [sender.py:send():234] send: history
+2022-01-30 22:47:50,678 DEBUG   SenderThread:23196 [sender.py:send():234] send: summary
+2022-01-30 22:47:50,679 INFO    SenderThread:23196 [sender.py:_save_file():939] saving file wandb-summary.json with policy end
+2022-01-30 22:47:51,607 INFO    Thread-8  :23196 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220130_224738-2uzt3kt1/files/wandb-summary.json
+2022-01-30 22:47:52,608 INFO    Thread-8  :23196 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220130_224738-2uzt3kt1/files/output.log
+2022-01-30 22:47:54,610 INFO    Thread-8  :23196 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220130_224738-2uzt3kt1/files/output.log
+2022-01-30 22:47:55,488 DEBUG   HandlerThread:23196 [handler.py:handle_request():130] handle_request: stop_status
+2022-01-30 22:47:55,488 DEBUG   SenderThread:23196 [sender.py:send_request():248] send_request: stop_status
+2022-01-30 22:47:56,611 INFO    Thread-8  :23196 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220130_224738-2uzt3kt1/files/output.log
+2022-01-30 22:48:07,838 DEBUG   SenderThread:23196 [sender.py:send():234] send: stats
+2022-01-30 22:48:10,621 INFO    Thread-8  :23196 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220130_224738-2uzt3kt1/files/config.yaml
+2022-01-30 22:48:10,741 DEBUG   HandlerThread:23196 [handler.py:handle_request():130] handle_request: stop_status
+2022-01-30 22:48:10,742 DEBUG   SenderThread:23196 [sender.py:send_request():248] send_request: stop_status

wandb/run-20220130_224738-2uzt3kt1/logs/debug.log

@@ -0,0 +1,24 @@
+2022-01-30 22:47:38,315 INFO    MainThread:22602 [wandb_setup.py:_flush():71] setting env: {}
+2022-01-30 22:47:38,315 INFO    MainThread:22602 [wandb_setup.py:_flush():71] setting login settings: {}
+2022-01-30 22:47:38,315 INFO    MainThread:22602 [wandb_init.py:_log_setup():371] Logging user logs to /workspace/xls-r-300m-fr/wandb/run-20220130_224738-2uzt3kt1/logs/debug.log
+2022-01-30 22:47:38,315 INFO    MainThread:22602 [wandb_init.py:_log_setup():372] Logging internal logs to /workspace/xls-r-300m-fr/wandb/run-20220130_224738-2uzt3kt1/logs/debug-internal.log
+2022-01-30 22:47:38,316 INFO    MainThread:22602 [wandb_init.py:init():404] calling init triggers
+2022-01-30 22:47:38,316 INFO    MainThread:22602 [wandb_init.py:init():409] wandb.init called with sweep_config: {}
+config: {}
+2022-01-30 22:47:38,316 INFO    MainThread:22602 [wandb_init.py:init():460] starting backend
+2022-01-30 22:47:38,316 INFO    MainThread:22602 [backend.py:_multiprocessing_setup():99] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
+2022-01-30 22:47:38,420 INFO    MainThread:22602 [backend.py:ensure_launched():216] starting backend process...
+2022-01-30 22:47:38,497 INFO    MainThread:22602 [backend.py:ensure_launched():221] started backend process with pid: 23196
+2022-01-30 22:47:38,499 INFO    MainThread:22602 [wandb_init.py:init():469] backend started and connected
+2022-01-30 22:47:38,508 INFO    MainThread:22602 [wandb_init.py:init():533] updated telemetry
+2022-01-30 22:47:38,663 INFO    MainThread:22602 [wandb_init.py:init():563] communicating current version
+2022-01-30 22:47:39,375 INFO    MainThread:22602 [wandb_init.py:init():568] got version response 
+2022-01-30 22:47:39,375 INFO    MainThread:22602 [wandb_init.py:init():578] communicating run to backend with 30 second timeout
+2022-01-30 22:47:39,596 INFO    MainThread:22602 [wandb_init.py:init():606] starting run threads in backend
+2022-01-30 22:47:40,195 INFO    MainThread:22602 [wandb_run.py:_console_start():1810] atexit reg
+2022-01-30 22:47:40,196 INFO    MainThread:22602 [wandb_run.py:_redirect():1684] redirect: SettingsConsole.REDIRECT
+2022-01-30 22:47:40,197 INFO    MainThread:22602 [wandb_run.py:_redirect():1689] Redirecting console.
+2022-01-30 22:47:40,203 INFO    MainThread:22602 [wandb_run.py:_redirect():1745] Redirects installed.
+2022-01-30 22:47:40,203 INFO    MainThread:22602 [wandb_init.py:init():633] run started, returning control to user process
+2022-01-30 22:47:40,206 INFO    MainThread:22602 [wandb_run.py:_config_callback():956] config_cb None None {'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'float32', 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 20, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'chunk_size_feed_forward': 0, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'architectures': ['Wav2Vec2ForPreTraining'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 1, 'pad_token_id': 40, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': None, 'task_specific_params': None, 'problem_type': None, '_name_or_path': 'facebook/wav2vec2-xls-r-300m', 'transformers_version': '4.17.0.dev0', 'feat_extract_dropout': 0.0, 'model_type': 'wav2vec2', 'num_feat_extract_layers': 7, 'hidden_size': 1024, 'feat_extract_norm': 'layer', 'feat_extract_activation': 'gelu', 'conv_dim': [512, 512, 512, 512, 512, 512, 512], 'conv_stride': [5, 2, 2, 2, 2, 2, 2], 'conv_kernel': [10, 3, 3, 3, 3, 2, 2], 'conv_bias': True, 'num_conv_pos_embeddings': 128, 'num_conv_pos_embedding_groups': 16, 'num_hidden_layers': 24, 'intermediate_size': 4096, 'hidden_act': 'gelu', 'num_attention_heads': 16, 'hidden_dropout': 0.0, 'attention_dropout': 0.0, 'activation_dropout': 0.1, 'feat_proj_dropout': 0.0, 'final_dropout': 0.0, 'layerdrop': 0.0, 'layer_norm_eps': 1e-05, 'initializer_range': 0.02, 'vocab_size': 41, 'do_stable_layer_norm': True, 'use_weighted_layer_sum': False, 'apply_spec_augment': True, 'mask_time_prob': 0.75, 'mask_time_length': 10, 'mask_time_min_masks': 2, 'mask_feature_prob': 0.25, 'mask_feature_length': 64, 'mask_feature_min_masks': 0, 'num_codevectors_per_group': 320, 'num_codevector_groups': 2, 'contrastive_logits_temperature': 0.1, 'feat_quantizer_dropout': 0.0, 'num_negatives': 100, 'codevector_dim': 768, 'proj_codevector_dim': 768, 'diversity_loss_weight': 0.1, 'ctc_loss_reduction': 'mean', 'ctc_zero_infinity': False, 'add_adapter': False, 'adapter_kernel_size': 3, 'adapter_stride': 2, 'num_adapter_layers': 3, 'output_hidden_size': 1024, 'classifier_proj_size': 256, 'tdnn_dim': [512, 512, 512, 512, 1500], 'tdnn_kernel': [5, 3, 3, 1, 1], 'tdnn_dilation': [1, 2, 3, 1, 1], 'xvector_output_dim': 512, 'output_dir': './', 'overwrite_output_dir': True, 'do_train': True, 'do_eval': True, 'do_predict': False, 'evaluation_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 8, 'per_device_eval_batch_size': 8, 'per_gpu_train_batch_size': 'None', 'per_gpu_eval_batch_size': 'None', 'gradient_accumulation_steps': 8, 'eval_accumulation_steps': 'None', 'learning_rate': 7.5e-05, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 0.2, 'max_steps': -1, 'lr_scheduler_type': 'linear', 'warmup_ratio': 0.0, 'warmup_steps': 2000, 'log_level': -1, 'log_level_replica': -1, 'log_on_each_node': True, 'logging_dir': './runs/Jan30_22-46-41_job-3261699b-76eb-4c28-8419-66a66c5c9199', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 100, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': 3, 'save_on_each_node': False, 'no_cuda': False, 'seed': 42, 'bf16': False, 'fp16': True, 'fp16_opt_level': 'O1', 'half_precision_backend': 'amp', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': 'None', 'local_rank': -1, 'xpu_backend': 'None', 'tpu_num_cores': 'None', 'tpu_metrics_debug': False, 'debug': '[]', 'dataloader_drop_last': False, 'eval_steps': 500, 'dataloader_num_workers': 0, 'past_index': -1, 'run_name': './', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': 'None', 'load_best_model_at_end': True, 'metric_for_best_model': 'loss', 'greater_is_better': False, 'ignore_data_skip': False, 'sharded_ddp': '[]', 'deepspeed': 'None', 'label_smoothing_factor': 0.0, 'optim': 'adamw_hf', 'adafactor': False, 'group_by_length': True, 'length_column_name': 'input_length', 'report_to': "['wandb']", 'ddp_find_unused_parameters': 'None', 'ddp_bucket_cap_mb': 'None', 'dataloader_pin_memory': True, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': True, 'resume_from_checkpoint': 'None', 'hub_model_id': 'None', 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'gradient_checkpointing': True, 'fp16_backend': 'auto', 'push_to_hub_model_id': 'None', 'push_to_hub_organization': 'None', 'push_to_hub_token': '<PUSH_TO_HUB_TOKEN>', '_n_gpu': 1, 'mp_parameters': '', 'train_batch_size': 8, 'eval_batch_size': 8}
+2022-01-30 22:47:40,212 INFO    MainThread:22602 [wandb_watch.py:watch():43] Watching