update model card README.md

.ipynb_checkpoints/run-checkpoint.sh

@@ -1,7 +1,7 @@
 WANDB_PROJECT=auto-speech-recognition-french
 python run_speech_recognition_ctc.py \
 --dataset_name="mozilla-foundation/common_voice_8_0" \
---model_name_or_path="facebook/wav2vec2-xls-r-300m" \
+--model_name_or_path="./checkpoint-18000" \
 --dataset_config_name="fr" \
 --tokenizer_name_or_path="./" \
 --output_dir="./" \
@@ -36,6 +36,6 @@ python run_speech_recognition_ctc.py \
 --fp16 \
 --group_by_length \
 --preprocessing_num_workers="64" \
---do_eval \
+--do_train --do_eval \
 --load_best_model_at_end \
 --push_to_hub

README.md

@@ -3,8 +3,8 @@ language:
 - fr
 license: apache-2.0
 tags:
-- robust-speech-event
 - automatic-speech-recognition
+- robust-speech-event
 - mozilla-foundation/common_voice_8_0
 - generated_from_trainer
 datasets:
@@ -20,9 +20,6 @@ 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_8_0 - FR dataset.
-It achieves the following results on the evaluation set:
-- Loss: 0.2388
-- Wer: 0.3681
 
 ## Model description
 
@@ -47,15 +44,10 @@ The following hyperparameters were used during training:
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 1500
-- num_epochs: 2.0
+- lr_scheduler_warmup_steps: 2700
+- num_epochs: 4.0
 - mixed_precision_training: Native AMP
 
-### Training results
-
-
-
-
 ### Framework versions
 
 - Transformers 4.17.0.dev0

run.sh

@@ -1,7 +1,7 @@
 WANDB_PROJECT=auto-speech-recognition-french
 python run_speech_recognition_ctc.py \
 --dataset_name="mozilla-foundation/common_voice_8_0" \
---model_name_or_path="facebook/wav2vec2-xls-r-300m" \
+--model_name_or_path="./checkpoint-18000" \
 --dataset_config_name="fr" \
 --tokenizer_name_or_path="./" \
 --output_dir="./" \

wandb/run-20220201_103004-1yfj7vwy/files/output.log

@@ -9,3 +9,22 @@
 Saving model checkpoint to ./
 Configuration saved in ./config.json
 Model weights saved in ./pytorch_model.bin
+Configuration saved in ./preprocessor_config.json
+02/01/2022 10:31:23 - WARNING - huggingface_hub.repository - Several commits (53) will be pushed upstream.
+02/01/2022 10:31:23 - WARNING - huggingface_hub.repository - The progress bars may be unreliable.
+Several commits (53) will be pushed upstream.
+The progress bars may be unreliable.
+Upload file pytorch_model.bin:   1%|▉                                                                                                      | 11.0M/1.18G [00:01<01:50, 11.3MB/s]
+Upload file training_args.bin: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2.98k/2.98k [00:00<?, ?B/s]
+
+
+
+
+
+Upload file wandb/run-20220130_191954-1mtmnz5y/run-1mtmnz5y.wandb: 100%|█████████████████████████████████████████████████████████████████████| 201M/201M [00:21<00:00, 25.5MB/s]
+02/01/2022 10:32:41 - WARNING - huggingface_hub.repository - To https://huggingface.co/AlexN/xls-r-300m-fr
+   d6fc661..f25a924  main -> main
+Upload file pytorch_model.bin: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████| 1.18G/1.18G [01:12<00:00, 17.3MB/s]
+Upload file training_args.bin: 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2.98k/2.98k [01:12<?, ?B/s]
+Upload file wandb/run-20220130_191954-1mtmnz5y/run-1mtmnz5y.wandb: 100%|█████████████████████████████████████████████████████████████████████| 201M/201M [01:12<00:00, 2.89MB/s]
+

wandb/run-20220201_103004-1yfj7vwy/logs/debug-internal.log

@@ -56,3 +56,25 @@
 2022-02-01 10:31:07,163 DEBUG   SenderThread:61045 [sender.py:send_request():248] send_request: stop_status
 2022-02-01 10:31:22,319 DEBUG   HandlerThread:61045 [handler.py:handle_request():130] handle_request: stop_status
 2022-02-01 10:31:22,320 DEBUG   SenderThread:61045 [sender.py:send_request():248] send_request: stop_status
+2022-02-01 10:31:24,949 INFO    Thread-8  :61045 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220201_103004-1yfj7vwy/files/output.log
+2022-02-01 10:31:30,952 INFO    Thread-8  :61045 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220201_103004-1yfj7vwy/files/output.log
+2022-02-01 10:31:32,953 INFO    Thread-8  :61045 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220201_103004-1yfj7vwy/files/output.log
+2022-02-01 10:31:34,785 DEBUG   SenderThread:61045 [sender.py:send():234] send: stats
+2022-02-01 10:31:34,954 INFO    Thread-8  :61045 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220201_103004-1yfj7vwy/files/output.log
+2022-02-01 10:31:36,955 INFO    Thread-8  :61045 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220201_103004-1yfj7vwy/files/output.log
+2022-02-01 10:31:37,488 DEBUG   HandlerThread:61045 [handler.py:handle_request():130] handle_request: stop_status
+2022-02-01 10:31:37,489 DEBUG   SenderThread:61045 [sender.py:send_request():248] send_request: stop_status
+2022-02-01 10:31:38,957 INFO    Thread-8  :61045 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220201_103004-1yfj7vwy/files/output.log
+2022-02-01 10:31:52,642 DEBUG   HandlerThread:61045 [handler.py:handle_request():130] handle_request: stop_status
+2022-02-01 10:31:52,643 DEBUG   SenderThread:61045 [sender.py:send_request():248] send_request: stop_status
+2022-02-01 10:31:52,963 INFO    Thread-8  :61045 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220201_103004-1yfj7vwy/files/output.log
+2022-02-01 10:32:05,018 DEBUG   SenderThread:61045 [sender.py:send():234] send: stats
+2022-02-01 10:32:07,832 DEBUG   HandlerThread:61045 [handler.py:handle_request():130] handle_request: stop_status
+2022-02-01 10:32:07,833 DEBUG   SenderThread:61045 [sender.py:send_request():248] send_request: stop_status
+2022-02-01 10:32:23,003 DEBUG   HandlerThread:61045 [handler.py:handle_request():130] handle_request: stop_status
+2022-02-01 10:32:23,003 DEBUG   SenderThread:61045 [sender.py:send_request():248] send_request: stop_status
+2022-02-01 10:32:35,259 DEBUG   SenderThread:61045 [sender.py:send():234] send: stats
+2022-02-01 10:32:38,164 DEBUG   HandlerThread:61045 [handler.py:handle_request():130] handle_request: stop_status
+2022-02-01 10:32:38,165 DEBUG   SenderThread:61045 [sender.py:send_request():248] send_request: stop_status
+2022-02-01 10:32:42,989 INFO    Thread-8  :61045 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220201_103004-1yfj7vwy/files/output.log
+2022-02-01 10:32:44,990 INFO    Thread-8  :61045 [dir_watcher.py:_on_file_modified():230] file/dir modified: /workspace/xls-r-300m-fr/wandb/run-20220201_103004-1yfj7vwy/files/output.log