Training in progress, step 500

.gitattributes

@@ -56,3 +56,4 @@ wandb/run-20220302_085255-16llzpbl/run-16llzpbl.wandb filter=lfs diff=lfs merge=
 wandb/run-20220302_114207-c11kvzxa/run-c11kvzxa.wandb filter=lfs diff=lfs merge=lfs -text
 wandb/run-20220302_135216-v36bieky/run-v36bieky.wandb filter=lfs diff=lfs merge=lfs -text
 wandb/run-20220302_154455-17zs7rwf/run-17zs7rwf.wandb filter=lfs diff=lfs merge=lfs -text
+wandb/run-20220302_180214-gd4yxtv7/run-gd4yxtv7.wandb filter=lfs diff=lfs merge=lfs -text

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:10fb64407c91b1f7887407a888238b97f28e28b48d23e30379036dc620527113
+oid sha256:2def16a7d4d0cdfa93daaebef23a338fdee1afff26cd8968f49f55a11206c14e
 size 3210531882

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a0b256199bd919199c275f7eb37fb99f8e7f197271a355bcf25f9a59f78af747
+oid sha256:6b692dcb0344065a7ff0122b7f7ee937d91bcefa1e0922c4b04076a65723d16e
 size 3119

wandb/debug-internal.log

@@ -1 +1 @@
-run-20220302_154455-17zs7rwf/logs/debug-internal.log
+run-20220302_180214-gd4yxtv7/logs/debug-internal.log

wandb/debug.log

@@ -1 +1 @@
-run-20220302_154455-17zs7rwf/logs/debug.log
+run-20220302_180214-gd4yxtv7/logs/debug.log

wandb/latest-run

@@ -1 +1 @@
-run-20220302_154455-17zs7rwf
+run-20220302_180214-gd4yxtv7

wandb/run-20220302_154455-17zs7rwf/files/config.yaml

@@ -10673,7 +10673,14 @@ _wandb:
       - 1
       - 5
       - 11
+      2:
+      - 1
+      - 5
+      - 11
+      - 12
       3:
+      - 1
+      - 7
       - 13
       4: 3.9.5
       5: 0.12.10

wandb/run-20220302_154455-17zs7rwf/files/output.log

@@ -4501,3 +4501,10 @@ Upload file wandb/run-20220302_154455-17zs7rwf/run-17zs7rwf.wandb: 100%|██
   eval_steps_per_second   =      0.238
   eval_wer                =     2.4429
 [INFO|trainer.py:2369] 2022-03-02 17:44:12,720 >>   Batch size = 12100%|███████████| 35.9M/35.9M [00:02<00:00, 18.7MB/s][WARNING|modeling_utils.py:388] 2022-03-02 16:06:42,638 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+[INFO|trainer.py:2369] 2022-03-02 17:44:12,720 >>   Batch size = 12100%|███████████| 35.9M/35.9M [00:02<00:00, 18.7MB/s][WARNING|modeling_utils.py:388] 2022-03-02 16:06:42,638 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+[INFO|trainer.py:2369] 2022-03-02 17:44:12,720 >>   Batch size = 12100%|███████████| 35.9M/35.9M [00:02<00:00, 18.7MB/s][WARNING|modeling_utils.py:388] 2022-03-02 16:06:42,638 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+03/02/2022 18:00:30 - WARNING - huggingface_hub.repository - To https://huggingface.co/sanchit-gandhi/wav2vec2-gpt2-wandb-grid-search
+Upload file wandb/run-20220302_154455-17zs7rwf/run-17zs7rwf.wandb: 100%|███████████| 36.0M/36.0M [00:02<00:00, 18.8MB/s][WARNING|modeling_utils.py:388] 2022-03-02 16:06:42,638 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+    return ModelInfo(**d)6, in __init__ansformers/src/transformers/trainer.py", line 2725, in create_model_card31, in mainARNING|modeling_utils.py:388] 2022-03-02 16:06:42,638 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+    return ModelInfo(**d)6, in __init__ansformers/src/transformers/trainer.py", line 2725, in create_model_card31, in mainARNING|modeling_utils.py:388] 2022-03-02 16:06:42,638 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed
+    return ModelInfo(**d)6, in __init__ansformers/src/transformers/trainer.py", line 2725, in create_model_card31, in mainARNING|modeling_utils.py:388] 2022-03-02 16:06:42,638 >> Could not estimate the number of tokens of the input, floating-point operations will not be computed

wandb/run-20220302_154455-17zs7rwf/logs/debug-internal.log

@@ -12461,3 +12461,117 @@
 2022-03-02 18:00:08,582 DEBUG   SenderThread:261319 [sender.py:send():235] send: stats
 2022-03-02 18:00:13,125 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: stop_status
 2022-03-02 18:00:13,125 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: stop_status
+2022-03-02 18:00:28,197 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: stop_status
+2022-03-02 18:00:28,197 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: stop_status
+2022-03-02 18:00:29,497 INFO    Thread-8  :261319 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/run-20220302_154455-17zs7rwf/files/output.log
+2022-03-02 18:00:31,497 INFO    Thread-8  :261319 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/run-20220302_154455-17zs7rwf/files/output.log
+2022-03-02 18:00:35,499 INFO    Thread-8  :261319 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/run-20220302_154455-17zs7rwf/files/output.log
+2022-03-02 18:00:35,609 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: poll_exit
+2022-03-02 18:00:35,610 DEBUG   SenderThread:261319 [sender.py:send():235] send: telemetry
+2022-03-02 18:00:35,610 DEBUG   SenderThread:261319 [sender.py:send():235] send: exit
+2022-03-02 18:00:35,610 INFO    SenderThread:261319 [sender.py:send_exit():371] handling exit code: 1
+2022-03-02 18:00:35,610 INFO    SenderThread:261319 [sender.py:send_exit():373] handling runtime: 8138
+2022-03-02 18:00:35,666 INFO    SenderThread:261319 [sender.py:_save_file():944] saving file wandb-summary.json with policy end
+2022-03-02 18:00:35,667 INFO    SenderThread:261319 [sender.py:send_exit():379] send defer
+2022-03-02 18:00:35,667 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: poll_exit
+2022-03-02 18:00:35,667 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: defer
+2022-03-02 18:00:35,667 INFO    HandlerThread:261319 [handler.py:handle_request_defer():154] handle defer: 0
+2022-03-02 18:00:35,668 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: defer
+2022-03-02 18:00:35,668 INFO    SenderThread:261319 [sender.py:send_request_defer():388] handle sender defer: 0
+2022-03-02 18:00:35,668 INFO    SenderThread:261319 [sender.py:transition_state():392] send defer: 1
+2022-03-02 18:00:35,668 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: defer
+2022-03-02 18:00:35,668 INFO    HandlerThread:261319 [handler.py:handle_request_defer():154] handle defer: 1
+2022-03-02 18:00:35,791 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: poll_exit
+2022-03-02 18:00:35,791 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: defer
+2022-03-02 18:00:35,792 INFO    SenderThread:261319 [sender.py:send_request_defer():388] handle sender defer: 1
+2022-03-02 18:00:35,792 INFO    SenderThread:261319 [sender.py:transition_state():392] send defer: 2
+2022-03-02 18:00:35,792 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: poll_exit
+2022-03-02 18:00:35,792 DEBUG   SenderThread:261319 [sender.py:send():235] send: stats
+2022-03-02 18:00:35,792 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: defer
+2022-03-02 18:00:35,793 INFO    HandlerThread:261319 [handler.py:handle_request_defer():154] handle defer: 2
+2022-03-02 18:00:35,793 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: defer
+2022-03-02 18:00:35,793 INFO    SenderThread:261319 [sender.py:send_request_defer():388] handle sender defer: 2
+2022-03-02 18:00:35,793 INFO    SenderThread:261319 [sender.py:transition_state():392] send defer: 3
+2022-03-02 18:00:35,793 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: defer
+2022-03-02 18:00:35,793 INFO    HandlerThread:261319 [handler.py:handle_request_defer():154] handle defer: 3
+2022-03-02 18:00:35,851 DEBUG   SenderThread:261319 [sender.py:send():235] send: summary
+2022-03-02 18:00:35,938 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: poll_exit
+2022-03-02 18:00:35,941 INFO    SenderThread:261319 [sender.py:_save_file():944] saving file wandb-summary.json with policy end
+2022-03-02 18:00:35,942 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: defer
+2022-03-02 18:00:35,942 INFO    SenderThread:261319 [sender.py:send_request_defer():388] handle sender defer: 3
+2022-03-02 18:00:35,942 INFO    SenderThread:261319 [sender.py:transition_state():392] send defer: 4
+2022-03-02 18:00:35,942 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: poll_exit
+2022-03-02 18:00:35,943 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: defer
+2022-03-02 18:00:35,943 INFO    HandlerThread:261319 [handler.py:handle_request_defer():154] handle defer: 4
+2022-03-02 18:00:35,943 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: defer
+2022-03-02 18:00:35,943 INFO    SenderThread:261319 [sender.py:send_request_defer():388] handle sender defer: 4
+2022-03-02 18:00:36,044 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: poll_exit
+2022-03-02 18:00:36,540 INFO    Thread-8  :261319 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/run-20220302_154455-17zs7rwf/files/output.log
+2022-03-02 18:00:36,550 INFO    Thread-8  :261319 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/run-20220302_154455-17zs7rwf/files/wandb-summary.json
+2022-03-02 18:00:36,840 INFO    SenderThread:261319 [sender.py:transition_state():392] send defer: 5
+2022-03-02 18:00:36,840 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: poll_exit
+2022-03-02 18:00:36,841 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: defer
+2022-03-02 18:00:36,841 INFO    HandlerThread:261319 [handler.py:handle_request_defer():154] handle defer: 5
+2022-03-02 18:00:36,841 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: defer
+2022-03-02 18:00:36,841 INFO    SenderThread:261319 [sender.py:send_request_defer():388] handle sender defer: 5
+2022-03-02 18:00:36,841 INFO    SenderThread:261319 [dir_watcher.py:finish():283] shutting down directory watcher
+2022-03-02 18:00:36,942 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: poll_exit
+2022-03-02 18:00:37,530 INFO    Thread-8  :261319 [dir_watcher.py:_on_file_modified():230] file/dir modified: /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/run-20220302_154455-17zs7rwf/files/config.yaml
+2022-03-02 18:00:37,530 INFO    SenderThread:261319 [dir_watcher.py:finish():313] scan: /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/run-20220302_154455-17zs7rwf/files
+2022-03-02 18:00:37,531 INFO    SenderThread:261319 [dir_watcher.py:finish():327] scan save: /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/run-20220302_154455-17zs7rwf/files/wandb-metadata.json wandb-metadata.json
+2022-03-02 18:00:37,531 INFO    SenderThread:261319 [dir_watcher.py:finish():327] scan save: /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/run-20220302_154455-17zs7rwf/files/output.log output.log
+2022-03-02 18:00:37,531 INFO    SenderThread:261319 [dir_watcher.py:finish():327] scan save: /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/run-20220302_154455-17zs7rwf/files/wandb-summary.json wandb-summary.json
+2022-03-02 18:00:37,534 INFO    SenderThread:261319 [dir_watcher.py:finish():327] scan save: /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/run-20220302_154455-17zs7rwf/files/requirements.txt requirements.txt
+2022-03-02 18:00:37,537 INFO    SenderThread:261319 [dir_watcher.py:finish():327] scan save: /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/run-20220302_154455-17zs7rwf/files/config.yaml config.yaml
+2022-03-02 18:00:37,537 INFO    SenderThread:261319 [sender.py:transition_state():392] send defer: 6
+2022-03-02 18:00:37,538 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: poll_exit
+2022-03-02 18:00:37,541 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: defer
+2022-03-02 18:00:37,541 INFO    HandlerThread:261319 [handler.py:handle_request_defer():154] handle defer: 6
+2022-03-02 18:00:37,544 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: defer
+2022-03-02 18:00:37,544 INFO    SenderThread:261319 [sender.py:send_request_defer():388] handle sender defer: 6
+2022-03-02 18:00:37,544 INFO    SenderThread:261319 [file_pusher.py:finish():177] shutting down file pusher
+2022-03-02 18:00:37,641 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: poll_exit
+2022-03-02 18:00:37,641 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: poll_exit
+2022-03-02 18:00:37,743 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: poll_exit
+2022-03-02 18:00:37,743 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: poll_exit
+2022-03-02 18:00:37,844 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: poll_exit
+2022-03-02 18:00:37,845 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: poll_exit
+2022-03-02 18:00:37,858 INFO    Thread-14 :261319 [upload_job.py:push():137] Uploaded file /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/run-20220302_154455-17zs7rwf/files/requirements.txt
+2022-03-02 18:00:37,888 INFO    Thread-15 :261319 [upload_job.py:push():137] Uploaded file /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/run-20220302_154455-17zs7rwf/files/config.yaml
+2022-03-02 18:00:37,901 INFO    Thread-13 :261319 [upload_job.py:push():137] Uploaded file /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/run-20220302_154455-17zs7rwf/files/wandb-summary.json
+2022-03-02 18:00:37,933 INFO    Thread-12 :261319 [upload_job.py:push():137] Uploaded file /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/run-20220302_154455-17zs7rwf/files/output.log
+2022-03-02 18:00:37,946 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: poll_exit
+2022-03-02 18:00:37,946 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: poll_exit
+2022-03-02 18:00:38,048 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: poll_exit
+2022-03-02 18:00:38,048 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: poll_exit
+2022-03-02 18:00:38,134 INFO    Thread-7  :261319 [sender.py:transition_state():392] send defer: 7
+2022-03-02 18:00:38,134 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: defer
+2022-03-02 18:00:38,134 INFO    HandlerThread:261319 [handler.py:handle_request_defer():154] handle defer: 7
+2022-03-02 18:00:38,135 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: defer
+2022-03-02 18:00:38,135 INFO    SenderThread:261319 [sender.py:send_request_defer():388] handle sender defer: 7
+2022-03-02 18:00:38,150 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: poll_exit
+2022-03-02 18:00:39,632 INFO    SenderThread:261319 [sender.py:transition_state():392] send defer: 8
+2022-03-02 18:00:39,633 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: poll_exit
+2022-03-02 18:00:39,633 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: defer
+2022-03-02 18:00:39,633 INFO    HandlerThread:261319 [handler.py:handle_request_defer():154] handle defer: 8
+2022-03-02 18:00:39,633 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: defer
+2022-03-02 18:00:39,633 INFO    SenderThread:261319 [sender.py:send_request_defer():388] handle sender defer: 8
+2022-03-02 18:00:39,634 INFO    SenderThread:261319 [sender.py:transition_state():392] send defer: 9
+2022-03-02 18:00:39,635 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: defer
+2022-03-02 18:00:39,635 INFO    HandlerThread:261319 [handler.py:handle_request_defer():154] handle defer: 9
+2022-03-02 18:00:39,636 DEBUG   SenderThread:261319 [sender.py:send():235] send: final
+2022-03-02 18:00:39,636 DEBUG   SenderThread:261319 [sender.py:send():235] send: footer
+2022-03-02 18:00:39,636 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: defer
+2022-03-02 18:00:39,636 INFO    SenderThread:261319 [sender.py:send_request_defer():388] handle sender defer: 9
+2022-03-02 18:00:39,734 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: poll_exit
+2022-03-02 18:00:39,734 DEBUG   SenderThread:261319 [sender.py:send_request():249] send_request: poll_exit
+2022-03-02 18:00:39,735 INFO    SenderThread:261319 [file_pusher.py:join():182] waiting for file pusher
+2022-03-02 18:00:39,809 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: get_summary
+2022-03-02 18:00:39,917 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: sampled_history
+2022-03-02 18:00:39,920 DEBUG   HandlerThread:261319 [handler.py:handle_request():131] handle_request: shutdown
+2022-03-02 18:00:39,920 INFO    HandlerThread:261319 [handler.py:finish():739] shutting down handler
+2022-03-02 18:00:40,635 INFO    WriterThread:261319 [datastore.py:close():281] close: /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/run-20220302_154455-17zs7rwf/run-17zs7rwf.wandb
+2022-03-02 18:00:40,808 INFO    SenderThread:261319 [sender.py:finish():1075] shutting down sender
+2022-03-02 18:00:40,808 INFO    SenderThread:261319 [file_pusher.py:finish():177] shutting down file pusher
+2022-03-02 18:00:40,808 INFO    SenderThread:261319 [file_pusher.py:join():182] waiting for file pusher
+2022-03-02 18:00:40,816 INFO    MainThread:261319 [internal.py:handle_exit():79] Internal process exited

wandb/run-20220302_154455-17zs7rwf/logs/debug.log

@@ -25,3 +25,109 @@ config: {}
 2022-03-02 15:44:57,210 INFO    MainThread:261220 [wandb_init.py:init():651] run started, returning control to user process
 2022-03-02 15:44:57,213 INFO    MainThread:261220 [wandb_run.py:_config_callback():966] config_cb None None {'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'torch.float32', 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': False, 'is_encoder_decoder': True, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 50, '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': ['SpeechEncoderDecoderModel'], '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': None, 'pad_token_id': 50256, 'eos_token_id': 50256, 'sep_token_id': None, 'decoder_start_token_id': 50256, 'task_specific_params': None, 'problem_type': None, '_name_or_path': './', 'transformers_version': None, 'decoder': {'vocab_size': 50257, 'n_positions': 1024, 'n_embd': 1024, 'n_layer': 24, 'n_head': 16, 'n_inner': None, 'activation_function': 'gelu_new', 'resid_pdrop': 0.0, 'embd_pdrop': 0.0, 'attn_pdrop': 0.0, 'layer_norm_epsilon': 1e-05, 'initializer_range': 0.02, 'summary_type': 'cls_index', 'summary_use_proj': True, 'summary_activation': None, 'summary_first_dropout': 0.0, 'summary_proj_to_labels': True, 'scale_attn_weights': True, 'use_cache': False, 'scale_attn_by_inverse_layer_idx': False, 'reorder_and_upcast_attn': False, 'bos_token_id': 50256, 'eos_token_id': 50256, 'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': None, 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'is_encoder_decoder': False, 'is_decoder': True, 'cross_attention_hidden_size': None, 'add_cross_attention': True, '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': ['GPT2LMHeadModel'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'pad_token_id': None, 'sep_token_id': None, 'decoder_start_token_id': None, 'task_specific_params': {'text-generation': {'do_sample': True, 'max_length': 50}}, 'problem_type': None, '_name_or_path': 'gpt2-medium', 'transformers_version': '4.17.0.dev0', 'n_ctx': 1024, 'n_special': 0, 'predict_special_tokens': True, 'model_type': 'gpt2'}, 'encoder': {'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': None, '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': 0, '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-large-lv60', 'transformers_version': '4.17.0.dev0', 'feat_extract_dropout': 0.0, 'gradient_checkpointing': False, 'hidden_dropout_prob': 0.0, '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': 0.0, 'feat_proj_dropout': 0.0, 'final_dropout': 0.0, 'layerdrop': 0.0, 'layer_norm_eps': 1e-05, 'initializer_range': 0.02, 'vocab_size': 32, 'do_stable_layer_norm': True, 'use_weighted_layer_sum': False, 'apply_spec_augment': False, 'mask_time_prob': 0.0, 'mask_time_length': 10, 'mask_time_min_masks': 2, 'mask_feature_prob': 0.0, 'mask_feature_length': 10, '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': 'sum', 'ctc_zero_infinity': False, 'add_adapter': True, '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, 'model_type': 'wav2vec2'}, 'model_type': 'speech-encoder-decoder', 'processor_class': 'Wav2Vec2Processor', 'use_cache': False, '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': 12, 'per_device_eval_batch_size': 12, 'per_gpu_train_batch_size': 'None', 'per_gpu_eval_batch_size': 'None', 'gradient_accumulation_steps': 2, 'eval_accumulation_steps': 'None', 'learning_rate': 0.0003, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 1.0, 'max_steps': -1, 'lr_scheduler_type': 'linear', 'warmup_ratio': 0.0, 'warmup_steps': 500, 'log_level': -1, 'log_level_replica': -1, 'log_on_each_node': True, 'logging_dir': './runs/Mar02_15-44-13_sanchit--v100', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 1, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': 1, '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': False, 'metric_for_best_model': 'None', 'greater_is_better': 'None', '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': '', 'sortish_sampler': False, 'predict_with_generate': True, 'generation_max_length': 40, 'generation_num_beams': 1, 'train_batch_size': 12, 'eval_batch_size': 12}
 2022-03-02 15:44:57,216 INFO    MainThread:261220 [wandb_watch.py:watch():43] Watching
+2022-03-02 18:00:33,516 INFO    MainThread:261220 [wandb_run.py:_atexit_cleanup():1797] got exitcode: 1
+2022-03-02 18:00:33,520 INFO    MainThread:261220 [wandb_run.py:_restore():1769] restore
+2022-03-02 18:00:35,667 INFO    MainThread:261220 [wandb_run.py:_wait_for_finish():1929] got exit ret: file_counts {
+  wandb_count: 1
+}
+pusher_stats {
+  uploaded_bytes: 2095
+  total_bytes: 2095
+}
+
+2022-03-02 18:00:35,793 INFO    MainThread:261220 [wandb_run.py:_wait_for_finish():1929] got exit ret: file_counts {
+  wandb_count: 1
+}
+pusher_stats {
+  uploaded_bytes: 2095
+  total_bytes: 2095
+}
+
+2022-03-02 18:00:35,943 INFO    MainThread:261220 [wandb_run.py:_wait_for_finish():1929] got exit ret: file_counts {
+  wandb_count: 1
+}
+pusher_stats {
+  uploaded_bytes: 2095
+  total_bytes: 2095
+}
+
+2022-03-02 18:00:36,841 INFO    MainThread:261220 [wandb_run.py:_wait_for_finish():1929] got exit ret: file_counts {
+  wandb_count: 1
+}
+pusher_stats {
+  uploaded_bytes: 2095
+  total_bytes: 2095
+}
+
+2022-03-02 18:00:37,538 INFO    MainThread:261220 [wandb_run.py:_wait_for_finish():1929] got exit ret: file_counts {
+  wandb_count: 4
+}
+pusher_stats {
+  uploaded_bytes: 2095
+  total_bytes: 2653576
+}
+
+2022-03-02 18:00:37,642 INFO    MainThread:261220 [wandb_run.py:_wait_for_finish():1929] got exit ret: file_counts {
+  wandb_count: 5
+}
+pusher_stats {
+  uploaded_bytes: 34863
+  total_bytes: 2951743
+}
+
+2022-03-02 18:00:37,744 INFO    MainThread:261220 [wandb_run.py:_wait_for_finish():1929] got exit ret: file_counts {
+  wandb_count: 5
+}
+pusher_stats {
+  uploaded_bytes: 2951743
+  total_bytes: 2951743
+}
+
+2022-03-02 18:00:37,845 INFO    MainThread:261220 [wandb_run.py:_wait_for_finish():1929] got exit ret: file_counts {
+  wandb_count: 5
+}
+pusher_stats {
+  uploaded_bytes: 2951743
+  total_bytes: 2951743
+}
+
+2022-03-02 18:00:37,947 INFO    MainThread:261220 [wandb_run.py:_wait_for_finish():1929] got exit ret: file_counts {
+  wandb_count: 5
+}
+pusher_stats {
+  uploaded_bytes: 2951743
+  total_bytes: 2951743
+}
+
+2022-03-02 18:00:38,049 INFO    MainThread:261220 [wandb_run.py:_wait_for_finish():1929] got exit ret: file_counts {
+  wandb_count: 5
+}
+pusher_stats {
+  uploaded_bytes: 2951743
+  total_bytes: 2951743
+}
+
+2022-03-02 18:00:39,633 INFO    MainThread:261220 [wandb_run.py:_wait_for_finish():1929] got exit ret: file_counts {
+  wandb_count: 5
+}
+pusher_stats {
+  uploaded_bytes: 2951743
+  total_bytes: 2951743
+}
+
+2022-03-02 18:00:39,808 INFO    MainThread:261220 [wandb_run.py:_wait_for_finish():1929] got exit ret: done: true
+exit_result {
+}
+file_counts {
+  wandb_count: 5
+}
+pusher_stats {
+  uploaded_bytes: 2951743
+  total_bytes: 2951743
+}
+local_info {
+}
+
+2022-03-02 18:00:40,957 INFO    MainThread:261220 [wandb_run.py:_append_history():2144] rendering history
+2022-03-02 18:00:40,958 INFO    MainThread:261220 [wandb_run.py:_append_summary():2102] rendering summary
+2022-03-02 18:00:40,959 INFO    MainThread:261220 [wandb_run.py:_append_files():2194] logging synced files

wandb/run-20220302_154455-17zs7rwf/run-17zs7rwf.wandb

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3b7c65fdeb18a35ae15dbf98835ae05ace68d5bc95ac05a51f8928cbc68c9f4b
-size 37720132
+oid sha256:b47d1c25c6eeb525c82f99d052c62f47fa740e62f254b1f4fdb459bbd479c5e5
+size 39299803

wandb/run-20220302_180214-gd4yxtv7/files/requirements.txt

@@ -0,0 +1,184 @@
+absl-py==1.0.0
+aiohttp==3.8.1
+aiosignal==1.2.0
+anyio==3.5.0
+appdirs==1.4.4
+argon2-cffi-bindings==21.2.0
+argon2-cffi==21.3.0
+asttokens==2.0.5
+async-timeout==4.0.2
+attrs==21.4.0
+audioread==2.1.9
+babel==2.9.1
+backcall==0.2.0
+bitsandbytes-cuda113==0.26.0
+black==22.1.0
+bleach==4.1.0
+cachetools==5.0.0
+certifi==2021.10.8
+cffi==1.15.0
+charset-normalizer==2.0.11
+chex==0.1.0
+click==8.0.3
+clldutils==3.10.1
+colorlog==6.6.0
+csvw==1.11.0
+cycler==0.11.0
+datasets==1.18.3
+debugpy==1.5.1
+decorator==5.1.1
+defusedxml==0.7.1
+dill==0.3.4
+dlinfo==1.2.1
+dm-tree==0.1.6
+docker-pycreds==0.4.0
+entrypoints==0.4
+executing==0.8.2
+filelock==3.4.2
+flatbuffers==2.0
+flax==0.4.0
+fonttools==4.29.1
+frozenlist==1.3.0
+fsspec==2022.1.0
+gitdb==4.0.9
+gitpython==3.1.27
+google-auth-oauthlib==0.4.6
+google-auth==2.6.0
+grpcio==1.43.0
+huggingface-hub==0.4.0
+hypothesis==6.36.1
+idna==3.3
+importlib-metadata==4.10.1
+ipykernel==6.8.0
+ipython-genutils==0.2.0
+ipython==8.0.1
+ipywidgets==7.6.5
+isodate==0.6.1
+jax==0.2.28
+jaxlib==0.1.76+cuda11.cudnn82
+jedi==0.18.1
+jinja2==3.0.3
+jiwer==2.3.0
+joblib==1.1.0
+json5==0.9.6
+jsonschema==4.4.0
+jupyter-client==7.1.2
+jupyter-console==6.4.0
+jupyter-core==4.9.1
+jupyter-server==1.13.5
+jupyter==1.0.0
+jupyterlab-pygments==0.1.2
+jupyterlab-server==2.10.3
+jupyterlab-widgets==1.0.2
+jupyterlab==3.2.9
+kiwisolver==1.3.2
+librosa==0.8.1
+llvmlite==0.38.0
+markdown==3.3.6
+markupsafe==2.0.1
+matplotlib-inline==0.1.3
+matplotlib==3.5.1
+mistune==0.8.4
+msgpack==1.0.3
+multidict==6.0.2
+multiprocess==0.70.12.2
+mypy-extensions==0.4.3
+nbclassic==0.3.5
+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==1.21.5
+oauthlib==3.2.0
+opt-einsum==3.3.0
+optax==0.1.0
+packaging==21.3
+pandas==1.4.0
+pandocfilters==1.5.0
+parso==0.8.3
+pathspec==0.9.0
+pathtools==0.1.2
+pexpect==4.8.0
+phonemizer==3.0.1
+pickleshare==0.7.5
+pillow==9.0.0
+pip==22.0.2
+pkg-resources==0.0.0
+platformdirs==2.4.1
+pooch==1.6.0
+prometheus-client==0.13.1
+promise==2.3
+prompt-toolkit==3.0.26
+protobuf==3.19.4
+psutil==5.9.0
+ptyprocess==0.7.0
+pure-eval==0.2.2
+pyarrow==6.0.1
+pyasn1-modules==0.2.8
+pyasn1==0.4.8
+pycparser==2.21
+pyctcdecode==0.3.0
+pygments==2.11.2
+pygtrie==2.4.2
+pyparsing==3.0.7
+pyrsistent==0.18.1
+python-dateutil==2.8.2
+python-levenshtein==0.12.2
+pytz==2021.3
+pyyaml==6.0
+pyzmq==22.3.0
+qtconsole==5.2.2
+qtpy==2.0.1
+regex==2022.1.18
+requests-oauthlib==1.3.1
+requests==2.27.1
+resampy==0.2.2
+rfc3986==2.0.0
+rsa==4.8
+sacremoses==0.0.47
+scikit-learn==1.0.2
+scipy==1.7.3
+segments==2.2.0
+send2trash==1.8.0
+sentry-sdk==1.5.6
+setuptools==44.1.1
+shortuuid==1.0.8
+six==1.16.0
+smmap==5.0.0
+sniffio==1.2.0
+sortedcontainers==2.4.0
+soundfile==0.10.3.post1
+stack-data==0.1.4
+tabulate==0.8.9
+tensorboard-data-server==0.6.1
+tensorboard-plugin-wit==1.8.1
+tensorboard==2.8.0
+termcolor==1.1.0
+terminado==0.13.1
+testpath==0.5.0
+threadpoolctl==3.1.0
+tokenizers==0.11.4
+tomli==2.0.0
+toolz==0.11.2
+torch==1.10.2+cu113
+torchaudio==0.10.2+cu113
+tornado==6.1
+tqdm==4.62.3
+traitlets==5.1.1
+transformers==4.17.0.dev0
+typing-extensions==3.10.0.2
+uritemplate==4.1.1
+urllib3==1.26.8
+wandb==0.12.10
+wcwidth==0.2.5
+webencodings==0.5.1
+websocket-client==1.2.3
+werkzeug==2.0.2
+wheel==0.37.1
+widgetsnbextension==3.5.2
+xxhash==2.0.2
+yarl==1.7.2
+yaspin==2.1.0
+zipp==3.7.0

wandb/run-20220302_180214-gd4yxtv7/files/wandb-metadata.json

@@ -0,0 +1,60 @@
+{
+    "os": "Linux-5.11.0-1028-gcp-x86_64-with-glibc2.33",
+    "python": "3.9.5",
+    "heartbeatAt": "2022-03-02T18:02:16.043267",
+    "startedAt": "2022-03-02T18:02:14.973524",
+    "docker": null,
+    "gpu": "Tesla V100-SXM2-16GB",
+    "gpu_count": 2,
+    "cpu_count": 16,
+    "cuda": null,
+    "args": [
+        "--dataset_name=librispeech_asr",
+        "--model_name_or_path=./",
+        "--tokenizer_name=./",
+        "--dataset_config_name=clean",
+        "--train_split_name=train.100",
+        "--eval_split_name=validation",
+        "--output_dir=./",
+        "--preprocessing_num_workers=1",
+        "--length_column_name=input_length",
+        "--overwrite_output_dir",
+        "--num_train_epochs=1",
+        "--per_device_train_batch_size=12",
+        "--per_device_eval_batch_size=12",
+        "--gradient_accumulation_steps=4",
+        "--generation_max_length=40",
+        "--generation_num_beams=1",
+        "--learning_rate=3e-4",
+        "--warmup_steps=500",
+        "--evaluation_strategy=steps",
+        "--text_column_name=text",
+        "--save_steps=500",
+        "--eval_steps=500",
+        "--logging_steps=1",
+        "--save_total_limit=1",
+        "--freeze_feature_encoder",
+        "--gradient_checkpointing",
+        "--fp16",
+        "--group_by_length",
+        "--predict_with_generate",
+        "--do_lower_case",
+        "--do_train",
+        "--do_eval",
+        "--report_to=wandb",
+        "--push_to_hub",
+        "--use_auth_token"
+    ],
+    "state": "running",
+    "program": "/home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/run_speech_recognition_seq2seq.py",
+    "codePath": "run_speech_recognition_seq2seq.py",
+    "git": {
+        "remote": "https://huggingface.co/sanchit-gandhi/wav2vec2-gpt2-wandb-grid-search",
+        "commit": "45dc0be6c5c02a6d931977ada65d1a4d583e2ad3"
+    },
+    "email": "sanchit@huggingface.co",
+    "root": "/home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search",
+    "host": "sanchit--v100",
+    "username": "sanchit_huggingface_co",
+    "executable": "/home/sanchit_huggingface_co/gcp/bin/python"
+}

wandb/run-20220302_180214-gd4yxtv7/logs/debug.log

@@ -0,0 +1,27 @@
+2022-03-02 18:02:14,980 INFO    MainThread:262859 [wandb_setup.py:_flush():75] Loading settings from /home/sanchit_huggingface_co/.config/wandb/settings
+2022-03-02 18:02:14,980 INFO    MainThread:262859 [wandb_setup.py:_flush():75] Loading settings from /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/settings
+2022-03-02 18:02:14,980 INFO    MainThread:262859 [wandb_setup.py:_flush():75] Loading settings from environment variables: {}
+2022-03-02 18:02:14,980 INFO    MainThread:262859 [wandb_setup.py:_flush():75] Inferring run settings from compute environment: {'program_relpath': 'run_speech_recognition_seq2seq.py', 'program': '/home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/run_speech_recognition_seq2seq.py'}
+2022-03-02 18:02:14,981 INFO    MainThread:262859 [wandb_init.py:_log_setup():386] Logging user logs to /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/run-20220302_180214-gd4yxtv7/logs/debug.log
+2022-03-02 18:02:14,981 INFO    MainThread:262859 [wandb_init.py:_log_setup():387] Logging internal logs to /home/sanchit_huggingface_co/wav2vec2-gpt2-wandb-grid-search/wandb/run-20220302_180214-gd4yxtv7/logs/debug-internal.log
+2022-03-02 18:02:14,981 INFO    MainThread:262859 [wandb_init.py:init():420] calling init triggers
+2022-03-02 18:02:14,981 INFO    MainThread:262859 [wandb_init.py:init():425] wandb.init called with sweep_config: {}
+config: {}
+2022-03-02 18:02:14,981 INFO    MainThread:262859 [wandb_init.py:init():471] starting backend
+2022-03-02 18:02:14,981 INFO    MainThread:262859 [backend.py:_multiprocessing_setup():99] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
+2022-03-02 18:02:15,034 INFO    MainThread:262859 [backend.py:ensure_launched():219] starting backend process...
+2022-03-02 18:02:15,085 INFO    MainThread:262859 [backend.py:ensure_launched():224] started backend process with pid: 262958
+2022-03-02 18:02:15,087 INFO    MainThread:262859 [wandb_init.py:init():480] backend started and connected
+2022-03-02 18:02:15,097 INFO    MainThread:262859 [wandb_init.py:init():550] updated telemetry
+2022-03-02 18:02:15,219 INFO    MainThread:262859 [wandb_init.py:init():581] communicating current version
+2022-03-02 18:02:15,929 INFO    MainThread:262859 [wandb_init.py:init():586] got version response upgrade_message: "wandb version 0.12.11 is available!  To upgrade, please run:\n $ pip install wandb --upgrade"
+
+2022-03-02 18:02:15,929 INFO    MainThread:262859 [wandb_init.py:init():596] communicating run to backend with 30 second timeout
+2022-03-02 18:02:16,035 INFO    MainThread:262859 [wandb_init.py:init():624] starting run threads in backend
+2022-03-02 18:02:16,186 INFO    MainThread:262859 [wandb_run.py:_console_start():1827] atexit reg
+2022-03-02 18:02:16,186 INFO    MainThread:262859 [wandb_run.py:_redirect():1701] redirect: SettingsConsole.REDIRECT
+2022-03-02 18:02:16,187 INFO    MainThread:262859 [wandb_run.py:_redirect():1706] Redirecting console.
+2022-03-02 18:02:16,188 INFO    MainThread:262859 [wandb_run.py:_redirect():1762] Redirects installed.
+2022-03-02 18:02:16,189 INFO    MainThread:262859 [wandb_init.py:init():651] run started, returning control to user process
+2022-03-02 18:02:16,191 INFO    MainThread:262859 [wandb_run.py:_config_callback():966] config_cb None None {'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'torch.float32', 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': False, 'is_encoder_decoder': True, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 50, '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': ['SpeechEncoderDecoderModel'], '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': None, 'pad_token_id': 50256, 'eos_token_id': 50256, 'sep_token_id': None, 'decoder_start_token_id': 50256, 'task_specific_params': None, 'problem_type': None, '_name_or_path': './', 'transformers_version': None, 'decoder': {'vocab_size': 50257, 'n_positions': 1024, 'n_embd': 1024, 'n_layer': 24, 'n_head': 16, 'n_inner': None, 'activation_function': 'gelu_new', 'resid_pdrop': 0.0, 'embd_pdrop': 0.0, 'attn_pdrop': 0.0, 'layer_norm_epsilon': 1e-05, 'initializer_range': 0.02, 'summary_type': 'cls_index', 'summary_use_proj': True, 'summary_activation': None, 'summary_first_dropout': 0.0, 'summary_proj_to_labels': True, 'scale_attn_weights': True, 'use_cache': False, 'scale_attn_by_inverse_layer_idx': False, 'reorder_and_upcast_attn': False, 'bos_token_id': 50256, 'eos_token_id': 50256, 'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': None, 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'is_encoder_decoder': False, 'is_decoder': True, 'cross_attention_hidden_size': None, 'add_cross_attention': True, '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': ['GPT2LMHeadModel'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'pad_token_id': None, 'sep_token_id': None, 'decoder_start_token_id': None, 'task_specific_params': {'text-generation': {'do_sample': True, 'max_length': 50}}, 'problem_type': None, '_name_or_path': 'gpt2-medium', 'transformers_version': '4.17.0.dev0', 'n_ctx': 1024, 'n_special': 0, 'predict_special_tokens': True, 'model_type': 'gpt2'}, 'encoder': {'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': None, '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': 0, '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-large-lv60', 'transformers_version': '4.17.0.dev0', 'feat_extract_dropout': 0.0, 'gradient_checkpointing': False, 'hidden_dropout_prob': 0.0, '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': 0.0, 'feat_proj_dropout': 0.0, 'final_dropout': 0.0, 'layerdrop': 0.0, 'layer_norm_eps': 1e-05, 'initializer_range': 0.02, 'vocab_size': 32, 'do_stable_layer_norm': True, 'use_weighted_layer_sum': False, 'apply_spec_augment': False, 'mask_time_prob': 0.0, 'mask_time_length': 10, 'mask_time_min_masks': 2, 'mask_feature_prob': 0.0, 'mask_feature_length': 10, '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': 'sum', 'ctc_zero_infinity': False, 'add_adapter': True, '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, 'model_type': 'wav2vec2'}, 'model_type': 'speech-encoder-decoder', 'processor_class': 'Wav2Vec2Processor', 'use_cache': False, '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': 12, 'per_device_eval_batch_size': 12, 'per_gpu_train_batch_size': 'None', 'per_gpu_eval_batch_size': 'None', 'gradient_accumulation_steps': 4, 'eval_accumulation_steps': 'None', 'learning_rate': 0.0003, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'num_train_epochs': 1.0, 'max_steps': -1, 'lr_scheduler_type': 'linear', 'warmup_ratio': 0.0, 'warmup_steps': 500, 'log_level': -1, 'log_level_replica': -1, 'log_on_each_node': True, 'logging_dir': './runs/Mar02_18-01-32_sanchit--v100', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 1, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': 1, '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': False, 'metric_for_best_model': 'None', 'greater_is_better': 'None', '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': '', 'sortish_sampler': False, 'predict_with_generate': True, 'generation_max_length': 40, 'generation_num_beams': 1, 'train_batch_size': 12, 'eval_batch_size': 12}
+2022-03-02 18:02:16,194 INFO    MainThread:262859 [wandb_watch.py:watch():43] Watching

wandb/run-20220302_180214-gd4yxtv7/run-gd4yxtv7.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25885843ba8c9adc78d4b12f06633bf8d1b745fe46c55c7641fc2d834c25738f
+size 31341881