Upload folder using huggingface_hub

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "facebook/wav2vec2-large-xlsr-53",
+  "_name_or_path": "/scratch/leferran/model_computel/Morisyen_xlsr53_mapped_French/checkpoint-12360/",
   "activation_dropout": 0.0,
   "adapter_attn_dim": null,
   "adapter_kernel_size": 3,

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:07c3e5d9397b063848cd8bd133c981e6b296aad7d079a6213d618233825939e7
+oid sha256:741aa870f4811d13bb1eccd5ef2bc457c63d6606f715b0722a5dd54ab30af56b
 size 2524087290

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:71215bfb130595fd02c10e68d26eb247fb66346c4f45006b5b88b702b64e3b68
+oid sha256:ccebe4ddc7b01c1c4bfc46dbf7c96a74d2d6d3a26395fe32d9ff8cc690ab3515
 size 1262066346

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:803af5ced29da46d863ffd8eba565a6aeb0e5bac498eff657677f280fc46dba7
+oid sha256:12d1023535178ebc143e435d35c76e93e0c4c17ea9eb293f6b305741186bb603
 size 14308

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1e2b657ee850cbb8c2a27d8a729fb15acb16559d36f8453bda8818c30cbb9613
+oid sha256:9c907955359884f550451d911a6f12b4ff75e81e328ecf6bfe2e59859437179a
 size 1064

trainer_state.json

@@ -1,415 +1,112 @@
 {
-  "best_metric": 0.34946778711484594,
-  "best_model_checkpoint": "/scratch/leferran/model_computel/Morisyen_xlsr53_mapped_French/checkpoint-11948",
-  "epoch": 29.0,
-  "global_step": 11948,
+  "best_metric": 0.4995602462620932,
+  "best_model_checkpoint": "/scratch/leferran/model_computel/Morisyen_xlsr53_2phase_mapped_French/checkpoint-830",
+  "epoch": 10.0,
+  "global_step": 830,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_loss": 0.715461254119873,
-      "eval_runtime": 14.8409,
-      "eval_samples_per_second": 176.203,
-      "eval_steps_per_second": 22.034,
-      "eval_wer": 0.6281605975723623,
-      "step": 412
-    },
-    {
-      "epoch": 1.21,
-      "learning_rate": 0.00028793689320388344,
-      "loss": 1.6254,
-      "step": 500
+      "eval_loss": 0.7071983218193054,
+      "eval_runtime": 0.6963,
+      "eval_samples_per_second": 231.238,
+      "eval_steps_per_second": 30.162,
+      "eval_wer": 0.6086191732629728,
+      "step": 83
     },
     {
       "epoch": 2.0,
-      "eval_loss": 0.5745924115180969,
-      "eval_runtime": 14.7351,
-      "eval_samples_per_second": 177.467,
-      "eval_steps_per_second": 22.192,
-      "eval_wer": 0.4969561157796452,
-      "step": 824
-    },
-    {
-      "epoch": 2.43,
-      "learning_rate": 0.00027580097087378635,
-      "loss": 0.7008,
-      "step": 1000
+      "eval_loss": 0.6636466383934021,
+      "eval_runtime": 0.6661,
+      "eval_samples_per_second": 241.713,
+      "eval_steps_per_second": 31.528,
+      "eval_wer": 0.5593667546174143,
+      "step": 166
     },
     {
       "epoch": 3.0,
-      "eval_loss": 0.5181881785392761,
-      "eval_runtime": 14.6117,
-      "eval_samples_per_second": 178.966,
-      "eval_steps_per_second": 22.379,
-      "eval_wer": 0.4519514472455649,
-      "step": 1236
-    },
-    {
-      "epoch": 3.64,
-      "learning_rate": 0.0002636650485436893,
-      "loss": 0.5679,
-      "step": 1500
+      "eval_loss": 0.7175659537315369,
+      "eval_runtime": 0.6672,
+      "eval_samples_per_second": 241.298,
+      "eval_steps_per_second": 31.474,
+      "eval_wer": 0.5461741424802111,
+      "step": 249
     },
     {
       "epoch": 4.0,
-      "eval_loss": 0.500670313835144,
-      "eval_runtime": 14.5751,
-      "eval_samples_per_second": 179.415,
-      "eval_steps_per_second": 22.436,
-      "eval_wer": 0.43682539682539684,
-      "step": 1648
-    },
-    {
-      "epoch": 4.85,
-      "learning_rate": 0.0002515291262135922,
-      "loss": 0.4896,
-      "step": 2000
+      "eval_loss": 0.7306005954742432,
+      "eval_runtime": 0.6669,
+      "eval_samples_per_second": 241.429,
+      "eval_steps_per_second": 31.491,
+      "eval_wer": 0.5285839929639402,
+      "step": 332
     },
     {
       "epoch": 5.0,
-      "eval_loss": 0.5165295600891113,
-      "eval_runtime": 14.5743,
-      "eval_samples_per_second": 179.425,
-      "eval_steps_per_second": 22.437,
-      "eval_wer": 0.42603174603174604,
-      "step": 2060
+      "eval_loss": 0.7698836326599121,
+      "eval_runtime": 0.6667,
+      "eval_samples_per_second": 241.506,
+      "eval_steps_per_second": 31.501,
+      "eval_wer": 0.5461741424802111,
+      "step": 415
     },
     {
       "epoch": 6.0,
-      "eval_loss": 0.49643170833587646,
-      "eval_runtime": 14.6008,
-      "eval_samples_per_second": 179.1,
-      "eval_steps_per_second": 22.396,
-      "eval_wer": 0.4053408029878618,
-      "step": 2472
+      "eval_loss": 0.7805858254432678,
+      "eval_runtime": 0.6659,
+      "eval_samples_per_second": 241.763,
+      "eval_steps_per_second": 31.534,
+      "eval_wer": 0.5576077396657871,
+      "step": 498
     },
     {
-      "epoch": 6.07,
-      "learning_rate": 0.00023939320388349515,
-      "loss": 0.4278,
-      "step": 2500
+      "epoch": 6.02,
+      "learning_rate": 0.00023987951807228914,
+      "loss": 0.6806,
+      "step": 500
     },
     {
       "epoch": 7.0,
-      "eval_loss": 0.4752729833126068,
-      "eval_runtime": 14.5968,
-      "eval_samples_per_second": 179.149,
-      "eval_steps_per_second": 22.402,
-      "eval_wer": 0.3961531279178338,
-      "step": 2884
-    },
-    {
-      "epoch": 7.28,
-      "learning_rate": 0.00022725728155339805,
-      "loss": 0.3807,
-      "step": 3000
+      "eval_loss": 0.8358508944511414,
+      "eval_runtime": 0.668,
+      "eval_samples_per_second": 241.027,
+      "eval_steps_per_second": 31.438,
+      "eval_wer": 0.5417766051011433,
+      "step": 581
     },
     {
       "epoch": 8.0,
-      "eval_loss": 0.4915193021297455,
-      "eval_runtime": 14.5679,
-      "eval_samples_per_second": 179.504,
-      "eval_steps_per_second": 22.447,
-      "eval_wer": 0.400186741363212,
-      "step": 3296
-    },
-    {
-      "epoch": 8.5,
-      "learning_rate": 0.00021512135922330096,
-      "loss": 0.357,
-      "step": 3500
+      "eval_loss": 0.9229233860969543,
+      "eval_runtime": 0.6689,
+      "eval_samples_per_second": 240.71,
+      "eval_steps_per_second": 31.397,
+      "eval_wer": 0.5408970976253298,
+      "step": 664
     },
     {
       "epoch": 9.0,
-      "eval_loss": 0.5117114782333374,
-      "eval_runtime": 14.6118,
-      "eval_samples_per_second": 178.965,
-      "eval_steps_per_second": 22.379,
-      "eval_wer": 0.3938001867413632,
-      "step": 3708
-    },
-    {
-      "epoch": 9.71,
-      "learning_rate": 0.00020298543689320386,
-      "loss": 0.3137,
-      "step": 4000
+      "eval_loss": 0.9406667947769165,
+      "eval_runtime": 0.6681,
+      "eval_samples_per_second": 240.977,
+      "eval_steps_per_second": 31.432,
+      "eval_wer": 0.5074758135444152,
+      "step": 747
     },
     {
       "epoch": 10.0,
-      "eval_loss": 0.5042166709899902,
-      "eval_runtime": 14.6066,
-      "eval_samples_per_second": 179.029,
-      "eval_steps_per_second": 22.387,
-      "eval_wer": 0.3908870214752568,
-      "step": 4120
-    },
-    {
-      "epoch": 10.92,
-      "learning_rate": 0.00019087378640776696,
-      "loss": 0.2947,
-      "step": 4500
-    },
-    {
-      "epoch": 11.0,
-      "eval_loss": 0.5115295052528381,
-      "eval_runtime": 14.5991,
-      "eval_samples_per_second": 179.12,
-      "eval_steps_per_second": 22.399,
-      "eval_wer": 0.38110177404295054,
-      "step": 4532
-    },
-    {
-      "epoch": 12.0,
-      "eval_loss": 0.542365550994873,
-      "eval_runtime": 14.6099,
-      "eval_samples_per_second": 178.988,
-      "eval_steps_per_second": 22.382,
-      "eval_wer": 0.3961157796451914,
-      "step": 4944
-    },
-    {
-      "epoch": 12.14,
-      "learning_rate": 0.00017873786407766987,
-      "loss": 0.2738,
-      "step": 5000
-    },
-    {
-      "epoch": 13.0,
-      "eval_loss": 0.5087167024612427,
-      "eval_runtime": 14.5831,
-      "eval_samples_per_second": 179.317,
-      "eval_steps_per_second": 22.423,
-      "eval_wer": 0.3839402427637722,
-      "step": 5356
-    },
-    {
-      "epoch": 13.35,
-      "learning_rate": 0.00016660194174757277,
-      "loss": 0.2473,
-      "step": 5500
-    },
-    {
-      "epoch": 14.0,
-      "eval_loss": 0.5282618999481201,
-      "eval_runtime": 14.6081,
-      "eval_samples_per_second": 179.011,
-      "eval_steps_per_second": 22.385,
-      "eval_wer": 0.3946218487394958,
-      "step": 5768
-    },
-    {
-      "epoch": 14.56,
-      "learning_rate": 0.00015446601941747573,
-      "loss": 0.23,
-      "step": 6000
-    },
-    {
-      "epoch": 15.0,
-      "eval_loss": 0.5473937392234802,
-      "eval_runtime": 14.6354,
-      "eval_samples_per_second": 178.677,
-      "eval_steps_per_second": 22.343,
-      "eval_wer": 0.38595704948646126,
-      "step": 6180
-    },
-    {
-      "epoch": 15.78,
-      "learning_rate": 0.00014233009708737864,
-      "loss": 0.2167,
-      "step": 6500
-    },
-    {
-      "epoch": 16.0,
-      "eval_loss": 0.5802648067474365,
-      "eval_runtime": 14.6253,
-      "eval_samples_per_second": 178.8,
-      "eval_steps_per_second": 22.359,
-      "eval_wer": 0.3760224089635854,
-      "step": 6592
-    },
-    {
-      "epoch": 16.99,
-      "learning_rate": 0.00013019417475728154,
-      "loss": 0.2038,
-      "step": 7000
-    },
-    {
-      "epoch": 17.0,
-      "eval_loss": 0.5503666400909424,
-      "eval_runtime": 14.6287,
-      "eval_samples_per_second": 178.758,
-      "eval_steps_per_second": 22.353,
-      "eval_wer": 0.37628384687208216,
-      "step": 7004
-    },
-    {
-      "epoch": 18.0,
-      "eval_loss": 0.5478885173797607,
-      "eval_runtime": 14.6192,
-      "eval_samples_per_second": 178.874,
-      "eval_steps_per_second": 22.368,
-      "eval_wer": 0.3692623716153128,
-      "step": 7416
-    },
-    {
-      "epoch": 18.2,
-      "learning_rate": 0.00011805825242718445,
-      "loss": 0.1791,
-      "step": 7500
-    },
-    {
-      "epoch": 19.0,
-      "eval_loss": 0.5444658994674683,
-      "eval_runtime": 14.6206,
-      "eval_samples_per_second": 178.857,
-      "eval_steps_per_second": 22.366,
-      "eval_wer": 0.3685901027077498,
-      "step": 7828
-    },
-    {
-      "epoch": 19.42,
-      "learning_rate": 0.00010594660194174756,
-      "loss": 0.1696,
-      "step": 8000
-    },
-    {
-      "epoch": 20.0,
-      "eval_loss": 0.5748639702796936,
-      "eval_runtime": 14.6234,
-      "eval_samples_per_second": 178.823,
-      "eval_steps_per_second": 22.361,
-      "eval_wer": 0.36952380952380953,
-      "step": 8240
-    },
-    {
-      "epoch": 20.63,
-      "learning_rate": 9.381067961165047e-05,
-      "loss": 0.1623,
-      "step": 8500
-    },
-    {
-      "epoch": 21.0,
-      "eval_loss": 0.6076938509941101,
-      "eval_runtime": 14.6105,
-      "eval_samples_per_second": 178.981,
-      "eval_steps_per_second": 22.381,
-      "eval_wer": 0.36593837535014007,
-      "step": 8652
-    },
-    {
-      "epoch": 21.84,
-      "learning_rate": 8.16747572815534e-05,
-      "loss": 0.1497,
-      "step": 9000
-    },
-    {
-      "epoch": 22.0,
-      "eval_loss": 0.6302914023399353,
-      "eval_runtime": 14.5946,
-      "eval_samples_per_second": 179.176,
-      "eval_steps_per_second": 22.406,
-      "eval_wer": 0.3715032679738562,
-      "step": 9064
-    },
-    {
-      "epoch": 23.0,
-      "eval_loss": 0.5975435376167297,
-      "eval_runtime": 14.6019,
-      "eval_samples_per_second": 179.086,
-      "eval_steps_per_second": 22.394,
-      "eval_wer": 0.3733706816059757,
-      "step": 9476
-    },
-    {
-      "epoch": 23.06,
-      "learning_rate": 6.95388349514563e-05,
-      "loss": 0.1357,
-      "step": 9500
-    },
-    {
-      "epoch": 24.0,
-      "eval_loss": 0.6168835163116455,
-      "eval_runtime": 14.6062,
-      "eval_samples_per_second": 179.033,
-      "eval_steps_per_second": 22.388,
-      "eval_wer": 0.3604855275443511,
-      "step": 9888
-    },
-    {
-      "epoch": 24.27,
-      "learning_rate": 5.740291262135922e-05,
-      "loss": 0.1252,
-      "step": 10000
-    },
-    {
-      "epoch": 25.0,
-      "eval_loss": 0.6086588501930237,
-      "eval_runtime": 14.6378,
-      "eval_samples_per_second": 178.647,
-      "eval_steps_per_second": 22.339,
-      "eval_wer": 0.3648926237161531,
-      "step": 10300
-    },
-    {
-      "epoch": 25.49,
-      "learning_rate": 4.526699029126213e-05,
-      "loss": 0.119,
-      "step": 10500
-    },
-    {
-      "epoch": 26.0,
-      "eval_loss": 0.6258659958839417,
-      "eval_runtime": 14.6218,
-      "eval_samples_per_second": 178.842,
-      "eval_steps_per_second": 22.364,
-      "eval_wer": 0.3592530345471522,
-      "step": 10712
-    },
-    {
-      "epoch": 26.7,
-      "learning_rate": 3.315533980582524e-05,
-      "loss": 0.1039,
-      "step": 11000
-    },
-    {
-      "epoch": 27.0,
-      "eval_loss": 0.6336789131164551,
-      "eval_runtime": 14.62,
-      "eval_samples_per_second": 178.865,
-      "eval_steps_per_second": 22.367,
-      "eval_wer": 0.3522315592903828,
-      "step": 11124
-    },
-    {
-      "epoch": 27.91,
-      "learning_rate": 2.1019417475728154e-05,
-      "loss": 0.1029,
-      "step": 11500
-    },
-    {
-      "epoch": 28.0,
-      "eval_loss": 0.6366888284683228,
-      "eval_runtime": 14.6461,
-      "eval_samples_per_second": 178.546,
-      "eval_steps_per_second": 22.327,
-      "eval_wer": 0.3557796451914099,
-      "step": 11536
-    },
-    {
-      "epoch": 29.0,
-      "eval_loss": 0.6363462805747986,
-      "eval_runtime": 14.6045,
-      "eval_samples_per_second": 179.054,
-      "eval_steps_per_second": 22.39,
-      "eval_wer": 0.34946778711484594,
-      "step": 11948
+      "eval_loss": 0.9393610954284668,
+      "eval_runtime": 0.6679,
+      "eval_samples_per_second": 241.051,
+      "eval_steps_per_second": 31.441,
+      "eval_wer": 0.4995602462620932,
+      "step": 830
     }
   ],
-  "max_steps": 12360,
+  "max_steps": 2490,
   "num_train_epochs": 30,
-  "total_flos": 3.2239711217706242e+19,
+  "total_flos": 3.21639689183166e+18,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2bb88c1740f3e2f2f1dff31dba9a1a639267659428b010388050b2f0de9d1c80
-size 4472
+oid sha256:08c8d98a422b9433b5d4287d36d02da44d12d91f134185824038634ab95902fe
+size 4536