Training in progress, step 4000, checkpoint
Browse files
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 966995080
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b810b7f0fe83995300521a646137d370d36d549e50472fcf6c1da2a177919ec3
|
| 3 |
size 966995080
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1925064044
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3a70bf563a60fc19871fd719056a40461333af72260151f6bade95b428cbc93c
|
| 3 |
size 1925064044
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a7c93a397e9322e49f4ed50d18f810eaf2c39ecdb2985c95d248cd7a2fa2aa47
|
| 3 |
size 14244
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fd54311344b834087a4b1c20d06544579c7f43d33908960b6b3b61734dbde46d
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint": "./whisper-small-taiwanese/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 1000,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -874,6 +874,295 @@
|
|
| 874 |
"eval_samples_per_second": 2.385,
|
| 875 |
"eval_steps_per_second": 0.299,
|
| 876 |
"step": 3000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 877 |
}
|
| 878 |
],
|
| 879 |
"logging_steps": 25,
|
|
@@ -881,7 +1170,7 @@
|
|
| 881 |
"num_input_tokens_seen": 0,
|
| 882 |
"num_train_epochs": 3,
|
| 883 |
"save_steps": 1000,
|
| 884 |
-
"total_flos": 1.
|
| 885 |
"train_batch_size": 16,
|
| 886 |
"trial_name": null,
|
| 887 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 29.080310880829014,
|
| 3 |
+
"best_model_checkpoint": "./whisper-small-taiwanese/checkpoint-4000",
|
| 4 |
+
"epoch": 2.5806451612903225,
|
| 5 |
"eval_steps": 1000,
|
| 6 |
+
"global_step": 4000,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 874 |
"eval_samples_per_second": 2.385,
|
| 875 |
"eval_steps_per_second": 0.299,
|
| 876 |
"step": 3000
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 1.9516129032258065,
|
| 880 |
+
"grad_norm": 9.115569114685059,
|
| 881 |
+
"learning_rate": 3.882915173237754e-06,
|
| 882 |
+
"loss": 0.2929,
|
| 883 |
+
"step": 3025
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 1.967741935483871,
|
| 887 |
+
"grad_norm": 10.029779434204102,
|
| 888 |
+
"learning_rate": 3.823178016726404e-06,
|
| 889 |
+
"loss": 0.2939,
|
| 890 |
+
"step": 3050
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 1.9838709677419355,
|
| 894 |
+
"grad_norm": 10.219369888305664,
|
| 895 |
+
"learning_rate": 3.763440860215054e-06,
|
| 896 |
+
"loss": 0.3228,
|
| 897 |
+
"step": 3075
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 2.0,
|
| 901 |
+
"grad_norm": 10.639910697937012,
|
| 902 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 903 |
+
"loss": 0.3236,
|
| 904 |
+
"step": 3100
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 2.0161290322580645,
|
| 908 |
+
"grad_norm": 4.525390148162842,
|
| 909 |
+
"learning_rate": 3.643966547192354e-06,
|
| 910 |
+
"loss": 0.1713,
|
| 911 |
+
"step": 3125
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 2.032258064516129,
|
| 915 |
+
"grad_norm": 6.762115478515625,
|
| 916 |
+
"learning_rate": 3.584229390681004e-06,
|
| 917 |
+
"loss": 0.1892,
|
| 918 |
+
"step": 3150
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 2.0483870967741935,
|
| 922 |
+
"grad_norm": 8.670353889465332,
|
| 923 |
+
"learning_rate": 3.5244922341696534e-06,
|
| 924 |
+
"loss": 0.1902,
|
| 925 |
+
"step": 3175
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 2.064516129032258,
|
| 929 |
+
"grad_norm": 6.066471576690674,
|
| 930 |
+
"learning_rate": 3.4647550776583037e-06,
|
| 931 |
+
"loss": 0.157,
|
| 932 |
+
"step": 3200
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 2.0806451612903225,
|
| 936 |
+
"grad_norm": 5.877708911895752,
|
| 937 |
+
"learning_rate": 3.4050179211469536e-06,
|
| 938 |
+
"loss": 0.18,
|
| 939 |
+
"step": 3225
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 2.096774193548387,
|
| 943 |
+
"grad_norm": 5.906160354614258,
|
| 944 |
+
"learning_rate": 3.3452807646356034e-06,
|
| 945 |
+
"loss": 0.198,
|
| 946 |
+
"step": 3250
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 2.1129032258064515,
|
| 950 |
+
"grad_norm": 7.664149761199951,
|
| 951 |
+
"learning_rate": 3.2855436081242537e-06,
|
| 952 |
+
"loss": 0.1889,
|
| 953 |
+
"step": 3275
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 2.129032258064516,
|
| 957 |
+
"grad_norm": 6.261497497558594,
|
| 958 |
+
"learning_rate": 3.225806451612903e-06,
|
| 959 |
+
"loss": 0.157,
|
| 960 |
+
"step": 3300
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 2.1451612903225805,
|
| 964 |
+
"grad_norm": 6.205556869506836,
|
| 965 |
+
"learning_rate": 3.1660692951015535e-06,
|
| 966 |
+
"loss": 0.1635,
|
| 967 |
+
"step": 3325
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 2.161290322580645,
|
| 971 |
+
"grad_norm": 5.549154758453369,
|
| 972 |
+
"learning_rate": 3.1063321385902034e-06,
|
| 973 |
+
"loss": 0.1799,
|
| 974 |
+
"step": 3350
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 2.1774193548387095,
|
| 978 |
+
"grad_norm": 7.939329624176025,
|
| 979 |
+
"learning_rate": 3.0465949820788532e-06,
|
| 980 |
+
"loss": 0.159,
|
| 981 |
+
"step": 3375
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 2.193548387096774,
|
| 985 |
+
"grad_norm": 7.168279647827148,
|
| 986 |
+
"learning_rate": 2.9868578255675035e-06,
|
| 987 |
+
"loss": 0.1772,
|
| 988 |
+
"step": 3400
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 2.2096774193548385,
|
| 992 |
+
"grad_norm": 7.339049816131592,
|
| 993 |
+
"learning_rate": 2.9271206690561534e-06,
|
| 994 |
+
"loss": 0.161,
|
| 995 |
+
"step": 3425
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 2.225806451612903,
|
| 999 |
+
"grad_norm": 7.939656734466553,
|
| 1000 |
+
"learning_rate": 2.867383512544803e-06,
|
| 1001 |
+
"loss": 0.163,
|
| 1002 |
+
"step": 3450
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 2.241935483870968,
|
| 1006 |
+
"grad_norm": 6.338183403015137,
|
| 1007 |
+
"learning_rate": 2.807646356033453e-06,
|
| 1008 |
+
"loss": 0.1671,
|
| 1009 |
+
"step": 3475
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 2.258064516129032,
|
| 1013 |
+
"grad_norm": 4.27256441116333,
|
| 1014 |
+
"learning_rate": 2.747909199522103e-06,
|
| 1015 |
+
"loss": 0.1696,
|
| 1016 |
+
"step": 3500
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 2.274193548387097,
|
| 1020 |
+
"grad_norm": 7.593271255493164,
|
| 1021 |
+
"learning_rate": 2.688172043010753e-06,
|
| 1022 |
+
"loss": 0.1558,
|
| 1023 |
+
"step": 3525
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 2.2903225806451615,
|
| 1027 |
+
"grad_norm": 3.984323501586914,
|
| 1028 |
+
"learning_rate": 2.6284348864994032e-06,
|
| 1029 |
+
"loss": 0.1727,
|
| 1030 |
+
"step": 3550
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 2.306451612903226,
|
| 1034 |
+
"grad_norm": 6.01830530166626,
|
| 1035 |
+
"learning_rate": 2.5686977299880527e-06,
|
| 1036 |
+
"loss": 0.1681,
|
| 1037 |
+
"step": 3575
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 2.3225806451612905,
|
| 1041 |
+
"grad_norm": 8.704211235046387,
|
| 1042 |
+
"learning_rate": 2.5089605734767026e-06,
|
| 1043 |
+
"loss": 0.1703,
|
| 1044 |
+
"step": 3600
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 2.338709677419355,
|
| 1048 |
+
"grad_norm": 7.5924506187438965,
|
| 1049 |
+
"learning_rate": 2.4492234169653525e-06,
|
| 1050 |
+
"loss": 0.1723,
|
| 1051 |
+
"step": 3625
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 2.3548387096774195,
|
| 1055 |
+
"grad_norm": 5.732957363128662,
|
| 1056 |
+
"learning_rate": 2.3894862604540028e-06,
|
| 1057 |
+
"loss": 0.1586,
|
| 1058 |
+
"step": 3650
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 2.370967741935484,
|
| 1062 |
+
"grad_norm": 7.640561580657959,
|
| 1063 |
+
"learning_rate": 2.3297491039426526e-06,
|
| 1064 |
+
"loss": 0.1781,
|
| 1065 |
+
"step": 3675
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 2.3870967741935485,
|
| 1069 |
+
"grad_norm": 7.3015313148498535,
|
| 1070 |
+
"learning_rate": 2.2700119474313025e-06,
|
| 1071 |
+
"loss": 0.1571,
|
| 1072 |
+
"step": 3700
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 2.403225806451613,
|
| 1076 |
+
"grad_norm": 7.6669440269470215,
|
| 1077 |
+
"learning_rate": 2.2102747909199524e-06,
|
| 1078 |
+
"loss": 0.1532,
|
| 1079 |
+
"step": 3725
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 2.4193548387096775,
|
| 1083 |
+
"grad_norm": 4.207671165466309,
|
| 1084 |
+
"learning_rate": 2.1505376344086023e-06,
|
| 1085 |
+
"loss": 0.1658,
|
| 1086 |
+
"step": 3750
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 2.435483870967742,
|
| 1090 |
+
"grad_norm": 6.316219806671143,
|
| 1091 |
+
"learning_rate": 2.0908004778972526e-06,
|
| 1092 |
+
"loss": 0.1681,
|
| 1093 |
+
"step": 3775
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 2.4516129032258065,
|
| 1097 |
+
"grad_norm": 6.380753040313721,
|
| 1098 |
+
"learning_rate": 2.031063321385902e-06,
|
| 1099 |
+
"loss": 0.1595,
|
| 1100 |
+
"step": 3800
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 2.467741935483871,
|
| 1104 |
+
"grad_norm": 7.425994873046875,
|
| 1105 |
+
"learning_rate": 1.9713261648745523e-06,
|
| 1106 |
+
"loss": 0.1457,
|
| 1107 |
+
"step": 3825
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 2.4838709677419355,
|
| 1111 |
+
"grad_norm": 7.0207600593566895,
|
| 1112 |
+
"learning_rate": 1.911589008363202e-06,
|
| 1113 |
+
"loss": 0.1557,
|
| 1114 |
+
"step": 3850
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 2.5,
|
| 1118 |
+
"grad_norm": 7.421679496765137,
|
| 1119 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 1120 |
+
"loss": 0.17,
|
| 1121 |
+
"step": 3875
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 2.5161290322580645,
|
| 1125 |
+
"grad_norm": 7.43884801864624,
|
| 1126 |
+
"learning_rate": 1.792114695340502e-06,
|
| 1127 |
+
"loss": 0.1605,
|
| 1128 |
+
"step": 3900
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 2.532258064516129,
|
| 1132 |
+
"grad_norm": 6.753660202026367,
|
| 1133 |
+
"learning_rate": 1.7323775388291518e-06,
|
| 1134 |
+
"loss": 0.1467,
|
| 1135 |
+
"step": 3925
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 2.5483870967741935,
|
| 1139 |
+
"grad_norm": 5.872158050537109,
|
| 1140 |
+
"learning_rate": 1.6726403823178017e-06,
|
| 1141 |
+
"loss": 0.1511,
|
| 1142 |
+
"step": 3950
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 2.564516129032258,
|
| 1146 |
+
"grad_norm": 5.571343421936035,
|
| 1147 |
+
"learning_rate": 1.6129032258064516e-06,
|
| 1148 |
+
"loss": 0.1476,
|
| 1149 |
+
"step": 3975
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 2.5806451612903225,
|
| 1153 |
+
"grad_norm": 9.308771133422852,
|
| 1154 |
+
"learning_rate": 1.5531660692951017e-06,
|
| 1155 |
+
"loss": 0.1566,
|
| 1156 |
+
"step": 4000
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 2.5806451612903225,
|
| 1160 |
+
"eval_cer": 29.080310880829014,
|
| 1161 |
+
"eval_loss": 0.37441545724868774,
|
| 1162 |
+
"eval_runtime": 959.8469,
|
| 1163 |
+
"eval_samples_per_second": 2.376,
|
| 1164 |
+
"eval_steps_per_second": 0.298,
|
| 1165 |
+
"step": 4000
|
| 1166 |
}
|
| 1167 |
],
|
| 1168 |
"logging_steps": 25,
|
|
|
|
| 1170 |
"num_input_tokens_seen": 0,
|
| 1171 |
"num_train_epochs": 3,
|
| 1172 |
"save_steps": 1000,
|
| 1173 |
+
"total_flos": 1.84665797664768e+19,
|
| 1174 |
"train_batch_size": 16,
|
| 1175 |
"trial_name": null,
|
| 1176 |
"trial_params": null
|