Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
Safetensors
whisper
Generated from Trainer
Instructions to use byoussef/whisper-large-v2-AiHub with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use byoussef/whisper-large-v2-AiHub with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="byoussef/whisper-large-v2-AiHub")# pip install -U transformers accelerate # Load model directly from transformers import AutoProcessor, AutoModelForSpeechSeq2Seq processor = AutoProcessor.from_pretrained("byoussef/whisper-large-v2-AiHub") model = AutoModelForSpeechSeq2Seq.from_pretrained("byoussef/whisper-large-v2-AiHub", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download trainer_state.json from byoussef/whisper-large-v2-AiHub: direct link, hf CLI and curl.
- Browser
- Download file 25.8 kB
-
https://huggingface.co/byoussef/whisper-large-v2-AiHub/resolve/main/trainer_state.json
- Command line
-
hf download hf://byoussef/whisper-large-v2-AiHub/trainer_state.json
-
curl -L -o trainer_state.json https://huggingface.co/byoussef/whisper-large-v2-AiHub/resolve/main/trainer_state.json
25.8 kB
{
"best_metric": 0.22299687523689737,
"best_model_checkpoint": "./whisper-large-v2-AiHub/checkpoint-5000",
"epoch": 0.502563071665494,
"global_step": 5000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 4.6000000000000004e-07,
"loss": 1.0382,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 9.600000000000001e-07,
"loss": 0.8725,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 1.46e-06,
"loss": 0.5618,
"step": 75
},
{
"epoch": 0.01,
"learning_rate": 1.9600000000000003e-06,
"loss": 0.4216,
"step": 100
},
{
"epoch": 0.01,
"learning_rate": 2.46e-06,
"loss": 0.3748,
"step": 125
},
{
"epoch": 0.02,
"learning_rate": 2.96e-06,
"loss": 0.3507,
"step": 150
},
{
"epoch": 0.02,
"learning_rate": 3.44e-06,
"loss": 0.3454,
"step": 175
},
{
"epoch": 0.02,
"learning_rate": 3.94e-06,
"loss": 0.3512,
"step": 200
},
{
"epoch": 0.02,
"learning_rate": 4.440000000000001e-06,
"loss": 0.3427,
"step": 225
},
{
"epoch": 0.03,
"learning_rate": 4.92e-06,
"loss": 0.3455,
"step": 250
},
{
"epoch": 0.03,
"learning_rate": 5.420000000000001e-06,
"loss": 0.3427,
"step": 275
},
{
"epoch": 0.03,
"learning_rate": 5.9e-06,
"loss": 0.3385,
"step": 300
},
{
"epoch": 0.03,
"learning_rate": 6.4000000000000006e-06,
"loss": 0.3348,
"step": 325
},
{
"epoch": 0.04,
"learning_rate": 6.9e-06,
"loss": 0.3412,
"step": 350
},
{
"epoch": 0.04,
"learning_rate": 7.4e-06,
"loss": 0.3366,
"step": 375
},
{
"epoch": 0.04,
"learning_rate": 7.9e-06,
"loss": 0.3444,
"step": 400
},
{
"epoch": 0.04,
"learning_rate": 8.400000000000001e-06,
"loss": 0.342,
"step": 425
},
{
"epoch": 0.05,
"learning_rate": 8.900000000000001e-06,
"loss": 0.3478,
"step": 450
},
{
"epoch": 0.05,
"learning_rate": 9.4e-06,
"loss": 0.3419,
"step": 475
},
{
"epoch": 0.05,
"learning_rate": 9.9e-06,
"loss": 0.3477,
"step": 500
},
{
"epoch": 0.05,
"learning_rate": 9.955555555555556e-06,
"loss": 0.3503,
"step": 525
},
{
"epoch": 0.06,
"learning_rate": 9.9e-06,
"loss": 0.3502,
"step": 550
},
{
"epoch": 0.06,
"learning_rate": 9.844444444444446e-06,
"loss": 0.3456,
"step": 575
},
{
"epoch": 0.06,
"learning_rate": 9.78888888888889e-06,
"loss": 0.3414,
"step": 600
},
{
"epoch": 0.06,
"learning_rate": 9.733333333333334e-06,
"loss": 0.3462,
"step": 625
},
{
"epoch": 0.07,
"learning_rate": 9.677777777777778e-06,
"loss": 0.3479,
"step": 650
},
{
"epoch": 0.07,
"learning_rate": 9.622222222222222e-06,
"loss": 0.3387,
"step": 675
},
{
"epoch": 0.07,
"learning_rate": 9.566666666666668e-06,
"loss": 0.3422,
"step": 700
},
{
"epoch": 0.07,
"learning_rate": 9.511111111111112e-06,
"loss": 0.3349,
"step": 725
},
{
"epoch": 0.08,
"learning_rate": 9.455555555555557e-06,
"loss": 0.3361,
"step": 750
},
{
"epoch": 0.08,
"learning_rate": 9.4e-06,
"loss": 0.3475,
"step": 775
},
{
"epoch": 0.08,
"learning_rate": 9.346666666666666e-06,
"loss": 0.3388,
"step": 800
},
{
"epoch": 0.08,
"learning_rate": 9.291111111111112e-06,
"loss": 0.3324,
"step": 825
},
{
"epoch": 0.09,
"learning_rate": 9.235555555555556e-06,
"loss": 0.3349,
"step": 850
},
{
"epoch": 0.09,
"learning_rate": 9.180000000000002e-06,
"loss": 0.3419,
"step": 875
},
{
"epoch": 0.09,
"learning_rate": 9.124444444444444e-06,
"loss": 0.3369,
"step": 900
},
{
"epoch": 0.09,
"learning_rate": 9.06888888888889e-06,
"loss": 0.3454,
"step": 925
},
{
"epoch": 0.1,
"learning_rate": 9.013333333333334e-06,
"loss": 0.3379,
"step": 950
},
{
"epoch": 0.1,
"learning_rate": 8.957777777777778e-06,
"loss": 0.3353,
"step": 975
},
{
"epoch": 0.1,
"learning_rate": 8.902222222222224e-06,
"loss": 0.3373,
"step": 1000
},
{
"epoch": 0.1,
"eval_loss": 0.31461259722709656,
"eval_runtime": 11916.9934,
"eval_samples_per_second": 16.681,
"eval_steps_per_second": 0.075,
"eval_wer": 0.25069307983843153,
"step": 1000
},
{
"epoch": 0.1,
"learning_rate": 8.846666666666668e-06,
"loss": 0.3292,
"step": 1025
},
{
"epoch": 0.11,
"learning_rate": 8.791111111111112e-06,
"loss": 0.3303,
"step": 1050
},
{
"epoch": 0.11,
"learning_rate": 8.735555555555556e-06,
"loss": 0.3272,
"step": 1075
},
{
"epoch": 0.11,
"learning_rate": 8.68e-06,
"loss": 0.3329,
"step": 1100
},
{
"epoch": 0.11,
"learning_rate": 8.624444444444446e-06,
"loss": 0.3249,
"step": 1125
},
{
"epoch": 0.12,
"learning_rate": 8.56888888888889e-06,
"loss": 0.3333,
"step": 1150
},
{
"epoch": 0.12,
"learning_rate": 8.513333333333335e-06,
"loss": 0.3322,
"step": 1175
},
{
"epoch": 0.12,
"learning_rate": 8.457777777777778e-06,
"loss": 0.3323,
"step": 1200
},
{
"epoch": 0.12,
"learning_rate": 8.402222222222223e-06,
"loss": 0.3223,
"step": 1225
},
{
"epoch": 0.13,
"learning_rate": 8.346666666666668e-06,
"loss": 0.3291,
"step": 1250
},
{
"epoch": 0.13,
"learning_rate": 8.291111111111112e-06,
"loss": 0.3312,
"step": 1275
},
{
"epoch": 0.13,
"learning_rate": 8.235555555555557e-06,
"loss": 0.3263,
"step": 1300
},
{
"epoch": 0.13,
"learning_rate": 8.18e-06,
"loss": 0.3241,
"step": 1325
},
{
"epoch": 0.14,
"learning_rate": 8.124444444444445e-06,
"loss": 0.3298,
"step": 1350
},
{
"epoch": 0.14,
"learning_rate": 8.06888888888889e-06,
"loss": 0.3253,
"step": 1375
},
{
"epoch": 0.14,
"learning_rate": 8.013333333333333e-06,
"loss": 0.3238,
"step": 1400
},
{
"epoch": 0.14,
"learning_rate": 7.957777777777779e-06,
"loss": 0.328,
"step": 1425
},
{
"epoch": 0.15,
"learning_rate": 7.902222222222223e-06,
"loss": 0.3269,
"step": 1450
},
{
"epoch": 0.15,
"learning_rate": 7.846666666666667e-06,
"loss": 0.3215,
"step": 1475
},
{
"epoch": 0.15,
"learning_rate": 7.791111111111111e-06,
"loss": 0.323,
"step": 1500
},
{
"epoch": 0.15,
"learning_rate": 7.735555555555557e-06,
"loss": 0.3231,
"step": 1525
},
{
"epoch": 0.16,
"learning_rate": 7.680000000000001e-06,
"loss": 0.3259,
"step": 1550
},
{
"epoch": 0.16,
"learning_rate": 7.624444444444445e-06,
"loss": 0.3199,
"step": 1575
},
{
"epoch": 0.16,
"learning_rate": 7.56888888888889e-06,
"loss": 0.3153,
"step": 1600
},
{
"epoch": 0.16,
"learning_rate": 7.513333333333334e-06,
"loss": 0.321,
"step": 1625
},
{
"epoch": 0.17,
"learning_rate": 7.457777777777778e-06,
"loss": 0.3242,
"step": 1650
},
{
"epoch": 0.17,
"learning_rate": 7.402222222222223e-06,
"loss": 0.3197,
"step": 1675
},
{
"epoch": 0.17,
"learning_rate": 7.346666666666668e-06,
"loss": 0.3137,
"step": 1700
},
{
"epoch": 0.17,
"learning_rate": 7.291111111111112e-06,
"loss": 0.3164,
"step": 1725
},
{
"epoch": 0.18,
"learning_rate": 7.235555555555556e-06,
"loss": 0.3215,
"step": 1750
},
{
"epoch": 0.18,
"learning_rate": 7.180000000000001e-06,
"loss": 0.321,
"step": 1775
},
{
"epoch": 0.18,
"learning_rate": 7.124444444444445e-06,
"loss": 0.318,
"step": 1800
},
{
"epoch": 0.18,
"learning_rate": 7.06888888888889e-06,
"loss": 0.3184,
"step": 1825
},
{
"epoch": 0.19,
"learning_rate": 7.0133333333333345e-06,
"loss": 0.3027,
"step": 1850
},
{
"epoch": 0.19,
"learning_rate": 6.9577777777777785e-06,
"loss": 0.3107,
"step": 1875
},
{
"epoch": 0.19,
"learning_rate": 6.902222222222223e-06,
"loss": 0.3203,
"step": 1900
},
{
"epoch": 0.19,
"learning_rate": 6.846666666666667e-06,
"loss": 0.3173,
"step": 1925
},
{
"epoch": 0.2,
"learning_rate": 6.7911111111111115e-06,
"loss": 0.3171,
"step": 1950
},
{
"epoch": 0.2,
"learning_rate": 6.735555555555556e-06,
"loss": 0.3114,
"step": 1975
},
{
"epoch": 0.2,
"learning_rate": 6.680000000000001e-06,
"loss": 0.3131,
"step": 2000
},
{
"epoch": 0.2,
"eval_loss": 0.29671308398246765,
"eval_runtime": 11814.166,
"eval_samples_per_second": 16.827,
"eval_steps_per_second": 0.075,
"eval_wer": 0.24084092760972095,
"step": 2000
},
{
"epoch": 0.2,
"learning_rate": 6.6244444444444445e-06,
"loss": 0.318,
"step": 2025
},
{
"epoch": 0.21,
"learning_rate": 6.568888888888889e-06,
"loss": 0.3161,
"step": 2050
},
{
"epoch": 0.21,
"learning_rate": 6.513333333333333e-06,
"loss": 0.3103,
"step": 2075
},
{
"epoch": 0.21,
"learning_rate": 6.457777777777778e-06,
"loss": 0.3102,
"step": 2100
},
{
"epoch": 0.21,
"learning_rate": 6.404444444444446e-06,
"loss": 0.3102,
"step": 2125
},
{
"epoch": 0.22,
"learning_rate": 6.348888888888889e-06,
"loss": 0.3128,
"step": 2150
},
{
"epoch": 0.22,
"learning_rate": 6.293333333333334e-06,
"loss": 0.3093,
"step": 2175
},
{
"epoch": 0.22,
"learning_rate": 6.237777777777778e-06,
"loss": 0.3081,
"step": 2200
},
{
"epoch": 0.22,
"learning_rate": 6.182222222222223e-06,
"loss": 0.3078,
"step": 2225
},
{
"epoch": 0.23,
"learning_rate": 6.126666666666668e-06,
"loss": 0.3069,
"step": 2250
},
{
"epoch": 0.23,
"learning_rate": 6.0711111111111125e-06,
"loss": 0.3115,
"step": 2275
},
{
"epoch": 0.23,
"learning_rate": 6.015555555555556e-06,
"loss": 0.3106,
"step": 2300
},
{
"epoch": 0.23,
"learning_rate": 5.9600000000000005e-06,
"loss": 0.312,
"step": 2325
},
{
"epoch": 0.24,
"learning_rate": 5.9044444444444446e-06,
"loss": 0.3076,
"step": 2350
},
{
"epoch": 0.24,
"learning_rate": 5.8488888888888895e-06,
"loss": 0.3104,
"step": 2375
},
{
"epoch": 0.24,
"learning_rate": 5.793333333333334e-06,
"loss": 0.3069,
"step": 2400
},
{
"epoch": 0.24,
"learning_rate": 5.737777777777778e-06,
"loss": 0.3097,
"step": 2425
},
{
"epoch": 0.25,
"learning_rate": 5.682222222222222e-06,
"loss": 0.3068,
"step": 2450
},
{
"epoch": 0.25,
"learning_rate": 5.626666666666667e-06,
"loss": 0.3067,
"step": 2475
},
{
"epoch": 0.25,
"learning_rate": 5.571111111111111e-06,
"loss": 0.3023,
"step": 2500
},
{
"epoch": 0.25,
"learning_rate": 5.515555555555556e-06,
"loss": 0.3019,
"step": 2525
},
{
"epoch": 0.26,
"learning_rate": 5.460000000000001e-06,
"loss": 0.304,
"step": 2550
},
{
"epoch": 0.26,
"learning_rate": 5.404444444444444e-06,
"loss": 0.3061,
"step": 2575
},
{
"epoch": 0.26,
"learning_rate": 5.348888888888889e-06,
"loss": 0.2949,
"step": 2600
},
{
"epoch": 0.26,
"learning_rate": 5.293333333333334e-06,
"loss": 0.3023,
"step": 2625
},
{
"epoch": 0.27,
"learning_rate": 5.237777777777778e-06,
"loss": 0.3057,
"step": 2650
},
{
"epoch": 0.27,
"learning_rate": 5.182222222222223e-06,
"loss": 0.2998,
"step": 2675
},
{
"epoch": 0.27,
"learning_rate": 5.126666666666668e-06,
"loss": 0.3003,
"step": 2700
},
{
"epoch": 0.27,
"learning_rate": 5.071111111111111e-06,
"loss": 0.3031,
"step": 2725
},
{
"epoch": 0.28,
"learning_rate": 5.015555555555556e-06,
"loss": 0.3014,
"step": 2750
},
{
"epoch": 0.28,
"learning_rate": 4.960000000000001e-06,
"loss": 0.3064,
"step": 2775
},
{
"epoch": 0.28,
"learning_rate": 4.904444444444445e-06,
"loss": 0.2969,
"step": 2800
},
{
"epoch": 0.28,
"learning_rate": 4.848888888888889e-06,
"loss": 0.2996,
"step": 2825
},
{
"epoch": 0.29,
"learning_rate": 4.793333333333334e-06,
"loss": 0.3013,
"step": 2850
},
{
"epoch": 0.29,
"learning_rate": 4.737777777777779e-06,
"loss": 0.3061,
"step": 2875
},
{
"epoch": 0.29,
"learning_rate": 4.682222222222223e-06,
"loss": 0.2968,
"step": 2900
},
{
"epoch": 0.29,
"learning_rate": 4.626666666666667e-06,
"loss": 0.3045,
"step": 2925
},
{
"epoch": 0.3,
"learning_rate": 4.571111111111112e-06,
"loss": 0.2978,
"step": 2950
},
{
"epoch": 0.3,
"learning_rate": 4.515555555555556e-06,
"loss": 0.3003,
"step": 2975
},
{
"epoch": 0.3,
"learning_rate": 4.4600000000000005e-06,
"loss": 0.3053,
"step": 3000
},
{
"epoch": 0.3,
"eval_loss": 0.2815037965774536,
"eval_runtime": 11830.8408,
"eval_samples_per_second": 16.803,
"eval_steps_per_second": 0.075,
"eval_wer": 0.23461362758682003,
"step": 3000
},
{
"epoch": 0.3,
"learning_rate": 4.404444444444445e-06,
"loss": 0.2924,
"step": 3025
},
{
"epoch": 0.31,
"learning_rate": 4.348888888888889e-06,
"loss": 0.3026,
"step": 3050
},
{
"epoch": 0.31,
"learning_rate": 4.2933333333333334e-06,
"loss": 0.2934,
"step": 3075
},
{
"epoch": 0.31,
"learning_rate": 4.2377777777777775e-06,
"loss": 0.2992,
"step": 3100
},
{
"epoch": 0.31,
"learning_rate": 4.182222222222222e-06,
"loss": 0.291,
"step": 3125
},
{
"epoch": 0.32,
"learning_rate": 4.126666666666667e-06,
"loss": 0.2972,
"step": 3150
},
{
"epoch": 0.32,
"learning_rate": 4.071111111111111e-06,
"loss": 0.2916,
"step": 3175
},
{
"epoch": 0.32,
"learning_rate": 4.015555555555556e-06,
"loss": 0.2998,
"step": 3200
},
{
"epoch": 0.32,
"learning_rate": 3.96e-06,
"loss": 0.2936,
"step": 3225
},
{
"epoch": 0.33,
"learning_rate": 3.904444444444444e-06,
"loss": 0.2879,
"step": 3250
},
{
"epoch": 0.33,
"learning_rate": 3.848888888888889e-06,
"loss": 0.2901,
"step": 3275
},
{
"epoch": 0.33,
"learning_rate": 3.793333333333334e-06,
"loss": 0.2974,
"step": 3300
},
{
"epoch": 0.33,
"learning_rate": 3.737777777777778e-06,
"loss": 0.2978,
"step": 3325
},
{
"epoch": 0.34,
"learning_rate": 3.6822222222222225e-06,
"loss": 0.2932,
"step": 3350
},
{
"epoch": 0.34,
"learning_rate": 3.6266666666666674e-06,
"loss": 0.2953,
"step": 3375
},
{
"epoch": 0.34,
"learning_rate": 3.5711111111111114e-06,
"loss": 0.2905,
"step": 3400
},
{
"epoch": 0.34,
"learning_rate": 3.515555555555556e-06,
"loss": 0.2986,
"step": 3425
},
{
"epoch": 0.35,
"learning_rate": 3.46e-06,
"loss": 0.2943,
"step": 3450
},
{
"epoch": 0.35,
"learning_rate": 3.404444444444445e-06,
"loss": 0.2913,
"step": 3475
},
{
"epoch": 0.35,
"learning_rate": 3.3488888888888892e-06,
"loss": 0.2879,
"step": 3500
},
{
"epoch": 0.35,
"learning_rate": 3.2933333333333333e-06,
"loss": 0.2915,
"step": 3525
},
{
"epoch": 0.36,
"learning_rate": 3.237777777777778e-06,
"loss": 0.2925,
"step": 3550
},
{
"epoch": 0.36,
"learning_rate": 3.1822222222222226e-06,
"loss": 0.2876,
"step": 3575
},
{
"epoch": 0.36,
"learning_rate": 3.1266666666666667e-06,
"loss": 0.2875,
"step": 3600
},
{
"epoch": 0.36,
"learning_rate": 3.0711111111111115e-06,
"loss": 0.2947,
"step": 3625
},
{
"epoch": 0.37,
"learning_rate": 3.015555555555556e-06,
"loss": 0.2881,
"step": 3650
},
{
"epoch": 0.37,
"learning_rate": 2.96e-06,
"loss": 0.2887,
"step": 3675
},
{
"epoch": 0.37,
"learning_rate": 2.904444444444445e-06,
"loss": 0.2853,
"step": 3700
},
{
"epoch": 0.37,
"learning_rate": 2.8488888888888894e-06,
"loss": 0.2893,
"step": 3725
},
{
"epoch": 0.38,
"learning_rate": 2.7933333333333334e-06,
"loss": 0.2885,
"step": 3750
},
{
"epoch": 0.38,
"learning_rate": 2.7377777777777783e-06,
"loss": 0.2904,
"step": 3775
},
{
"epoch": 0.38,
"learning_rate": 2.6822222222222223e-06,
"loss": 0.2924,
"step": 3800
},
{
"epoch": 0.38,
"learning_rate": 2.6266666666666668e-06,
"loss": 0.2816,
"step": 3825
},
{
"epoch": 0.39,
"learning_rate": 2.5711111111111112e-06,
"loss": 0.285,
"step": 3850
},
{
"epoch": 0.39,
"learning_rate": 2.5155555555555557e-06,
"loss": 0.2857,
"step": 3875
},
{
"epoch": 0.39,
"learning_rate": 2.46e-06,
"loss": 0.284,
"step": 3900
},
{
"epoch": 0.39,
"learning_rate": 2.4044444444444446e-06,
"loss": 0.2901,
"step": 3925
},
{
"epoch": 0.4,
"learning_rate": 2.348888888888889e-06,
"loss": 0.2904,
"step": 3950
},
{
"epoch": 0.4,
"learning_rate": 2.2933333333333335e-06,
"loss": 0.2866,
"step": 3975
},
{
"epoch": 0.4,
"learning_rate": 2.237777777777778e-06,
"loss": 0.2812,
"step": 4000
},
{
"epoch": 0.4,
"eval_loss": 0.2694556713104248,
"eval_runtime": 11782.0085,
"eval_samples_per_second": 16.873,
"eval_steps_per_second": 0.075,
"eval_wer": 0.22669475632761552,
"step": 4000
},
{
"epoch": 0.4,
"learning_rate": 2.1822222222222225e-06,
"loss": 0.2804,
"step": 4025
},
{
"epoch": 0.41,
"learning_rate": 2.126666666666667e-06,
"loss": 0.2875,
"step": 4050
},
{
"epoch": 0.41,
"learning_rate": 2.0711111111111114e-06,
"loss": 0.2839,
"step": 4075
},
{
"epoch": 0.41,
"learning_rate": 2.0155555555555554e-06,
"loss": 0.2837,
"step": 4100
},
{
"epoch": 0.41,
"learning_rate": 1.9600000000000003e-06,
"loss": 0.2798,
"step": 4125
},
{
"epoch": 0.42,
"learning_rate": 1.9044444444444445e-06,
"loss": 0.2895,
"step": 4150
},
{
"epoch": 0.42,
"learning_rate": 1.848888888888889e-06,
"loss": 0.2829,
"step": 4175
},
{
"epoch": 0.42,
"learning_rate": 1.7933333333333337e-06,
"loss": 0.2836,
"step": 4200
},
{
"epoch": 0.42,
"learning_rate": 1.737777777777778e-06,
"loss": 0.2807,
"step": 4225
},
{
"epoch": 0.43,
"learning_rate": 1.6822222222222224e-06,
"loss": 0.2784,
"step": 4250
},
{
"epoch": 0.43,
"learning_rate": 1.6266666666666666e-06,
"loss": 0.2832,
"step": 4275
},
{
"epoch": 0.43,
"learning_rate": 1.5711111111111113e-06,
"loss": 0.2833,
"step": 4300
},
{
"epoch": 0.43,
"learning_rate": 1.5155555555555558e-06,
"loss": 0.2831,
"step": 4325
},
{
"epoch": 0.44,
"learning_rate": 1.46e-06,
"loss": 0.2795,
"step": 4350
},
{
"epoch": 0.44,
"learning_rate": 1.4044444444444447e-06,
"loss": 0.2827,
"step": 4375
},
{
"epoch": 0.44,
"learning_rate": 1.3488888888888891e-06,
"loss": 0.2838,
"step": 4400
},
{
"epoch": 0.44,
"learning_rate": 1.2933333333333334e-06,
"loss": 0.2755,
"step": 4425
},
{
"epoch": 0.45,
"learning_rate": 1.2377777777777778e-06,
"loss": 0.2812,
"step": 4450
},
{
"epoch": 0.45,
"learning_rate": 1.1822222222222223e-06,
"loss": 0.2791,
"step": 4475
},
{
"epoch": 0.45,
"learning_rate": 1.1266666666666667e-06,
"loss": 0.2776,
"step": 4500
},
{
"epoch": 0.45,
"learning_rate": 1.0711111111111112e-06,
"loss": 0.2778,
"step": 4525
},
{
"epoch": 0.46,
"learning_rate": 1.0155555555555557e-06,
"loss": 0.28,
"step": 4550
},
{
"epoch": 0.46,
"learning_rate": 9.600000000000001e-07,
"loss": 0.2822,
"step": 4575
},
{
"epoch": 0.46,
"learning_rate": 9.044444444444445e-07,
"loss": 0.2852,
"step": 4600
},
{
"epoch": 0.46,
"learning_rate": 8.488888888888889e-07,
"loss": 0.2752,
"step": 4625
},
{
"epoch": 0.47,
"learning_rate": 7.933333333333335e-07,
"loss": 0.282,
"step": 4650
},
{
"epoch": 0.47,
"learning_rate": 7.377777777777779e-07,
"loss": 0.2763,
"step": 4675
},
{
"epoch": 0.47,
"learning_rate": 6.822222222222223e-07,
"loss": 0.2765,
"step": 4700
},
{
"epoch": 0.47,
"learning_rate": 6.266666666666667e-07,
"loss": 0.2815,
"step": 4725
},
{
"epoch": 0.48,
"learning_rate": 5.711111111111111e-07,
"loss": 0.2733,
"step": 4750
},
{
"epoch": 0.48,
"learning_rate": 5.155555555555556e-07,
"loss": 0.2831,
"step": 4775
},
{
"epoch": 0.48,
"learning_rate": 4.6000000000000004e-07,
"loss": 0.2784,
"step": 4800
},
{
"epoch": 0.48,
"learning_rate": 4.0444444444444445e-07,
"loss": 0.2785,
"step": 4825
},
{
"epoch": 0.49,
"learning_rate": 3.488888888888889e-07,
"loss": 0.2784,
"step": 4850
},
{
"epoch": 0.49,
"learning_rate": 2.9333333333333337e-07,
"loss": 0.2785,
"step": 4875
},
{
"epoch": 0.49,
"learning_rate": 2.3777777777777777e-07,
"loss": 0.28,
"step": 4900
},
{
"epoch": 0.5,
"learning_rate": 1.8222222222222226e-07,
"loss": 0.2753,
"step": 4925
},
{
"epoch": 0.5,
"learning_rate": 1.2666666666666666e-07,
"loss": 0.2799,
"step": 4950
},
{
"epoch": 0.5,
"learning_rate": 7.111111111111112e-08,
"loss": 0.278,
"step": 4975
},
{
"epoch": 0.5,
"learning_rate": 1.5555555555555557e-08,
"loss": 0.2765,
"step": 5000
},
{
"epoch": 0.5,
"eval_loss": 0.2623128890991211,
"eval_runtime": 11763.5863,
"eval_samples_per_second": 16.899,
"eval_steps_per_second": 0.075,
"eval_wer": 0.22299687523689737,
"step": 5000
},
{
"epoch": 0.5,
"step": 5000,
"total_flos": 2.3779516205911507e+21,
"train_loss": 0.31619301767349245,
"train_runtime": 115130.9341,
"train_samples_per_second": 9.728,
"train_steps_per_second": 0.043
}
],
"max_steps": 5000,
"num_train_epochs": 1,
"total_flos": 2.3779516205911507e+21,
"trial_name": null,
"trial_params": null
}
{
"best_metric": 0.22299687523689737,
"best_model_checkpoint": "./whisper-large-v2-AiHub/checkpoint-5000",
"epoch": 0.502563071665494,
"global_step": 5000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"learning_rate": 4.6000000000000004e-07,
"loss": 1.0382,
"step": 25
},
{
"epoch": 0.01,
"learning_rate": 9.600000000000001e-07,
"loss": 0.8725,
"step": 50
},
{
"epoch": 0.01,
"learning_rate": 1.46e-06,
"loss": 0.5618,
"step": 75
},
{
"epoch": 0.01,
"learning_rate": 1.9600000000000003e-06,
"loss": 0.4216,
"step": 100
},
{
"epoch": 0.01,
"learning_rate": 2.46e-06,
"loss": 0.3748,
"step": 125
},
{
"epoch": 0.02,
"learning_rate": 2.96e-06,
"loss": 0.3507,
"step": 150
},
{
"epoch": 0.02,
"learning_rate": 3.44e-06,
"loss": 0.3454,
"step": 175
},
{
"epoch": 0.02,
"learning_rate": 3.94e-06,
"loss": 0.3512,
"step": 200
},
{
"epoch": 0.02,
"learning_rate": 4.440000000000001e-06,
"loss": 0.3427,
"step": 225
},
{
"epoch": 0.03,
"learning_rate": 4.92e-06,
"loss": 0.3455,
"step": 250
},
{
"epoch": 0.03,
"learning_rate": 5.420000000000001e-06,
"loss": 0.3427,
"step": 275
},
{
"epoch": 0.03,
"learning_rate": 5.9e-06,
"loss": 0.3385,
"step": 300
},
{
"epoch": 0.03,
"learning_rate": 6.4000000000000006e-06,
"loss": 0.3348,
"step": 325
},
{
"epoch": 0.04,
"learning_rate": 6.9e-06,
"loss": 0.3412,
"step": 350
},
{
"epoch": 0.04,
"learning_rate": 7.4e-06,
"loss": 0.3366,
"step": 375
},
{