| { | |
| "best_metric": 0.36759981513023376, | |
| "best_model_checkpoint": "nrshoudi/wav2vec-arabic-V2-10/checkpoint-2952", | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 3280, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 5e-05, | |
| "loss": 22.0067, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 5.513646125793457, | |
| "eval_per": 1.0, | |
| "eval_runtime": 124.4365, | |
| "eval_samples_per_second": 6.533, | |
| "eval_steps_per_second": 3.271, | |
| "eval_wer": 1.0, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0001, | |
| "loss": 3.7377, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 3.274935483932495, | |
| "eval_per": 1.0, | |
| "eval_runtime": 123.268, | |
| "eval_samples_per_second": 6.595, | |
| "eval_steps_per_second": 3.302, | |
| "eval_wer": 1.0, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 9.444444444444444e-05, | |
| "loss": 3.2647, | |
| "step": 492 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 3.316337823867798, | |
| "eval_per": 1.0, | |
| "eval_runtime": 123.6463, | |
| "eval_samples_per_second": 6.575, | |
| "eval_steps_per_second": 3.292, | |
| "eval_wer": 1.0, | |
| "step": 492 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 3.2513, | |
| "step": 656 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 3.2318897247314453, | |
| "eval_per": 1.0, | |
| "eval_runtime": 123.6244, | |
| "eval_samples_per_second": 6.576, | |
| "eval_steps_per_second": 3.292, | |
| "eval_wer": 1.0, | |
| "step": 656 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 3.2314, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 3.290160894393921, | |
| "eval_per": 1.0, | |
| "eval_runtime": 123.9537, | |
| "eval_samples_per_second": 6.559, | |
| "eval_steps_per_second": 3.283, | |
| "eval_wer": 1.0, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 7.777777777777778e-05, | |
| "loss": 3.1754, | |
| "step": 984 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 3.0333683490753174, | |
| "eval_per": 1.0, | |
| "eval_runtime": 125.221, | |
| "eval_samples_per_second": 6.493, | |
| "eval_steps_per_second": 3.25, | |
| "eval_wer": 1.0, | |
| "step": 984 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 7.222222222222222e-05, | |
| "loss": 2.8137, | |
| "step": 1148 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 2.5538418292999268, | |
| "eval_per": 0.9959248091962687, | |
| "eval_runtime": 125.8623, | |
| "eval_samples_per_second": 6.459, | |
| "eval_steps_per_second": 3.234, | |
| "eval_wer": 0.9930474546218095, | |
| "step": 1148 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 2.1146, | |
| "step": 1312 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.657683253288269, | |
| "eval_per": 0.6457881843022708, | |
| "eval_runtime": 126.4952, | |
| "eval_samples_per_second": 6.427, | |
| "eval_steps_per_second": 3.218, | |
| "eval_wer": 0.6541925248471839, | |
| "step": 1312 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 6.111111111111112e-05, | |
| "loss": 1.2679, | |
| "step": 1476 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.8286688923835754, | |
| "eval_per": 0.18025063601243757, | |
| "eval_runtime": 126.0524, | |
| "eval_samples_per_second": 6.45, | |
| "eval_steps_per_second": 3.229, | |
| "eval_wer": 0.20890299099435397, | |
| "step": 1476 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 5.555555555555556e-05, | |
| "loss": 0.6981, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.580149233341217, | |
| "eval_per": 0.12289173654951474, | |
| "eval_runtime": 125.8066, | |
| "eval_samples_per_second": 6.462, | |
| "eval_steps_per_second": 3.235, | |
| "eval_wer": 0.14227054267183054, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4737, | |
| "step": 1804 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.48674020171165466, | |
| "eval_per": 0.10169132196362951, | |
| "eval_runtime": 126.8326, | |
| "eval_samples_per_second": 6.41, | |
| "eval_steps_per_second": 3.209, | |
| "eval_wer": 0.1219261816994074, | |
| "step": 1804 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.3733, | |
| "step": 1968 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.4481865465641022, | |
| "eval_per": 0.09288137190238387, | |
| "eval_runtime": 125.9688, | |
| "eval_samples_per_second": 6.454, | |
| "eval_steps_per_second": 3.231, | |
| "eval_wer": 0.10979422332135691, | |
| "step": 1968 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.2913, | |
| "step": 2132 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.4176279902458191, | |
| "eval_per": 0.09330538019410158, | |
| "eval_runtime": 125.8355, | |
| "eval_samples_per_second": 6.461, | |
| "eval_steps_per_second": 3.234, | |
| "eval_wer": 0.10918762540245439, | |
| "step": 2132 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.2358, | |
| "step": 2296 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.41998350620269775, | |
| "eval_per": 0.0915151229624046, | |
| "eval_runtime": 126.5555, | |
| "eval_samples_per_second": 6.424, | |
| "eval_steps_per_second": 3.216, | |
| "eval_wer": 0.10844104334842052, | |
| "step": 2296 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.2196, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.399908185005188, | |
| "eval_per": 0.08235183265806087, | |
| "eval_runtime": 126.1327, | |
| "eval_samples_per_second": 6.446, | |
| "eval_steps_per_second": 3.227, | |
| "eval_wer": 0.09868881526760301, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.1845, | |
| "step": 2624 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.39929312467575073, | |
| "eval_per": 0.08256383680391972, | |
| "eval_runtime": 126.7957, | |
| "eval_samples_per_second": 6.412, | |
| "eval_steps_per_second": 3.21, | |
| "eval_wer": 0.10060193178106482, | |
| "step": 2624 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.1831, | |
| "step": 2788 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.3811805844306946, | |
| "eval_per": 0.0653443889569396, | |
| "eval_runtime": 126.0916, | |
| "eval_samples_per_second": 6.448, | |
| "eval_steps_per_second": 3.228, | |
| "eval_wer": 0.08100415286267557, | |
| "step": 2788 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.1629, | |
| "step": 2952 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.36759981513023376, | |
| "eval_per": 0.06357768774144916, | |
| "eval_runtime": 126.4537, | |
| "eval_samples_per_second": 6.429, | |
| "eval_steps_per_second": 3.219, | |
| "eval_wer": 0.07848443843031123, | |
| "step": 2952 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.1578, | |
| "step": 3116 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.3679310083389282, | |
| "eval_per": 0.06576839724865731, | |
| "eval_runtime": 126.1007, | |
| "eval_samples_per_second": 6.447, | |
| "eval_steps_per_second": 3.228, | |
| "eval_wer": 0.08063086183565862, | |
| "step": 3116 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.1317, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.3686366677284241, | |
| "eval_per": 0.06133986620182795, | |
| "eval_runtime": 126.1017, | |
| "eval_samples_per_second": 6.447, | |
| "eval_steps_per_second": 3.228, | |
| "eval_wer": 0.07629135364658672, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 3280, | |
| "total_flos": 1.036683517600052e+18, | |
| "train_loss": 2.3987663961038357, | |
| "train_runtime": 4491.8663, | |
| "train_samples_per_second": 1.456, | |
| "train_steps_per_second": 0.73 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3280, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 1.036683517600052e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |