| { | |
| "best_metric": 50.680544435548434, | |
| "best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/canine/canine-base-finetuned-parsing-ud-Tamil-TTB/checkpoint-4000", | |
| "epoch": 500.0, | |
| "global_step": 6500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 7.680000000000001e-05, | |
| "loss": 3.657, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 7.948456375838927e-05, | |
| "loss": 0.6371, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 23.08, | |
| "learning_rate": 7.894765100671142e-05, | |
| "loss": 0.2692, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 30.77, | |
| "learning_rate": 7.841073825503357e-05, | |
| "loss": 0.1642, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 38.46, | |
| "learning_rate": 7.78738255033557e-05, | |
| "loss": 0.1262, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 38.46, | |
| "eval_las": 47.47798238590873, | |
| "eval_loss": 7.3044867515563965, | |
| "eval_runtime": 0.4605, | |
| "eval_samples_per_second": 173.711, | |
| "eval_steps_per_second": 21.714, | |
| "eval_uas": 59.887910328262606, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 46.15, | |
| "learning_rate": 7.733691275167786e-05, | |
| "loss": 0.1003, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 53.85, | |
| "learning_rate": 7.680000000000001e-05, | |
| "loss": 0.0887, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 61.54, | |
| "learning_rate": 7.626308724832216e-05, | |
| "loss": 0.0748, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 69.23, | |
| "learning_rate": 7.57261744966443e-05, | |
| "loss": 0.0732, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 76.92, | |
| "learning_rate": 7.518926174496645e-05, | |
| "loss": 0.0621, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 76.92, | |
| "eval_las": 49.159327461969575, | |
| "eval_loss": 9.363687515258789, | |
| "eval_runtime": 0.4593, | |
| "eval_samples_per_second": 174.192, | |
| "eval_steps_per_second": 21.774, | |
| "eval_uas": 61.0088070456365, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 84.62, | |
| "learning_rate": 7.46523489932886e-05, | |
| "loss": 0.0633, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 92.31, | |
| "learning_rate": 7.411543624161075e-05, | |
| "loss": 0.0595, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 7.35785234899329e-05, | |
| "loss": 0.0505, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 107.69, | |
| "learning_rate": 7.304161073825505e-05, | |
| "loss": 0.051, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 115.38, | |
| "learning_rate": 7.25046979865772e-05, | |
| "loss": 0.0511, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 115.38, | |
| "eval_las": 48.438751000800636, | |
| "eval_loss": 10.332539558410645, | |
| "eval_runtime": 0.4548, | |
| "eval_samples_per_second": 175.918, | |
| "eval_steps_per_second": 21.99, | |
| "eval_uas": 61.569255404323464, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 123.08, | |
| "learning_rate": 7.196778523489934e-05, | |
| "loss": 0.0512, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 130.77, | |
| "learning_rate": 7.143087248322148e-05, | |
| "loss": 0.043, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 138.46, | |
| "learning_rate": 7.089395973154363e-05, | |
| "loss": 0.0467, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 146.15, | |
| "learning_rate": 7.035704697986578e-05, | |
| "loss": 0.0471, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 153.85, | |
| "learning_rate": 6.982013422818792e-05, | |
| "loss": 0.0409, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 153.85, | |
| "eval_las": 49.71977582065652, | |
| "eval_loss": 10.502817153930664, | |
| "eval_runtime": 0.4442, | |
| "eval_samples_per_second": 180.102, | |
| "eval_steps_per_second": 22.513, | |
| "eval_uas": 61.809447558046436, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 161.54, | |
| "learning_rate": 6.928322147651007e-05, | |
| "loss": 0.0402, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 169.23, | |
| "learning_rate": 6.874630872483222e-05, | |
| "loss": 0.0397, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 176.92, | |
| "learning_rate": 6.820939597315437e-05, | |
| "loss": 0.0361, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 184.62, | |
| "learning_rate": 6.767248322147652e-05, | |
| "loss": 0.0443, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 192.31, | |
| "learning_rate": 6.713557046979866e-05, | |
| "loss": 0.0409, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 192.31, | |
| "eval_las": 49.23939151321057, | |
| "eval_loss": 10.978870391845703, | |
| "eval_runtime": 0.4462, | |
| "eval_samples_per_second": 179.297, | |
| "eval_steps_per_second": 22.412, | |
| "eval_uas": 60.368294635708565, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 200.0, | |
| "learning_rate": 6.659865771812081e-05, | |
| "loss": 0.0402, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 207.69, | |
| "learning_rate": 6.606174496644296e-05, | |
| "loss": 0.0389, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 215.38, | |
| "learning_rate": 6.55248322147651e-05, | |
| "loss": 0.0374, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 223.08, | |
| "learning_rate": 6.498791946308724e-05, | |
| "loss": 0.0369, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 230.77, | |
| "learning_rate": 6.445100671140939e-05, | |
| "loss": 0.04, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 230.77, | |
| "eval_las": 48.35868694955965, | |
| "eval_loss": 10.903715133666992, | |
| "eval_runtime": 0.469, | |
| "eval_samples_per_second": 170.59, | |
| "eval_steps_per_second": 21.324, | |
| "eval_uas": 60.848678943154525, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 238.46, | |
| "learning_rate": 6.391409395973154e-05, | |
| "loss": 0.0368, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 246.15, | |
| "learning_rate": 6.337718120805369e-05, | |
| "loss": 0.0361, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 253.85, | |
| "learning_rate": 6.284026845637584e-05, | |
| "loss": 0.0372, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 261.54, | |
| "learning_rate": 6.230335570469799e-05, | |
| "loss": 0.0371, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 269.23, | |
| "learning_rate": 6.176644295302013e-05, | |
| "loss": 0.036, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 269.23, | |
| "eval_las": 49.55964771817454, | |
| "eval_loss": 11.492471694946289, | |
| "eval_runtime": 0.4652, | |
| "eval_samples_per_second": 171.976, | |
| "eval_steps_per_second": 21.497, | |
| "eval_uas": 61.40912730184147, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 276.92, | |
| "learning_rate": 6.122953020134228e-05, | |
| "loss": 0.0406, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 284.62, | |
| "learning_rate": 6.069261744966444e-05, | |
| "loss": 0.0334, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 292.31, | |
| "learning_rate": 6.0155704697986585e-05, | |
| "loss": 0.0375, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 300.0, | |
| "learning_rate": 5.9618791946308734e-05, | |
| "loss": 0.0315, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 307.69, | |
| "learning_rate": 5.9081879194630875e-05, | |
| "loss": 0.0344, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 307.69, | |
| "eval_las": 50.680544435548434, | |
| "eval_loss": 11.751625061035156, | |
| "eval_runtime": 0.454, | |
| "eval_samples_per_second": 176.202, | |
| "eval_steps_per_second": 22.025, | |
| "eval_uas": 62.20976781425141, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 315.38, | |
| "learning_rate": 5.854496644295302e-05, | |
| "loss": 0.034, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 323.08, | |
| "learning_rate": 5.800805369127517e-05, | |
| "loss": 0.0336, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 330.77, | |
| "learning_rate": 5.747114093959732e-05, | |
| "loss": 0.0308, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 338.46, | |
| "learning_rate": 5.693422818791947e-05, | |
| "loss": 0.0338, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 346.15, | |
| "learning_rate": 5.6397315436241616e-05, | |
| "loss": 0.0333, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 346.15, | |
| "eval_las": 50.28022417934348, | |
| "eval_loss": 11.701101303100586, | |
| "eval_runtime": 0.4679, | |
| "eval_samples_per_second": 170.959, | |
| "eval_steps_per_second": 21.37, | |
| "eval_uas": 62.20976781425141, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 353.85, | |
| "learning_rate": 5.5860402684563764e-05, | |
| "loss": 0.0349, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 361.54, | |
| "learning_rate": 5.532348993288591e-05, | |
| "loss": 0.0338, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 369.23, | |
| "learning_rate": 5.478657718120806e-05, | |
| "loss": 0.0321, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 376.92, | |
| "learning_rate": 5.424966442953021e-05, | |
| "loss": 0.0334, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 384.62, | |
| "learning_rate": 5.371275167785236e-05, | |
| "loss": 0.0345, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 384.62, | |
| "eval_las": 50.200160128102475, | |
| "eval_loss": 10.622541427612305, | |
| "eval_runtime": 0.4673, | |
| "eval_samples_per_second": 171.194, | |
| "eval_steps_per_second": 21.399, | |
| "eval_uas": 62.44995996797438, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 392.31, | |
| "learning_rate": 5.31758389261745e-05, | |
| "loss": 0.033, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 400.0, | |
| "learning_rate": 5.263892617449665e-05, | |
| "loss": 0.03, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 407.69, | |
| "learning_rate": 5.2102013422818795e-05, | |
| "loss": 0.0295, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 415.38, | |
| "learning_rate": 5.1565100671140944e-05, | |
| "loss": 0.0325, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 423.08, | |
| "learning_rate": 5.102818791946309e-05, | |
| "loss": 0.0364, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 423.08, | |
| "eval_las": 49.879903923138514, | |
| "eval_loss": 11.19732666015625, | |
| "eval_runtime": 0.4631, | |
| "eval_samples_per_second": 172.758, | |
| "eval_steps_per_second": 21.595, | |
| "eval_uas": 62.770216172938355, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 430.77, | |
| "learning_rate": 5.049127516778524e-05, | |
| "loss": 0.033, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 438.46, | |
| "learning_rate": 4.995436241610739e-05, | |
| "loss": 0.0308, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 446.15, | |
| "learning_rate": 4.941744966442954e-05, | |
| "loss": 0.0337, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 453.85, | |
| "learning_rate": 4.8880536912751685e-05, | |
| "loss": 0.0323, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 461.54, | |
| "learning_rate": 4.834362416107383e-05, | |
| "loss": 0.0358, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 461.54, | |
| "eval_las": 49.079263410728586, | |
| "eval_loss": 11.815637588500977, | |
| "eval_runtime": 0.462, | |
| "eval_samples_per_second": 173.176, | |
| "eval_steps_per_second": 21.647, | |
| "eval_uas": 61.24899919935949, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 469.23, | |
| "learning_rate": 4.780671140939598e-05, | |
| "loss": 0.0279, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 476.92, | |
| "learning_rate": 4.726979865771813e-05, | |
| "loss": 0.0309, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 484.62, | |
| "learning_rate": 4.673288590604027e-05, | |
| "loss": 0.0327, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 492.31, | |
| "learning_rate": 4.619597315436242e-05, | |
| "loss": 0.0318, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 500.0, | |
| "learning_rate": 4.565906040268457e-05, | |
| "loss": 0.0287, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 500.0, | |
| "eval_las": 50.200160128102475, | |
| "eval_loss": 13.070086479187012, | |
| "eval_runtime": 0.4625, | |
| "eval_samples_per_second": 172.986, | |
| "eval_steps_per_second": 21.623, | |
| "eval_uas": 62.20976781425141, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 500.0, | |
| "step": 6500, | |
| "total_flos": 3.99259060224e+16, | |
| "train_loss": 0.11270012881205632, | |
| "train_runtime": 2562.2849, | |
| "train_samples_per_second": 187.333, | |
| "train_steps_per_second": 5.854 | |
| } | |
| ], | |
| "max_steps": 15000, | |
| "num_train_epochs": 1154, | |
| "total_flos": 3.99259060224e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |