learning_rate: 5e-05
train_batch_size: 10
eval_batch_size: 10
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: constant
lr_scheduler_warmup_steps: 0.2
num_epochs: 5

Training Loss	Epoch	Step	Validation Loss	Per Phonemes	Per Hams Or Jahr	Per Shidda Or Rakhawa	Per Tafkheem Or Taqeeq	Per Itbaq	Per Safeer	Per Qalqla	Per Tikraar	Per Tafashie	Per Istitala	Per Ghonna	Average Per
0.7398	1.0	19409	0.1096	0.0397	0.0784	0.0330	0.0190	0.0226	0.0296	0.0120	0.0307	0.0089	0.0093	0.0175	0.0273
0.0807	2.0	38818	0.0588	0.0224	0.0417	0.0169	0.0094	0.0095	0.0109	0.0065	0.0107	0.0051	0.0053	0.0097	0.0135
0.0501	3.0	58227	0.0428	0.0166	0.0346	0.0122	0.0069	0.0070	0.0067	0.0048	0.0102	0.0035	0.0038	0.0075	0.0103
0.0369	4.0	77636	0.0340	0.0126	0.0258	0.0101	0.0056	0.0060	0.0055	0.0043	0.0083	0.0030	0.0034	0.0061	0.0082
0.0294	5.0	97045	0.0269	0.0104	0.0257	0.0079	0.0043	0.0120	0.0076	0.0053	0.0142	0.0027	0.0030	0.0053	0.0090

AbdullahHedeya/h6_f1536_l16_d384_trained

h6_f1536_l16_d384_trained