Update README.md

README.md

@@ -96,11 +96,48 @@ model = AutoModelForCausalLM.from_pretrained("supermy/poetry")
 
 ## Training procedure
 
-在英伟达16G显卡训练了 4 天整，
-num_train_epochs=680。
-
-
-模型[GPT2](https://huggingface.co/gpt2) 
+模型：[GPT2](https://huggingface.co/gpt2) 
+训练环境：英伟达16G显卡
+
+bpe分词："vocab_size"=50000
+
+***** Running training *****
+  Num examples = 16431
+  Num Epochs = 680
+  Instantaneous batch size per device = 24
+  Total train batch size (w. parallel, distributed & accumulation) = 192
+  Gradient Accumulation steps = 8
+  Total optimization steps = 57800
+  Number of trainable parameters = 124242432
+GPT-2 size: 124.2M parameters
+  0%|          | 0/57800 [00:00<?, ?it/s]You're using a PreTrainedTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.
+    9%|▊         | 5000/57800 [6:58:57<72:53:18,  4.97s/it]***** Running Evaluation *****
+  Num examples = 1755
+  Batch size = 24
+{'loss': 3.1345, 'learning_rate': 0.0004939065828881268, 'epoch': 58.82}
+  9%|▊         | 5000/57800 [6:59:14<72:53:18, Saving model checkpoint to poetry-trainer/checkpoint-5000
+Configuration saved in poetry-trainer/checkpoint-5000/config.json
+Model weights saved in poetry-trainer/checkpoint-5000/pytorch_model.bin
+tokenizer config file saved in poetry-trainer/checkpoint-5000/tokenizer_config.json
+Special tokens file saved in poetry-trainer/checkpoint-5000/special_tokens_map.json
+ 17%|█▋        | 10000/57800 [13:55:32<65:40:41,  4.95s/it]***** Running Evaluation *****
+  Num examples = 1755
+  Batch size = 24
+{'eval_loss': 11.14090633392334, 'eval_runtime': 16.8326, 'eval_samples_per_second': 104.262, 'eval_steps_per_second': 4.396, 'epoch': 58.82}
+{'loss': 0.2511, 'learning_rate': 0.00046966687938531824, 'epoch': 117.64}
+ 17%|█▋        | 10000/57800 [13:55:48<65:40:41Saving model checkpoint to poetry-trainer/checkpoint-10000
+..........
+ 95%|█████████▌| 55000/57800 [76:06:46<3:59:33,  5.13s/it]***** Running Evaluation *****
+  Num examples = 1755
+  Batch size = 24
+{'eval_loss': 14.860174179077148, 'eval_runtime': 16.7826, 'eval_samples_per_second': 104.572, 'eval_steps_per_second': 4.409, 'epoch': 588.23}
+{'loss': 0.0083, 'learning_rate': 3.0262183266589473e-06, 'epoch': 647.06}
+ 95%|█████████▌| 55000/57800 [76:07:03<3:59:33,Saving model checkpoint to poetry-trainer/checkpoint-55000
+
+{'eval_loss': 14.830656051635742, 'eval_runtime': 16.7365, 'eval_samples_per_second': 104.86, 'eval_steps_per_second': 4.421, 'epoch': 647.06}
+{'train_runtime': 287920.5857, 'train_samples_per_second': 38.806, 'train_steps_per_second': 0.201, 'train_loss': 0.33751299874592816, 'epoch': 679.99}
+
+100%|██████████| 57800/57800 [79:58:40<00:00,  4.93s/it]  
 
 
 ```