LLM: add resume_from_checkpoint related section (#9705)
This commit is contained in:
		
							parent
							
								
									320110d158
								
							
						
					
					
						commit
						12df70953e
					
				
					 1 changed files with 3 additions and 2 deletions
				
			
		| 
						 | 
				
			
			@ -97,7 +97,8 @@ bash qalora_finetune_llama2_7b_arc_2_card.sh
 | 
			
		|||
bash qalora_finetune_llama2_7b_pvc_1550_1_tile.sh
 | 
			
		||||
```
 | 
			
		||||
 | 
			
		||||
**Important: If you fail to complete the whole finetuning process, it is suggested to resume training from a previously saved checkpoint by specifying `resume_from_checkpoint` to the local checkpoint folder as following:**
 | 
			
		||||
### 4. (Optional) Resume Training
 | 
			
		||||
If you fail to complete the whole finetuning process, it is suggested to resume training from a previously saved checkpoint by specifying `resume_from_checkpoint` to the local checkpoint folder as following:**
 | 
			
		||||
```bash
 | 
			
		||||
python ./alpaca_qlora_finetuning.py \
 | 
			
		||||
    --base_model "meta-llama/Llama-2-7b-hf" \
 | 
			
		||||
| 
						 | 
				
			
			@ -106,7 +107,7 @@ python ./alpaca_qlora_finetuning.py \
 | 
			
		|||
    --resume_from_checkpoint "./bigdl-qlora-alpaca/checkpoint-1100"
 | 
			
		||||
```
 | 
			
		||||
 | 
			
		||||
### 4. Sample Output
 | 
			
		||||
### 5. Sample Output
 | 
			
		||||
```log
 | 
			
		||||
{'loss': 1.9231, 'learning_rate': 2.9999945367033285e-05, 'epoch': 0.0}                                                                                                                            
 | 
			
		||||
{'loss': 1.8622, 'learning_rate': 2.9999781468531096e-05, 'epoch': 0.01}                                                                                                                           
 | 
			
		||||
| 
						 | 
				
			
			
 | 
			
		|||
		Loading…
	
		Reference in a new issue