InternLM-Math commited on
Commit
d706250
1 Parent(s): 72df0ed

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -13
README.md CHANGED
@@ -87,18 +87,8 @@ We suggest using [LMDeploy](https://github.com/InternLM/LMDeploy)(>=0.2.1) for i
87
  from lmdeploy import pipeline, TurbomindEngineConfig, ChatTemplateConfig
88
 
89
  backend_config = TurbomindEngineConfig(model_name='internlm2-chat-7b', tp=1, cache_max_entry_count=0.3)
90
- chat_template = ChatTemplateConfig(model_name='internlm2-chat-7b',
91
- system='',
92
- eosys='',
93
- meta_instruction='',
94
- user='<|im_start|>user\n',
95
- assistant='<|im_start|>assistant\n',
96
- eoh='<|im_end|>\n',
97
- eoa='<|im_end|>\n',
98
- stop_words=['<|im_end|>', '<|action_end|>'])
99
- pipe = pipeline(model_path='internlm/internlm2-math-7b',
100
- chat_template_config=chat_template,
101
- backend_config=backend_config)
102
 
103
  problem = '1+1='
104
  result = pipe([problem], request_output_len=1024, top_k=1)
@@ -112,7 +102,7 @@ tokenizer = AutoTokenizer.from_pretrained("internlm/internlm2-math-7b", trust_re
112
  # Set `torch_dtype=torch.float16` to load model in float16, otherwise it will be loaded as float32 and might cause OOM Error.
113
  model = AutoModelForCausalLM.from_pretrained("internlm/internlm2-math-7b", trust_remote_code=True, torch_dtype=torch.float16).cuda()
114
  model = model.eval()
115
- response, history = model.chat(tokenizer, "1+1=", history=[])
116
  print(response)
117
  ```
118
 
 
87
  from lmdeploy import pipeline, TurbomindEngineConfig, ChatTemplateConfig
88
 
89
  backend_config = TurbomindEngineConfig(model_name='internlm2-chat-7b', tp=1, cache_max_entry_count=0.3)
90
+ chat_template = ChatTemplateConfig(model_name='internlm2-chat-7b', system='', eosys='', meta_instruction='')
91
+ pipe = pipeline(model_path='internlm/internlm2-math-7b', chat_template_config=chat_template, backend_config=backend_config)
 
 
 
 
 
 
 
 
 
 
92
 
93
  problem = '1+1='
94
  result = pipe([problem], request_output_len=1024, top_k=1)
 
102
  # Set `torch_dtype=torch.float16` to load model in float16, otherwise it will be loaded as float32 and might cause OOM Error.
103
  model = AutoModelForCausalLM.from_pretrained("internlm/internlm2-math-7b", trust_remote_code=True, torch_dtype=torch.float16).cuda()
104
  model = model.eval()
105
+ response, history = model.chat(tokenizer, "1+1=", history=[], meta_instruction="")
106
  print(response)
107
  ```
108