From 734628a61effdd8d2d2239e81aac58d0bccd4bab Mon Sep 17 00:00:00 2001
From: ssbuild <462304@qq.com>
Date: Wed, 12 Jun 2024 23:21:44 +0800
Subject: [PATCH] glm4

---
 README.md                                          |  5 +++--
 src/deep_training/zoo/model_zoo/glm4v/llm_model.py | 10 ++++------
 2 files changed, 7 insertions(+), 8 deletions(-)

diff --git a/README.md b/README.md
index f55ad69..68e9d47 100644
--- a/README.md
+++ b/README.md
@@ -18,10 +18,11 @@ pip install -U git+https://github.com/ssbuild/deep_training.git --no-deps --forc
   
 ## update
 - <strong>2024-06-10</strong>
-   - 0.3.1 support glm4,glm4v
+   - 0.3.1 support glm4 https://github.com/ssbuild/glm4_finetuning 
+   glm4v https://github.com/ssbuild/glm4v_finetuning 
 
 - <strong>2024-02-15</strong>
-   - 0.2.11 support internlm2
+   - 0.2.11 support internlm2 https://github.com/ssbuild/internlm2_finetuning
 
 - <strong>2023-12-02</strong>
    - 0.2.10 update qwen model for 1.8b 7b 14b 72b
diff --git a/src/deep_training/zoo/model_zoo/glm4v/llm_model.py b/src/deep_training/zoo/model_zoo/glm4v/llm_model.py
index b4b896a..189f9fd 100644
--- a/src/deep_training/zoo/model_zoo/glm4v/llm_model.py
+++ b/src/deep_training/zoo/model_zoo/glm4v/llm_model.py
@@ -64,15 +64,14 @@ def process_response(self, output, history):
 
     @torch.inference_mode()
     def chat(self, tokenizer, query: str, history: List[Dict] = None, role: str = "user", image=None,
-             max_length: int = 8192, num_beams=1, do_sample=True, top_p=0.8, temperature=0.8, logits_processor=None,
+             logits_processor=None,
              **kwargs):
         if history is None:
             history = []
         if logits_processor is None:
             logits_processor = LogitsProcessorList()
         logits_processor.append(InvalidScoreLogitsProcessor())
-        gen_kwargs = {"max_length": max_length, "num_beams": num_beams, "do_sample": do_sample, "top_p": top_p,
-                      "temperature": temperature, "logits_processor": logits_processor, **kwargs}
+        gen_kwargs = {"logits_processor": logits_processor, **kwargs}
         message = {"role": role, "content": query}
         if image is not None:
             message["image"] = image
@@ -90,7 +89,7 @@ def chat(self, tokenizer, query: str, history: List[Dict] = None, role: str = "u
 
     @torch.inference_mode()
     def stream_chat(self, tokenizer, query: str, history: List[Dict] = None, role: str = "user", image=None,
-                    past_key_values=None, max_length: int = 8192, do_sample=True, top_p=0.8, temperature=0.8,
+                    past_key_values=None, 
                     logits_processor=None, return_past_key_values=False, **kwargs):
         if history is None:
             history = []
@@ -99,8 +98,7 @@ def stream_chat(self, tokenizer, query: str, history: List[Dict] = None, role: s
         logits_processor.append(InvalidScoreLogitsProcessor())
         eos_token_id = [tokenizer.eos_token_id, tokenizer.convert_tokens_to_ids("<|user|>"),
                         tokenizer.convert_tokens_to_ids("<|observation|>")]
-        gen_kwargs = {"max_length": max_length, "do_sample": do_sample, "top_p": top_p,
-                      "temperature": temperature, "logits_processor": logits_processor, **kwargs}
+        gen_kwargs = {"logits_processor": logits_processor, **kwargs}
         message = {"role": role, "content": "query"}
         if image is not None:
             message["image"] = image