Bump version to v1.5.0

Author: czczup

INSTALLATION.md

@@ -46,10 +46,10 @@
   mim install mmcv-full==1.6.2  # (optional, for mmsegmentation)
   ```
 
-- Install `transformers==4.36.2`:
+- Install `transformers==4.37.2`:
 
   ```bash
-  pip install transformers==4.36.2
+  pip install transformers==4.37.2
   ```
 
 - Install `apex` (optional):

internvl_chat/internvl/train/internvl_chat_finetune.py

@@ -42,7 +42,7 @@
 from transformers.utils.logging import (enable_default_handler,
                                         enable_explicit_format, set_verbosity)
 
-# Upgrade transformers to v4.36.2, we don't need it anymore
+# Upgrade transformers to v4.37.2, we don't need it anymore
 # replace_llama2_attn_with_flash_attn()
 replace_llama_rmsnorm_with_fused_rmsnorm()
 replace_train_sampler()

internvl_chat/internvl/train/internvl_chat_pretrain.py

@@ -43,7 +43,7 @@
 from transformers.utils.logging import (enable_default_handler,
                                         enable_explicit_format, set_verbosity)
 
-# Upgrade transformers to v4.36.2, we don't need it anymore
+# Upgrade transformers to v4.37.2, we don't need it anymore
 # replace_llama2_attn_with_flash_attn()
 replace_llama_rmsnorm_with_fused_rmsnorm()
 replace_train_sampler()

internvl_chat/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "internvl_chat"
-version = "1.2.3"
+version = "1.5.0"
 description = "Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks."
 readme = "README.md"
 requires-python = ">=3.8"
@@ -14,7 +14,7 @@ classifiers = [
 ]
 dependencies = [
     "torch==2.0.1", "torchvision==0.15.2",
-    "transformers==4.36.2", "tokenizers==0.15.1", "sentencepiece==0.1.99", "shortuuid",
+    "transformers==4.37.2", "tokenizers==0.15.1", "sentencepiece==0.1.99", "shortuuid",
     "accelerate", "peft>=0.4.0", "bitsandbytes==0.41.0",
     "pydantic<2,>=1", "markdown2[all]", "numpy", "scikit-learn>=1.2.2",
     "gradio==3.35.2", "gradio_client==0.2.9",

internvl_chat_llava/README.md

@@ -6,7 +6,7 @@ We mainly use the LLaVA codebase to evaluate InternVL in creating multimodal dia
 
 We have retained the original documentation of LLaVA 1.5 as a more detailed manual. In most cases, you will only need to refer to the new documentation that we have added.
 
-> Note: To unify the environment across different tasks, we have made some compatibility modifications to the LLaVA code, allowing it to support `transformers==4.36.2` (originally locked at 4.31.0). Please note that `transformers==4.36.2` should be installed.
+> Note: To unify the environment across different tasks, we have made some compatibility modifications to the LLaVA code, allowing it to support `transformers==4.37.2` (originally locked at 4.31.0). Please note that `transformers==4.37.2` should be installed.
 
 ## 🛠️ Installation
 

internvl_chat_llava/pyproject.toml

@@ -14,7 +14,7 @@ classifiers = [
 ]
 dependencies = [
     "torch==2.0.1", "torchvision==0.15.2",
-    "transformers==4.36.2", "tokenizers==0.15.1", "sentencepiece==0.1.99", "shortuuid",
+    "transformers==4.37.2", "tokenizers==0.15.1", "sentencepiece==0.1.99", "shortuuid",
     "accelerate", "peft>=0.4.0", "bitsandbytes==0.41.0",
     "pydantic<2,>=1", "markdown2[all]", "numpy", "scikit-learn>=1.2.2",
     "gradio==3.35.2", "gradio_client==0.2.9",