[ColossalChat] Update RLHF V2 (#5286)

* Add dpo. Fix sft, ppo, lora. Refactor all

* fix and tested ppo

* 2 nd round refactor

* add ci tests

* fix ci

* fix ci

* fix readme, style

* fix readme style

* fix style, fix benchmark

* reproduce benchmark result, remove useless files

* rename to ColossalChat

* use new image

* fix ci workflow

* fix ci

* use local model/tokenizer for ci tests

* fix ci

* fix ci

* fix ci

* fix ci timeout

* fix rm progress bar. fix ci timeout

* fix ci

* fix ci typo

* remove 3d plugin from ci temporary

* test environment

* cannot save optimizer

* support chat template

* fix readme

* fix path

* test ci locally

* restore build_or_pr

* fix ci data path

* fix benchmark

* fix ci, move ci tests to 3080, disable fast tokenizer

* move ci to 85

* support flash attention 2

* add all-in-one data preparation script. Fix colossal-llama2-chat chat template

* add hardware requirements

* move ci test data

* fix save_model, add unwrap

* fix missing bos

* fix missing bos; support grad accumulation with gemini

* fix ci

* fix ci

* fix ci

* fix llama2 chat template config

* debug sft

* debug sft

* fix colossalai version requirement

* fix ci

* add sanity check to prevent NaN loss

* fix requirements

* add dummy data generation script

* add dummy data generation script

* add dummy data generation script

* add dummy data generation script

* update readme

* update readme

* update readme and ignore

* fix logger bug

* support parallel_output

* modify data preparation logic

* fix tokenization

* update lr

* fix inference

* run pre-commit

---------

Co-authored-by: Tong Li <tong.li352711588@gmail.com>

Author: YeAnbang

.github/workflows/run_chatgpt_examples.yml

@@ -19,35 +19,44 @@ jobs:
     runs-on: [self-hosted, gpu]
     container:
       image: hpcaitech/pytorch-cuda:2.1.0-12.1.0
-      options: --gpus all --rm -v /data/scratch/github_actions/chat:/data/scratch/github_actions/chat --shm-size=10.24gb
-    timeout-minutes: 30
+      options: --gpus all --rm -v /data/scratch/examples-data:/data/scratch/examples-data --shm-size=10.24gb
+    timeout-minutes: 60
     defaults:
       run:
         shell: bash
     steps:
       - name: Checkout ColossalAI
         uses: actions/checkout@v2
 
+      - name: Install Colossal-AI
+        run: |
+          BUILD_EXT=1 pip install -v -e .
+
       - name: Install ChatGPT
         run: |
-          cd applications/Chat
+          cd applications/ColossalChat
           pip install -v .
+          export BUILD_EXT=1
           pip install -r examples/requirements.txt
 
       - name: Install Transformers
         run: |
-          pip install transformers==4.30.2
+          pip install transformers==4.34.1
 
       - name: Execute Examples
         run: |
-          cd applications/Chat
+          cd applications/ColossalChat
           rm -rf ~/.cache/colossalai
-          ./tests/test_inference.sh
-          ./tests/test_benchmarks.sh
+          mkdir models
+          mkdir sft_data
+          mkdir prompt_data
+          mkdir preference_data
+          ./tests/test_data_preparation.sh
           ./tests/test_train.sh
         env:
           NCCL_SHM_DISABLE: 1
           MAX_JOBS: 8
-          SFT_DATASET: /data/scratch/github_actions/chat/data.json
-          PROMPT_DATASET: /data/scratch/github_actions/chat/prompts_en.jsonl
-          PRETRAIN_DATASET: /data/scratch/github_actions/chat/alpaca_data.json
+          PRETRAINED_MODEL_PATH: ./models
+          SFT_DATASET: ./sft_data
+          PROMPT_DATASET: ./prompt_data
+          PREFERENCE_DATASET: ./preference_data

.github/workflows/run_chatgpt_unit_tests.yml

@@ -21,7 +21,7 @@ jobs:
     runs-on: [self-hosted, gpu]
     container:
       image: hpcaitech/pytorch-cuda:2.1.0-12.1.0
-      options: --gpus all --rm -v /data/scratch/chatgpt:/data/scratch/chatgpt
+      options: --gpus all --rm -v /data/scratch/examples-data:/data/scratch/examples-data
     timeout-minutes: 30
     defaults:
       run:
@@ -32,15 +32,17 @@ jobs:
 
       - name: Install ChatGPT
         run: |
-          cd applications/Chat
+          cd applications/ColossalChat
           pip install -v .
-          pip install -r requirements-test.txt
+          pip install pytest
 
       - name: Execute Unit Testing
         run: |
-          cd applications/Chat
+          cd applications/ColossalChat
           rm -rf ~/.cache/colossalai
           pytest tests/
+          cd ./tests
+          ./test_templating.sh
         env:
           NCCL_SHM_DISABLE: 1
           MAX_JOBS: 8

.gitignore

@@ -159,3 +159,7 @@ coverage.xml
 # ignore testmon and coverage files
 .coverage
 .testmondata*
+
+# log, test files - ColossalChat
+applications/ColossalChat/logs
+applications/ColossalChat/tests/logs