Update pulp version (#654)

alpa-projects · Aug 16, 2022 · 9802b3b · 9802b3b
1 parent 5eff300
commit 9802b3b
Show file tree

Hide file tree

Showing 4 changed files with 7 additions and 46 deletions.
diff --git a/README.md b/README.md
@@ -18,15 +18,15 @@ The key features of Alpa include:
 
 ✨ **Tight Integration with Machine Learning Ecosystem**. Alpa is backed by open-source, high-performance, and production-ready libraries such as [Jax](https://github.com/google/jax), [XLA](https://www.tensorflow.org/xla), and [Ray](https://github.com/ray-project/ray).
 
-
-## 🌟Try Alpa-served OPT-175B!
+## 👉 Try Alpa-served OPT-175B!
 Alpa provides a free, unlimited [OPT-175B](https://github.com/facebookresearch/metaseq/tree/main/projects/OPT) text generation service. Try our service at [https://opt.alpa.ai/](https://opt.alpa.ai/) 
 and share your [prompting results](examples/opt_serving/service/img.png)!
 
 Join [Alpa slack](https://forms.gle/YEZTCrtZD6EAVNBQ7) and let us know any new features you want!
 
 ## Quick Start
 Use Alpa's decorator ``@parallelize`` to scale your single-device training code to distributed clusters.
+
 ```python
 import alpa
 
@@ -49,32 +49,6 @@ for batch in data_loader:
 
 Check out the [Alpa Documentation](https://alpa-projects.github.io) site for installation instructions, tutorials, examples, and more.
 
-## Installation
-The quickest way to get started with Alpa is via pip. We push the latest Alpa wheels to the [PyPI index](https://pypi.org/project/alpa/#description) and the Alpa-modified jaxlib to 
-our [GitHub-hosted index](https://alpa-projects.github.io/wheels.html).
-
-```bash
-# Install alpa
-pip install alpa
-
-# install alpa-jaxlib compatible with CUDA >= 11.1 and cuDNN >= 8.0.5,
-pip install jaxlib==0.3.5+cuda111.cudnn805 -f https://alpa-projects.github.io/wheels.html
-
-# You can install for other CUDA versions via:
-pip install jaxlib==0.3.5+cuda{cuda_version}.cudnn{cudnn_version} -f https://alpa-projects.github.io/wheels.html
-```
-All supported CUDA and cuDNN versions are listed on the [index page](https://alpa-projects.github.io/wheels.html).
-
-After installation, validate the installation works as expected:
-```bash
-# Start the ray cluster first
-ray start --head
-python -m alpa.test_install
-```
-
-You can also install Alpa from source for development or other CUDA versions. 
-Follow this [detailed guide](https://alpa.ai/install.html#method-2-install-from-source) to install Alpa from source or troubleshooting if you meet any errors during the process.  
-
 ## Learning more
 - [Alpa OSDI 2022 paper](https://www.usenix.org/system/files/osdi22-zheng-lianmin.pdf)
 - [Google AI blog](https://ai.googleblog.com/2022/05/alpa-automated-model-parallel-deep.html)

diff --git a/alpa/shard_parallel/auto_sharding.py b/alpa/shard_parallel/auto_sharding.py
@@ -837,17 +837,14 @@ def get_non_zero_index(binary_vector):
 
     msg = verbose
     time_limit = 600
-    assert "COIN_CMD" in pulp.listSolvers(onlyAvailable=True), (
+    assert "PULP_CBC_CMD" in pulp.listSolvers(onlyAvailable=True), (
         "Please install ILP solvers by 'sudo apt install coinor-cbc'")
 
-    with warnings.catch_warnings():  # disable CBC warnings
-        warnings.simplefilter("ignore")
-        solver = pulp.COIN_CMD(mip=True,
+    solver = pulp.PULP_CBC_CMD(mip=True,
                                msg=msg,
                                timeLimit=time_limit,
                                threads=multiprocessing.cpu_count())
-        # solver = pulp.GLPK_CMD(mip=True, msg=msg, timeLimit=time_limit)
-        prob.solve(solver)
+    prob.solve(solver)
 
     status = prob.status
     objective = pulp.value(prob.objective)

diff --git a/docs/install.rst b/docs/install.rst
@@ -13,17 +13,7 @@ Regardless of installing from wheels or from source, there are a few prerequisit
   Follow the official guides to install `CUDA <https://developer.nvidia.com/cuda-toolkit>`_ and `cuDNN <https://developer.nvidia.com/cudnn>`_.
   Alpa requires CUDA >= 11.1 and  cuDNN >= 8.0.5.
 
-2. Install the ILP solver used by Alpa:
-
-    If you have sudo permission, use
-
-      .. code:: bash
-
-        sudo apt install coinor-cbc
-
-    Otherwise, please try to install via `binary <https://projects.coin-or.org/Cbc#DownloadandInstall>`_ or `conda <https://anaconda.org/conda-forge/coincbc>`_.
-
-3. Update pip version and install cupy:
+2. Update pip version and install cupy:
 
   .. code:: bash
 

diff --git a/setup.py b/setup.py
@@ -85,7 +85,7 @@ def get_cuda_version_str(no_dot=False):
     "ray>=1.13.0",
     "jax==0.3.5",
     "flax==0.4.1",
-    "pulp",
+    "pulp>=2.6.0",
     "numpy>=1.20",
     "numba",
 ]