Fix all docs for accelerate install directions (#17145)

docs/source/en/run_scripts.mdx

@@ -189,8 +189,9 @@ python run_summarization.py  \
 
 🤗 [Accelerate](https://huggingface.co/docs/accelerate/index.html) is a PyTorch-only library that offers a unified method for training a model on several types of setups (CPU-only, multiple GPUs, TPUs) while maintaining complete visibility into the PyTorch training loop. Make sure you have 🤗 Accelerate installed if you don't already have it:
 
+> Note: As Accelerate is rapidly developing, the git version of accelerate must be installed to run the scripts
 ```bash
-pip install accelerate
+pip install git+https://github.com/huggingface/accelerate
 ```
 
 Instead of the `run_summarization.py` script, you need to use the `run_summarization_no_trainer.py` script. 🤗 Accelerate supported scripts will have a `task_no_trainer.py` file in the folder. Begin by running the following command to create and save a configuration file:

examples/pytorch/image-classification/README.md

@@ -177,7 +177,7 @@ the means of the [🤗 `Accelerate`](https://github.com/huggingface/accelerate)
 after installing it:
 
 ```bash
-pip install accelerate
+pip install git+https://github.com/huggingface/accelerate
 ```
 
 You can then use your usual launchers to run in it in a distributed environment, but the easiest way is to run

examples/pytorch/multiple-choice/README.md

@@ -53,7 +53,7 @@ the mean of the [🤗 `Accelerate`](https://github.com/huggingface/accelerate) l
 after installing it:
 
 ```bash
-pip install accelerate
+pip install git+https://github.com/huggingface/accelerate
 ```
 
 then

examples/pytorch/question-answering/README.md

@@ -136,7 +136,7 @@ SQuAD or a similar dataset, the main difference is that this script exposes the
 You can use the script normally after installing it:
 
 ```bash
-pip install accelerate
+pip install git+https://github.com/huggingface/accelerate
 ```
 
 then

examples/pytorch/summarization/README.md

@@ -149,7 +149,7 @@ the mean of the [🤗 `Accelerate`](https://github.com/huggingface/accelerate) l
 after installing it:
 
 ```bash
-pip install accelerate
+pip install git+https://github.com/huggingface/accelerate
 ```
 
 then

examples/pytorch/text-classification/README.md

@@ -115,7 +115,7 @@ the mean of the [🤗 `Accelerate`](https://github.com/huggingface/accelerate) l
 after installing it:
 
 ```bash
-pip install accelerate
+pip install git+https://github.com/huggingface/accelerate
 ```
 
 then

examples/pytorch/token-classification/README.md

@@ -73,7 +73,7 @@ the mean of the [🤗 `Accelerate`](https://github.com/huggingface/accelerate) l
 after installing it:
 
 ```bash
-pip install accelerate
+pip install git+https://github.com/huggingface/accelerate
 ```
 
 then

examples/pytorch/translation/README.md

@@ -162,7 +162,7 @@ the mean of the [🤗 `Accelerate`](https://github.com/huggingface/accelerate) l
 after installing it:
 
 ```bash
-pip install accelerate
+pip install git+https://github.com/huggingface/accelerate
 ```
 
 then

examples/research_projects/luke/README.md

@@ -14,7 +14,7 @@ the mean of the [🤗 `Accelerate`](https://github.com/huggingface/accelerate) l
 after installing it:
 
 ```bash
-pip install accelerate
+pip install git+https://github.com/huggingface/accelerate
 ```
 
 then to train English LUKE on CoNLL2003: