Polish LlamaDemo README documentation (#128)

Author: Copilot

llm/android/LlamaDemo/README.md

@@ -103,15 +103,15 @@ mModule = new LlmModule(
 int loadResult = mModule.load();
 ```
 
-* `modelCategory`: Indicate whether it’s a text-only or vision model
-* `modePath`: path to the .pte file
-* `tokenizerPath`: path to the tokenizer file
-* `temperature`: model parameter to adjust the randomness of the model’s output
-* `dataPath`: path to one or a list of .ptd files
+* `modelCategory`: Indicates whether it’s a text-only or vision model
+* `modelPath`: Path to the .pte file
+* `tokenizerPath`: Path to the tokenizer file
+* `temperature`: Model parameter to adjust the randomness of the model’s output
+* `dataPath`: Path to one or a list of .ptd files
 
 
 ### User Prompt
-Once model is successfully loaded then enter any prompt and click the send (i.e. generate) button to send it to the model.
+Once the model is successfully loaded, enter any prompt and click the send (i.e., generate) button to send it to the model.
 <p align="center">
 <img src="https://raw.githubusercontent.com/pytorch/executorch/refs/heads/main/docs/source/_static/img/load_complete_and_start_prompt.png" style="width:300px">
 </p>
@@ -124,11 +124,11 @@ You can provide it more follow-up questions as well.
 #### ExecuTorch App API
 
 ```java
-mModule.generate(prompt,sequence_length, MainActivity.this);
+mModule.generate(prompt, sequence_length, MainActivity.this);
 ```
-* `prompt`: User formatted prompt
+* `prompt`: User-formatted prompt
 * `sequence_length`: Number of tokens to generate in response to a prompt
-* `MainActivity.this`: Indicate that the callback functions (OnResult(), OnStats()) are present in this class.
+* `MainActivity.this`: Indicates that the callback functions (`onResult()`, `onStats()`) are present in this class.
 
 [*LLaVA-1.5: Only for XNNPACK delegate*]
 
@@ -189,7 +189,7 @@ Ensure you have the following functions in your callback class that you provided
 ```
 
 ## Instrumentation Test
-You can run the instrumentation test for sanity check. The test loads a model pte file and tokenizer.bin file
+You can run the instrumentation test for a sanity check. The test loads a model .pte file and tokenizer.bin file
 under `/data/local/tmp/llama`.
 
 ### Model preparation
@@ -216,4 +216,4 @@ adb push tokenizer.bin /data/local/tmp/llama
 ```
 
 ## Reporting Issues
-If you encountered any bugs or issues following this tutorial please file a bug/issue here on [Github](https://github.com/pytorch/executorch/issues/new), or join our discord [here](https://lnkd.in/gWCM4ViK).
+If you encountered any bugs or issues following this tutorial, please file a bug/issue here on [GitHub](https://github.com/pytorch/executorch/issues/new), or join our Discord [here](https://lnkd.in/gWCM4ViK).

llm/android/LlamaDemo/SDK-quick-setup-guide.md

@@ -1,12 +1,12 @@
-# Guide to set up Java/SDK/NDK for Android
+# Guide to Set Up Java/SDK/NDK for Android
 
-Follow this doc if you haven't set up Java/SDK/NDK for Android development
+Follow this guide if you haven't set up Java/SDK/NDK for Android development
 already.
-This doc provides a CLI tutorial to set them up. Otherwise, you can do the same
-thing with Android Studio GUI.
+This guide provides a CLI tutorial to set them up. Alternatively, you can do the same
+thing with the Android Studio GUI.
 
-## Set up Java 17
-1. Download the archive from Oracle website.
+## Set Up Java 17
+1. Download the archive from the Oracle website.
 Make sure you have read and agree with the terms and conditions from the website before downloading.
 ```bash
 export DEV_HOME=<path-to-dev>
@@ -40,14 +40,14 @@ export PATH="$JAVA_HOME/bin:$PATH"
 Note: Oracle has tutorials for installing Java on
 [Linux](https://docs.oracle.com/en/java/javase/17/install/installation-jdk-linux-platforms.html#GUID-4A6BD592-1840-4BB4-A758-4CD49E9EE88B)
 and [macOS](https://docs.oracle.com/en/java/javase/17/install/installation-jdk-macos.html#GUID-E8A251B6-D9A9-4276-ABC8-CC0DAD62EA33).
-Some Linux distributions has JDK package in package manager. For example, Debian users can install
-openjdk-17-jdk package.
+Some Linux distributions have a JDK package in the package manager. For example, Debian users can install
+the openjdk-17-jdk package.
 
-## Set up Android SDK/NDK
-Android has a command line tool [sdkmanager](https://developer.android.com/tools/sdkmanager) which
-helps users managing SDK and other tools related to Android development.
+## Set Up Android SDK/NDK
+Android has a command-line tool [sdkmanager](https://developer.android.com/tools/sdkmanager) which
+helps users manage SDK and other tools related to Android development.
 
-1. Go to https://developer.android.com/studio and download the archive from "Command line tools
+1. Go to https://developer.android.com/studio and download the archive from the "Command line tools
 only" section. Make sure you have read and agree with the terms and conditions from the website.
 
 Linux:
@@ -58,7 +58,7 @@ macOS:
 ```bash
 curl https://dl.google.com/android/repository/commandlinetools-mac-11076708_latest.zip -o commandlinetools.zip
 ```
-2. Unzip.
+2. Unzip:
 ```bash
 unzip commandlinetools.zip
 ```
@@ -80,7 +80,7 @@ If you want to use Android Studio and never set up Java/SDK/NDK before, or if
 you use the newly installed ones, follow these steps to set Android Studio to use
 them.
 
-Copy these output paths to be used by Android Studio
+Copy these output paths to be used by Android Studio:
 ```bash
 echo $ANDROID_HOME
 echo $ANDROID_NDK

llm/android/LlamaDemo/docs/delegates/mediatek_README.md

@@ -1,27 +1,27 @@
-# Building ExecuTorch Android Demo for Llama running MediaTek
-This tutorial covers the end to end workflow for running Llama 3-8B-instruct inference on MediaTek AI accelerators on an Android device.
+# Building ExecuTorch Android Demo for Llama Running MediaTek
+This tutorial covers the end-to-end workflow for running Llama 3-8B-instruct inference on MediaTek AI accelerators on an Android device.
 More specifically, it covers:
 1. Export and quantization of Llama models against the MediaTek backend.
-2. Building and linking libraries that are required to inference on-device for Android platform using MediaTek AI accelerators.
+2. Building and linking libraries that are required for on-device inference on the Android platform using MediaTek AI accelerators.
 3. Loading the needed model files on the device and using the Android demo app to run inference.
 
-Verified on MacOS, Linux CentOS (model export), Python 3.10, Android NDK 26.3.11579264
-Phone verified: MediaTek Dimensity 9300 (D9300) chip.
+Verified on macOS, Linux CentOS (model export), Python 3.10, Android NDK 26.3.11579264.
+Phones verified: MediaTek Dimensity 9300 (D9300) chip.
 
 ## Prerequisites
 * Download and link the Android NDK, and MediaTek ExecuTorch Libraries from the MediaTek Backend Readme ([link](https://github.com/pytorch/executorch/tree/main/backends/mediatek/scripts#prerequisites)).
 * MediaTek Dimensity 9300 (D9300) chip device
-* Desired Llama 3 model weights. You can download them on HuggingFace [Example](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct)).
+* Desired Llama 3 model weights. You can download them on Hugging Face ([Example](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct)).
 * Download NeuroPilot Express SDK from the [MediaTek NeuroPilot Portal](https://neuropilot.mediatek.com/resources/public/npexpress/en/docs/npexpress):
   - `libneuronusdk_adapter.mtk.so`: This universal SDK contains the implementation required for executing target-dependent code on the MediaTek chip.
   - `libneuron_buffer_allocator.so`: This utility library is designed for allocating DMA buffers necessary for model inference.
-  - `mtk_converter-8.8.0.dev20240723+public.d1467db9-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl`: This library preprocess the model into a MediaTek representation.
+  - `mtk_converter-8.8.0.dev20240723+public.d1467db9-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl`: This library preprocesses the model into a MediaTek representation.
   - `mtk_neuron-8.2.2-py3-none-linux_x86_64.whl`: This library converts the model to binaries.
 
 ## Setup ExecuTorch
 In this section, we will need to set up the ExecuTorch repo first with Conda environment management. Make sure you have Conda available in your system (or follow the instructions to install it [here](https://anaconda.org/anaconda/conda)). The commands below are running on Linux (CentOS).
 
-Checkout ExecuTorch repo and sync submodules
+Checkout ExecuTorch repo and sync submodules:
 
 ```
 git clone -b viable/strict https://github.com/pytorch/executorch.git && cd executorch
@@ -39,7 +39,7 @@ Or a Conda environment:
 conda create -n et_xnnpack python=3.10.0 && conda activate et_xnnpack
 ```
 
-Install dependencies
+Install dependencies:
 ```
 ./install_executorch.sh
 ```
@@ -68,11 +68,11 @@ pip3 install mtk_neuron-8.2.2-py3-none-linux_x86_64.whl
 pip3 install mtk_converter-8.8.0.dev20240723+public.d1467db9-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
 ```
 
-This was tested with transformers version 4.40 and numpy version 1.23. If you do not have these version then, use the following commands:
+This was tested with transformers version 4.40 and numpy version 1.23. If you do not have these versions, use the following commands:
 ```
 pip install transformers==4.40
 
-pip install numpy=1.23
+pip install numpy==1.23
 ```
 
 ### Running Export
@@ -85,20 +85,20 @@ cd examples/mediatek
 source shell_scripts/export_llama.sh llama3 "" "" "" alpaca.txt
 ```
 
-There will be 3 main set of files generated:
-* num_chunks*2 pte files: half are for prompt and the other half are for generation. Generation pte files are denoted by “1t” in the file name.
+There will be 3 main sets of files generated:
+* num_chunks*2 .pte files: half are for prompt and the other half are for generation. Generation .pte files are denoted by “1t” in the file name.
 * Token embedding bin file: located in the weights folder where `config.json` is placed (`examples/mediatek/modes/llm_models/weight/<model_name>/embedding_<model_name>_fp32.bin`)
 * Tokenizer file: `tokenizer.model` file
 
-Note: Exporting model flow can take 2.5 hours (114GB RAM for num_chunks=4) to complete. (Results may vary depending on hardware)
+Note: The exporting model flow can take 2.5 hours (114GB RAM for num_chunks=4) to complete. Results may vary depending on hardware.
 
-Before continuing forward, make sure to modify the tokenizer, token embedding, and model paths in the  examples/mediatek/executor_runner/run_llama3_sample.sh.
+Before continuing forward, make sure to modify the tokenizer, token embedding, and model paths in the `examples/mediatek/executor_runner/run_llama3_sample.sh` file.
 
 ### Deploy
 First, make sure your Android phone’s chipset version is compatible with this demo (MediaTek Dimensity 9300 (D9300)) chip. Once you have the model, tokenizer, and runner generated ready, you can push them and the .so files to the device before we start running using the runner via shell.
 
 ```
-adb shell mkdir -p /data/local/tmp/et-mtk/ (or any other directory name)
+adb shell mkdir -p /data/local/tmp/et-mtk/  # or any other directory name
 adb push embedding_<model_name>_fp32.bin /data/local/tmp/et-mtk
 adb push tokenizer.model /data/local/tmp/et-mtk
 adb push <exported_prompt_model_0>.pte /data/local/tmp/et-mtk
@@ -113,13 +113,13 @@ adb push <exported_gen_model_n>.pte /data/local/tmp/et-mtk
 
 ## Populate Model Paths in Runner
 
-The Mediatek runner (`examples/mediatek/executor_runner/mtk_llama_runner.cpp`) contains the logic for implementing the function calls that come from the Android app.
+The MediaTek runner (`examples/mediatek/executor_runner/mtk_llama_runner.cpp`) contains the logic for implementing the function calls that come from the Android app.
 
-**Important!** Currently the model paths are set in the runner-level. Modify the values in `examples/mediatek/executor_runner/llama_runner/llm_helper/include/llama_runner_values.h` to set the model paths, tokenizer path, embedding file path, and other metadata.
+**Important!** Currently the model paths are set at the runner level. Modify the values in `examples/mediatek/executor_runner/llama_runner/llm_helper/include/llama_runner_values.h` to set the model paths, tokenizer path, embedding file path, and other metadata.
 
 
 ## Build AAR Library
-1. Open a terminal window and navigate to the root directory of the executorch
+1. Open a terminal window and navigate to the root directory of the ExecuTorch:
 ```sh
 cd $EXECUTORCH_ROOT
 ```
@@ -142,14 +142,14 @@ export BUILD_AAR_DIR=aar-out
 sh scripts/build_android_library.sh
 ```
 
-5. Now go to demo app root (containing main README.md) and copy the AAR to the app:
+5. Now go to the demo app root (containing the main README.md) and copy the AAR to the app:
 ```sh
 cd $DEMO_APP
 mkdir -p app/libs
 cp $EXECUTORCH_ROOT/aar-out/executorch.aar app/libs/executorch.aar
 ```
 
-If you were to unzip the .aar file or open it in Android Studio, verify it contains the following related to MediaTek backend:
+If you were to unzip the .aar file or open it in Android Studio, verify it contains the following files related to the MediaTek backend:
 * libneuron_buffer_allocator.so
 * libneuronusdk_adapter.mtk.so
 * libneuron_backend.so (generated during build)
@@ -160,14 +160,14 @@ If you were to unzip the .aar file or open it in Android Studio, verify it conta
 1. Open Android Studio and select “Open an existing Android Studio project” to open LlamaDemo.
 2. Run the app (^R). This builds and launches the app on the phone.
 
-### Alternative 2: Command line
-Without Android Studio UI, we can run gradle directly to build the app. We need to set up the Android SDK path and invoke gradle.
+### Alternative 2: Command Line
+Without Android Studio UI, we can run Gradle directly to build the app. We need to set up the Android SDK path and invoke Gradle.
 ```
 export ANDROID_HOME=<path_to_android_sdk_home>
 cd LlamaDemo
 ./gradlew :app:installDebug
 ```
-If the app successfully run on your device, you should see something like below:
+If the app successfully runs on your device, you should see something like the screenshot below:
 
 <p align="center">
 <img src="https://raw.githubusercontent.com/pytorch/executorch/refs/heads/main/docs/source/_static/img/opening_the_app_details.png" style="width:800px">
@@ -179,4 +179,4 @@ Once you've loaded the app on the device:
 3. Click the "Load Model" button. This will load the models from the Runner
 
 ## Reporting Issues
-If you encountered any bugs or issues following this tutorial please file a bug/issue here on [Github](https://github.com/pytorch/executorch/issues/new).
+If you encountered any bugs or issues following this tutorial, please file a bug/issue here on [GitHub](https://github.com/pytorch/executorch/issues/new).