Update README.md

README.md

@@ -1,90 +1,129 @@
 ---
-license: apache-2.0
-datasets:
-- cerebras/SlimPajama-627B
-- bigcode/starcoderdata
-- OpenAssistant/oasst_top1_2023-08-25
-language:
-- en
-tags:
-- GGUF
-- llamafile
-model_creator: TinyLlama
-model_name: TinyLlama-1.1B-Chat v1.0
-model_type: Pythia
-quantized_by: jartine
+slug: access-astra
+title: Access Astra Board
+authors: [asahu]
+tags: [connect,access]
 ---
 
-# TinyLlama-1.1B-Chat v1.0 w/ GGUF + llamafile
+# Guide to access Astra Machina Dev Board
+This tutorial will guide you through the steps required to access a Astra Machina Dev Board.
 
-- Model creator: [TinyLlama](https://huggingface.co/TinyLlama)
-- Original model: [TinyLlama-1.1B-Chat-v1.0](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v1.0)
+We can access our Astra Machina Board using:  
 
-<!-- description start -->
-## Description
 
-This repo contains both:
+**1. Using ADB on Host Machine** 
 
-- Prebuilt llamafiles for each quantization format that can be executed to launch a web server or cli interface
+**2. Using Tera Term / Putty on Host Machine** 
 
-- GGUF weights data files for each quantization format, which require either the [llamafile](https://github.com/mozilla-Ocho/llamafile) or [llama.cpp](https://github.com/ggerganov/llama.cpp) software to run
+**3. Using Wayland Terminal on Astra Board**
 
-## Prompt Template: ChatML
+## 1. ADB 
 
+### Connect to your board
+
+To communicate with your board via Android Debug Bridge (ADB) first connect a USB cable from your dev machine to the USB C port on the board marked "USB-2.0". 
+
+![board](https://synaptics-astra.github.io/doc/v/0.9.0/_images/usb-c.png)
+
+:::warning
+The other USB-C marked PWR_IN should be connected to the power supply - make sure you get them the right way around!
+:::
+
+
+If you don't have ADB,it can be installed as part of [Android SDK Platform-Tools](https://developer.android.com/tools/releases/platform-tools#downloads). You should then be able to open a MacOS, Linux or Windows shell inside folder `platform-tools` and run:
+
+```bash  
+adb devices
 ```
-<|im_start|>system
-{system_message}<|im_end|>
-<|im_start|>user
-{prompt}<|im_end|>
-<|im_start|>assistant
+
+You can check if the Synaptics Astra board is attached:
+
+<pre>List of devices attached
+SL16x0	device</pre>
+
+See the [user guide](https://synaptics-astra.github.io/doc/v/0.9.0/linux/index.html#the-shell-with-adb) for more information on ADB.
+
+
+### Copy across the sample image
+
+First, copy the example image from your dev machine to the board:
+
+```
+adb push ~/Downloads/expanse.jpg /home/weston/expanse.jpg
 ```
 
----
+### Start a shell session
 
-# TinyLlama-1.1B
-</div>
+Then start a shell session:
 
-https://github.com/jzhang38/TinyLlama
+```
+adb shell
+hostname
+```
 
-The TinyLlama project aims to **pretrain** a **1.1B Llama model on 3 trillion tokens**. With some proper optimization, we can achieve this within a span of "just" 90 days using 16 A100-40G GPUs 🚀🚀. The training has started on 2023-09-01. 
+You should see the prompt has changed and the board responds with a hostname:
 
+<pre>
+sh-5.1# hostname
+sl1680
+</pre>
 
-We adopted exactly the same architecture and tokenizer as Llama 2. This means TinyLlama can be plugged and played in many open-source projects built upon Llama. Besides, TinyLlama is compact with only 1.1B parameters. This compactness allows it to cater to a multitude of applications demanding a restricted computation and memory footprint.
 
-#### This Model
-This is the chat model finetuned on top of [TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T](https://huggingface.co/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T). **We follow [HF's Zephyr](https://huggingface.co/HuggingFaceH4/zephyr-7b-alpha/edit/main/README.md)'s training recipe.** The model was " initially fine-tuned on a variant of the [`UltraChat`](https://huggingface.co/datasets/stingning/ultrachat) dataset, which contains a diverse range of synthetic dialogues generated by ChatGPT. 
-We then further aligned the model with [🤗 TRL's](https://github.com/huggingface/trl) `DPOTrainer` on the [openbmb/UltraFeedback](https://huggingface.co/datasets/openbmb/UltraFeedback) dataset, which contain 64k prompts and model completions that are ranked by GPT-4." 
+## 2. Tera Term / Putty
 
+### Connect to your board
 
-#### How to use
-You will need the transformers>=4.34
-Do check the [TinyLlama](https://github.com/jzhang38/TinyLlama) github page for more information.
+To communicate with your board via Tera Term or Putty first connect a **UART** from your dev machine to the GPIO Header of the board following this image:
 
-```python
-# Install transformers from source - only needed for versions <= v4.34
-# pip install git+https://github.com/huggingface/transformers.git
-# pip install accelerate
+**VDD -> Pin 1 , GND -> Pin 6 , TX  -> Pin 8 and RX  -> Pin 10**.
 
-import torch
-from transformers import pipeline
 
-pipe = pipeline("text-generation", model="TinyLlama/TinyLlama-1.1B-Chat-v1.0", torch_dtype=torch.bfloat16, device_map="auto")
+![image](https://synaptics-astra.github.io/doc/v/1.0.0/_images/40-pin-connector.png )
+
+
+### Tera Term settings
+
+Open Tera Term: New connection then Select Serial -> Port: 'COM$: Silicon Labs CP210x USB to UART Bridge' 
+
+Go to 'Setup' and Open 'Serial Port'
+Then set 'Speed' -> '115200' and Click 'New Open'
+
+
+### Copy across the sample image
+
+Once in Tera Term is setup, Get the Inet address of your Astra Machina Board from here.
+<!-- ```bash title="ADB_Terminal" class="no-copy-prompt"
+sh-5.1# ifconfig eth0 | grep "inet addr" -->
+```bash
+ifconfig eth0 | grep "inet addr"
+``` 
+Copy the example image from your dev machine to the board using `scp ` to Machina board:
+
+```bash 
+scp ~/Downloads/expanse.jpg [email protected]:/home/root
+``` 
+
+
+## 3. Wayland Terminal 
+### Connect to your board
+
+Connect the Power supply in USB-C marked **PWR_IN**.  
+
+Astra Machina’s graphical desktop is enabled be default. It can be displayed on an external display connected to the HDMI port or a MIPI display. 
+
+
+Input can be provided by connecting a standard USB keyboard and mouse.
+
+![image](https://synaptics-astra.github.io/doc/v/0.9.0/_images/board-ports.png)
+
+:::warning
+HDMI is currently not supported on SL1620
+:::
+
+### Wayland Desktop
+
+Clicking on the icon in the top left corner will open a terminal.
+
+![image](https://synaptics-astra.github.io/doc/v/0.9.0/_images/wayland-terminal.jpg)
+
 
-# We use the tokenizer's chat template to format each message - see https://huggingface.co/docs/transformers/main/en/chat_templating
-messages = [
-    {
-        "role": "system",
-        "content": "You are a friendly chatbot who always responds in the style of a pirate",
-    },
-    {"role": "user", "content": "How many helicopters can a human eat in one sitting?"},
-]
-prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
-outputs = pipe(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
-print(outputs[0]["generated_text"])
-# <|system|>
-# You are a friendly chatbot who always responds in the style of a pirate.</s>
-# <|user|>
-# How many helicopters can a human eat in one sitting?</s>
-# <|assistant|>
-# ...
-```