testing v-0.0.1

app.py

@@ -0,0 +1,49 @@
+import subprocess
+import json
+import streamlit as st
+# import time
+
+prompt_format = \
+'''Below is an instruction that describes a task, paired with an input that provides further context. Write a response that appropriately completes the request.
+
+## Instruction:
+Normalize entities in a given sentence, including dates (various formats), currencies (multiple symbols and notations), and scientific units (single and compound). Convert them into their full, standardized textual representations in the same language.
+
+### Example Input:
+15/03/1990 को, वैज्ञानिक ने $120 में 500mg यौगिक का एक नमूना खरीदा।
+
+### Example Response:
+पंद्रह मार्च उन्नीस सौ नब्बे को, वैज्ञानिक ने एक सौ बीस अमेरिकी डॉलर में पाँच सौ मिलीग्राम यौगिक का एक नमूना खरीदा।
+
+Just as entities like dates, currencies, and scientific units have been normalized into simple terms, you must do the same. Do not leave any entity un-normalised.
+
+## Input:
+{}
+
+## Response:
+{}'''
+
+prompt = "हा अहवाल 30 pages लांब आणि 10 MB आकाराचा आहे."
+
+prompt = prompt_format.format (
+    prompt, # input
+        "", # output - leave this blank for generation!
+)
+
+prompt = prompt.replace('\n','\\n')
+
+command = \
+'''curl --request POST \
+    --url http://localhost:8081/completion \
+    --header "Content-Type: application/json" \
+    --data '{"prompt": "'''+prompt+'''", "n_predict": 256}\''''
+
+# print(command)
+# Display the variable on the page
+st.write("executing command ... \n")
+result = subprocess.run(command, shell=True, capture_output=True, text=True)
+output = json.loads(result.stdout)['content']
+print(output)
+# output = "hello me tasmay!"
+# time.sleep(5)
+st.write(f"Output:\n{output}")

init.sh

@@ -0,0 +1,44 @@
+#!/bin/bash
+set -e  # Exit immediately if a command fails
+
+echo "Starting initialization of llama.cpp environment..."
+
+# Clone llama.cpp repository if not already present
+if [ ! -d "llama.cpp" ]; then
+    echo "Cloning llama.cpp repository..."
+    git clone https://github.com/ggml-org/llama.cpp
+else
+    echo "llama.cpp repository already exists. Skipping clone."
+fi
+
+cd llama.cpp
+
+# Build llama-server
+echo "Configuring build with cmake..."
+cmake -B build
+echo "Building llama-server..."
+cmake --build build --config Release -t llama-server
+
+# Download the model if not already present
+if [ ! -f "models/sarvam_entity_normalisation_llama_3.1_8b_unsloth.Q4_K_M.gguf" ]; then
+    echo "Model not found. Downloading the model..."
+    cd models
+    wget https://huggingface.co/Tasmay-Tib/sarvam-entity-normalisation-llama-3.1-8b-gguf/resolve/main/sarvam_entity_normalisation_llama_3.1_8b_unsloth.Q4_K_M.gguf
+    cd ..
+else
+    echo "Model already exists. Skipping download."
+fi
+
+# Launch llama-server in the background
+echo "Starting llama-server in the background..."
+cd build
+./bin/llama-server \
+    -m ../llama.cpp/models/sarvam_entity_normalisation_llama_3.1_8b_unsloth.Q4_K_M.gguf \
+    -n 256 -c 1024 -t 2 -b 1 \
+    --temp 0.1 --repeat-penalty 1.1 --top-k 20 \
+    --port 8081 --mlock --numa numactl &
+
+echo "llama-server launched. Waiting for the server to initialize..."
+# (Optional) Wait a few seconds for the server to start before launching Streamlit
+sleep 5
+echo "Initialization complete. Proceeding with Streamlit app startup..."

requirements.txt

@@ -0,0 +1,4 @@
+requests
+json
+os
+subprocess