{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "import requests\n",
    "import json\n",
    "from urllib.request import urlretrieve\n",
    "import pandas as pd\n",
    "import time\n",
    "from allkeys import OPENAIKEY, GEMENIKEY"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import anvil.server\n",
    "anvil.server.connect('PLMOIU5VCGGUOJH2XORIBWV3-ZXZVFLWX7QFIIAF4')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def fetch_result(task_id):\n",
    "    while True:\n",
    "        result=anvil.server.call('poll',task_id)\n",
    "        if result!='In Progress' or result=='No such task': break\n",
    "        else: \n",
    "            time.sleep(1)\n",
    "            print(result)\n",
    "    print(result)\n",
    "    return result"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "text='write a python function to compute the nth digit of pi'\n",
    "model='gpt-3.5-turbo'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "task_id=anvil.server.call('launch','call_gemini',text,GEMENIKEY)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "task_id=anvil.server.call('launch','call_gpt',text,OPENAIKEY,model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "fetch_result(task_id)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(result)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(result[1],end='\\n')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pathlib\n",
    "import textwrap\n",
    "from IPython.display import display\n",
    "from IPython.display import Markdown\n",
    "\n",
    "def to_markdown(text):\n",
    "  text = text.replace('•', '  *')\n",
    "  return Markdown(textwrap.indent(text, '> ', predicate=lambda _: True))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "prompt='write code that defines a transformer network from scratch in pytorch'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "response=anvil.server.call('call_gemini',prompt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "anvil.server.call('encode_anvil',prompt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "to_markdown(response)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "anvil.server.call('encode_anvil','I am a robot')[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def encode(text,server='local'):\n",
    "    headers = {'Content-Type': 'application/json'}\n",
    "    if server=='local': url='http://127.0.0.1:7860/encode'\n",
    "    elif server=='hf': url='https://huggingface.co/spaces/gmshroff/gmserver/encode'\n",
    "    body={'text':text}\n",
    "    response=requests.post(url=url,data=json.dumps(body),headers = {'Content-Type': 'application/json'})\n",
    "    return response\n",
    "    return json.loads(response.content)['embedding']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "response=encode('I am a robot',server='local')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "response.content"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "headers = {'Content-Type': 'application/json'}\n",
    "# url='http://127.0.0.1:5000/run'\n",
    "url='https://huggingface.co/spaces/gmshroff/gmserver/'\n",
    "# url='http://127.0.0.1:7860/run'\n",
    "# body={\"script\":\"python update_valdata.py\"}\n",
    "# body={\"script\":\"pwd\"}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "response=requests.get(url=url)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "response.content"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# url='http://127.0.0.1:7860/encode'\n",
    "body={'text':'I am very good'}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "response=requests.post(url=url,data=json.dumps(body),headers = {'Content-Type': 'application/json'})\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "url"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(response)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(response.__dict__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(json.loads(response.content)['embedding'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "urlretrieve(url='http://127.0.0.1:7860/file/data.csv',filename='./returned_file.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df=pd.read_parquet('/tmp/validation_subset_int8.parquet')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "\n",
    "class Transformer(nn.Module):\n",
    "    def __init__(self, d_model, nhead, num_encoder_layers, num_decoder_layers, dim_feedforward, dropout=0.1):\n",
    "        super(Transformer, self).__init__()\n",
    "        self.transformer = nn.Transformer(d_model, nhead, num_encoder_layers, num_decoder_layers, dim_feedforward, dropout)\n",
    "\n",
    "    def forward(self, src, tgt):\n",
    "        output = self.transformer(src, tgt)\n",
    "        return output\n",
    "\n",
    "# Example usage:\n",
    "# Define the model parameters\n",
    "d_model = 512\n",
    "nhead = 8\n",
    "num_encoder_layers = 6\n",
    "num_decoder_layers = 6\n",
    "dim_feedforward = 2048\n",
    "dropout = 0.1\n",
    "\n",
    "# Initialize the model\n",
    "model = Transformer(d_model, nhead, num_encoder_layers, num_decoder_layers, dim_feedforward, dropout)\n",
    "\n",
    "# Generate some sample data\n",
    "src = torch.rand(10, 32, 512)\n",
    "tgt = torch.rand(20, 32, 512)\n",
    "\n",
    "# Pass the data through the model\n",
    "output = model(src, tgt)\n",
    "\n",
    "# Print the output shape\n",
    "print(output.shape)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}