7_final_code_and_metrics.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "eb6c1c46-bd53-43a4-be99-7e4a85b0f9ff",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 704,
     "status": "ok",
     "timestamp": 1726767108306,
     "user": {
      "displayName": "Juan Carlos Gómez Echevarría",
      "userId": "10113025764845079545"
     },
     "user_tz": -120
    },
    "id": "eb6c1c46-bd53-43a4-be99-7e4a85b0f9ff",
    "outputId": "f00fb05c-2dd8-4518-f850-c483ffe51d4f"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\jc\\AppData\\Roaming\\Python\\Python311\\site-packages\\tqdm\\auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The token has not been saved to the git credentials helper. Pass `add_to_git_credential=True` in this function directly or `--add-to-git-credential` if using via `huggingface-cli` if you want to set the git credential as well.\n",
      "Token is valid (permission: write).\n",
      "Your token has been saved to C:\\Users\\jc\\.cache\\huggingface\\token\n",
      "Login successful\n"
     ]
    }
   ],
   "source": [
    "from huggingface_hub import login\n",
    "hf_token = ''\n",
    "login(hf_token)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "e5c89a1b-1b2f-471e-bf81-71a98e3d1c35",
   "metadata": {
    "executionInfo": {
     "elapsed": 13417,
     "status": "ok",
     "timestamp": 1726767121721,
     "user": {
      "displayName": "Juan Carlos Gómez Echevarría",
      "userId": "10113025764845079545"
     },
     "user_tz": -120
    },
    "id": "e5c89a1b-1b2f-471e-bf81-71a98e3d1c35"
   },
   "outputs": [],
   "source": [
    "import torch\n",
    "import pickle\n",
    "import spacy\n",
    "import time\n",
    "import matplotlib.pyplot as plt\n",
    "from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline\n",
    "from sklearn.metrics.pairwise import cosine_similarity\n",
    "from datasets import load_dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "GweKrl6kqfC-",
   "metadata": {
    "executionInfo": {
     "elapsed": 2,
     "status": "ok",
     "timestamp": 1726767124937,
     "user": {
      "displayName": "Juan Carlos Gómez Echevarría",
      "userId": "10113025764845079545"
     },
     "user_tz": -120
    },
    "id": "GweKrl6kqfC-"
   },
   "outputs": [],
   "source": [
    "BASE_MODEL = \"jucargoe/fine-tuned-llama-3\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "4t38WrcYId43",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 49,
     "referenced_widgets": [
      "09da5199390a4079915baa233aa79f4d",
      "0dae19d1b7834166ac5c4071f048b4cb",
      "7b75194bb37c4be9bae59fa595ede223",
      "84dccd9123b64e89b06b57b926457bc1",
      "007faa50cbc94eb2b838749d432d1940",
      "9e63f4a6e13c47b5aebb4b4fa301e9ef",
      "68a8fe7a564040cd9a804630e2684092",
      "c0243034f60e48458ad7f65f09fc82ae",
      "5975278a9f9f42f8bc048f019ee58ebd",
      "22f3ebdbf39547a898d1b71002885141",
      "e3543eedc47d4263aad24aae557ca84a"
     ]
    },
    "executionInfo": {
     "elapsed": 89691,
     "status": "ok",
     "timestamp": 1726767214626,
     "user": {
      "displayName": "Juan Carlos Gómez Echevarría",
      "userId": "10113025764845079545"
     },
     "user_tz": -120
    },
    "id": "4t38WrcYId43",
    "outputId": "c4e3615b-79c9-477f-b4e6-af0e25911f41"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Unused kwargs: ['_load_in_4bit', '_load_in_8bit', 'quant_method']. These kwargs are not used in <class 'transformers.utils.quantization_config.BitsAndBytesConfig'>.\n",
      "Loading checkpoint shards: 100%|██████████| 2/2 [00:02<00:00,  1.48s/it]\n"
     ]
    }
   ],
   "source": [
    "model = AutoModelForCausalLM.from_pretrained(\n",
    "    BASE_MODEL,\n",
    "    torch_dtype=torch.float16,\n",
    "    device_map=\"auto\",\n",
    ")\n",
    "tokenizer = AutoTokenizer.from_pretrained(BASE_MODEL)\n",
    "nlp = spacy.load('en_core_web_md')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "dfe0469c-6808-4d7e-a37b-5daed0daf926",
   "metadata": {},
   "outputs": [],
   "source": [
    "pipe = pipeline(\n",
    "    \"text-generation\",\n",
    "    model=model,\n",
    "    tokenizer=tokenizer,\n",
    "    torch_dtype=torch.float16,\n",
    "    device_map=\"auto\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "3b9a9523-c57c-4a7a-82a9-1315bd964b34",
   "metadata": {},
   "outputs": [],
   "source": [
    "def calculate_context_of_sentence_in_corpus(corpus_vec, sentence):\n",
    "    query = nlp(sentence)\n",
    "    q_vec = query.vector\n",
    "    return cosine_similarity([corpus_vec], [q_vec])\n",
    "\n",
    "def transform_sentence(sentence):\n",
    "    sentence_transformed = []\n",
    "    nlp_sentence = nlp(sentence)\n",
    "    \n",
    "    for word in nlp_sentence:\n",
    "        if not word.is_stop and not word.is_punct and not word.like_num:\n",
    "            sentence_transformed.append(word.text)\n",
    "\n",
    "    return ' '.join(sentence_transformed)\n",
    "    \n",
    "def inference(user_input):\n",
    "    with open('sentences_vector_processed.pickle', 'rb') as file:\n",
    "        corpus_vec = pickle.load(file)\n",
    "    \n",
    "    context = calculate_context_of_sentence_in_corpus(corpus_vec, transform_sentence(user_input))\n",
    "    if context > 0.55:\n",
    "        prompt = f\"\"\"\n",
    "You are a powerful text-to-Python model.\n",
    "Your job is to answer questions about a Python.\n",
    "You are given a question regarding Python code.\n",
    "You must output the Python code that answers the question tabulated correctly.\n",
    "Only one response was allowed.\n",
    "### Input:\n",
    "{user_input}\n",
    "\n",
    "### Response:\n",
    "        \"\"\"\n",
    "        output = pipe(\n",
    "            prompt,\n",
    "            max_new_tokens=100,\n",
    "            do_sample=True,\n",
    "            temperature=0.7,\n",
    "            top_k=50,\n",
    "            top_p=0.95\n",
    "        )\n",
    "        output = output[0]['generated_text'].split('### Response:')\n",
    "        if len(output) > 1:\n",
    "            output = output[1]\n",
    "            if '###' in output:\n",
    "                output = output.split('###')[0]\n",
    "            \n",
    "            output = output.replace('```python', '')\n",
    "            output = output.replace('```', '')\n",
    "    else:\n",
    "        output = 'ERROR: The sentence you have entered is not related with Python code generation, please write a new sentence.'\n",
    "    \n",
    "    return output"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "8f31fb58-b1bb-4d81-ab79-d076d3937b47",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Using the latest cached version of the dataset since iamtarun/python_code_instructions_18k_alpaca couldn't be found on the Hugging Face Hub\n",
      "Found the latest cached dataset configuration 'default' at C:\\Users\\jc\\.cache\\huggingface\\datasets\\iamtarun___python_code_instructions_18k_alpaca\\default\\0.0.0\\7cae181e29701a8663a07a3ea43c8e105b663ba1 (last modified on Fri Oct 11 14:48:30 2024).\n"
     ]
    }
   ],
   "source": [
    "dataset = load_dataset(\"iamtarun/python_code_instructions_18k_alpaca\", split=\"train\")\n",
    "# train_dataset = dataset.train_test_split(test_size=0.2)[\"train\"]\n",
    "# eval_dataset = dataset.train_test_split(test_size=0.2)[\"test\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "a49cdd38-232f-42da-858a-d66817159573",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "==================================================================================\n",
      "Create a function to calculate the sum of a sequence of integers.\n",
      "===============EXPECTED===============\n",
      "# Python code\n",
      "def sum_sequence(sequence):\n",
      "  sum = 0\n",
      "  for num in sequence:\n",
      "    sum += num\n",
      "  return sum\n",
      "===============INFERENCE===============\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\jc\\AppData\\Roaming\\Python\\Python311\\site-packages\\transformers\\models\\llama\\modeling_llama.py:660: UserWarning: 1Torch was not compiled with flash attention. (Triggered internally at C:\\actions-runner\\_work\\pytorch\\pytorch\\builder\\windows\\pytorch\\aten\\src\\ATen\\native\\transformers\\cuda\\sdp_utils.cpp:555.)\n",
      "  attn_output = torch.nn.functional.scaled_dot_product_attention(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "         def sum(a):\n",
      "            s=0\n",
      "            for i in a:\n",
      "               s=s+i\n",
      "            return s\n",
      "\n",
      "--- 7.69370698928833 Execution time ---\n",
      "==================================================================================\n",
      "Generate a Python code for crawling a website for a specific type of data.\n",
      "===============EXPECTED===============\n",
      "import requests\n",
      "import re\n",
      "\n",
      "def crawl_website_for_phone_numbers(website):\n",
      "    response = requests.get(website)\n",
      "    phone_numbers = re.findall('\\d{3}-\\d{3}-\\d{4}', response.text)\n",
      "    return phone_numbers\n",
      "    \n",
      "if __name__ == '__main__':\n",
      "    print(crawl_website_for_phone_numbers('www.example.com'))\n",
      "===============INFERENCE===============\n",
      "\n",
      "         import requests\n",
      "         from bs4 import BeautifulSoup\n",
      "         r = requests.get(url)\n",
      "         soup = BeautifulSoup(r.text, \"html.parser\")\n",
      "         for tag in soup.find_all('a'):\n",
      "             print(tag.get('href'))\n",
      "\n",
      "\n",
      "--- 14.00899863243103 Execution time ---\n",
      "==================================================================================\n",
      "Create a Python list comprehension to get the squared values of a list [1, 2, 3, 5, 8, 13].\n",
      "===============EXPECTED===============\n",
      "[x*x for x in [1, 2, 3, 5, 8, 13]]\n",
      "===============INFERENCE===============\n",
      "\n",
      "         [1, 4, 9, 25, 64, 169]\n",
      "\n",
      "\n",
      "--- 14.28000020980835 Execution time ---\n",
      "==================================================================================\n",
      "Generate a python script to perform this action.\n",
      "===============EXPECTED===============\n",
      "def remove_duplicates(string): \n",
      "    result = \"\" \n",
      "    prev = '' \n",
      "\n",
      "    for char in string:\n",
      "        if char != prev: \n",
      "            result += char\n",
      "            prev = char\n",
      "    return result\n",
      "\n",
      "result = remove_duplicates(\"AAABBCCCD\")\n",
      "print(result)\n",
      "===============INFERENCE===============\n",
      "\n",
      "         # 1. Check if a string contains a substring\n",
      "         # Check if a string contains a substring\n",
      "         # Method 1: Use the 'in' keyword\n",
      "         # Method 2: Use the 'find' method\n",
      "         # Method 3: Use a loop to search for the substring\n",
      "\n",
      "         # 2. Check if a string starts with a substring\n",
      "         # Check if a string starts with a substring\n",
      "         # Method 1: Use the'startswith' method\n",
      "--- 14.074010610580444 Execution time ---\n",
      "==================================================================================\n",
      "Write a python script to generates random numbers between 0 and 9 that are divisible by 3.\n",
      "===============EXPECTED===============\n",
      "def generate_random_divisible_number():\n",
      "    import random\n",
      "    while True:\n",
      "        # Generate a random number\n",
      "        process = random.randint(0, 9)\n",
      "        # Check if the number is divisible by 3\n",
      "        if process % 3 == 0:\n",
      "            # If it is divisible, return it\n",
      "            return process\n",
      "===============INFERENCE===============\n",
      "\n",
      "         import random\n",
      "         random_number = random.randint(0,9)\n",
      "         if random_number % 3 == 0:\n",
      "             print(random_number)\n",
      "         else:\n",
      "             print(\"The number is not divisible by 3\")\n",
      "\n",
      "--- 9.033001899719238 Execution time ---\n",
      "==================================================================================\n",
      "Write a Python code to get the third largest element in a given row.\n",
      "===============EXPECTED===============\n",
      "def third_largest(lst):\n",
      "    if len(lst) < 3:\n",
      "        return\n",
      "    distinct = []\n",
      "    for i in lst:\n",
      "        if i not in distinct:\n",
      "            distinct.append(i)\n",
      "    distinct.sort(reverse=True)\n",
      "    return distinct[2]\n",
      "===============INFERENCE===============\n",
      "\n",
      "        \n",
      "\n",
      "\n",
      "\n",
      "a = [1,2,3,4,5,6,7,8,9,10]\n",
      "print(sorted(a)[len(a)//2])\n",
      "\n",
      "\n",
      "--- 8.389000415802002 Execution time ---\n",
      "==================================================================================\n",
      "Generate a REST API with Python and Flask that allows users to create, read, update, and delete records from a database.\n",
      "===============EXPECTED===============\n",
      "from flask import Flask, request\n",
      "from flask_sqlalchemy import SQLAlchemy\n",
      "\n",
      "app = Flask(name)\n",
      "app.config['SQLALCHEMY_DATABASE_URI'] = 'sqlite:////tmp/test.db'\n",
      "db = SQLAlchemy(app)\n",
      "\n",
      "class Record(db.Model):\n",
      " id = db.Column(db.Integer, primary_key=True)\n",
      " name = db.Column(db.String(120), unique=True)\n",
      "\n",
      "db.create_all()\n",
      "\n",
      "@app.route('/records', methods=['GET'])\n",
      "def get_records():\n",
      " records = Record.query.all()\n",
      " return {'records':[record.name for record in records]}\n",
      "\n",
      "@app.route('/records', methods=['POST'])\n",
      "def create_record():\n",
      " record = Record(name=request.json['name'])\n",
      " db.session.add(record)\n",
      " db.session.commit()\n",
      " return {'id': record.id}\n",
      "\n",
      "@app.route('/records/int:id', methods=['PUT'])\n",
      "def update_record(id):\n",
      " record = Record.query.get(id)\n",
      " record.name = request.json['name']\n",
      " db.session.commit()\n",
      " return {'message': 'Record updated'}\n",
      "\n",
      "@app.route('/records/int:id', methods=['DELETE'])\n",
      "def delete_record(id):\n",
      " Record.query.filter_by(id=id).delete()\n",
      " db.session.commit()\n",
      " return {'message': 'Record deleted'}\n",
      "\n",
      "if name == 'main':\n",
      " app.run(debug=True)\n",
      "===============INFERENCE===============\n",
      "\n",
      "         from flask import Flask, jsonify\n",
      "         from flask_sqlalchemy import SQLAlchemy\n",
      "         from flask_marshmallow import Marshmallow\n",
      "         import os\n",
      "\n",
      "         app = Flask(__name__)\n",
      "         basedir = os.path.abspath(os.path.dirname(__file__))\n",
      "         app.config['SQLALCHEMY_DATABASE_URI'] ='sqlite:///' + os.path.join(basedir, 'app.db')\n",
      "         db = SQLAlchemy(app)\n",
      "         ma = Marshmallow(app)\n",
      "\n",
      "         class Record(db.Model):\n",
      "             id = db.Column(db\n",
      "--- 14.147000312805176 Execution time ---\n",
      "==================================================================================\n",
      "Create a Python function that takes in a string and a list of words and returns true if the string contains all the words in the list.\n",
      "===============EXPECTED===============\n",
      "def contains_words(input_string, words):\n",
      " for word in words:\n",
      " if word not in input_string:\n",
      " return False\n",
      " return True\n",
      "\n",
      "print(contains_words('This is a test', ['test', 'this', 'is'])) # True\n",
      "===============INFERENCE===============\n",
      "\n",
      "         def contains_all_words(str, lst):\n",
      "            return all(word in str.split() for word in lst)\n",
      "\n",
      "--- 12.197022438049316 Execution time ---\n",
      "==================================================================================\n",
      "What should this Python program do?\n",
      "===============EXPECTED===============\n",
      "def is_palindrome(s):\n",
      "    '''This function returns True if the string s is a palindrome and False if it isn't.''' \n",
      "    return s == s[::-1]  # Checks if s is the same as its reverse.\n",
      "===============INFERENCE===============\n",
      "\n",
      "         print(\"Hello World\")\n",
      "\n",
      "--- 14.26701283454895 Execution time ---\n",
      "==================================================================================\n",
      "Create a Python program to sort and print out the elements of an array of integers.\n",
      "===============EXPECTED===============\n",
      "arr = [17, 41, 5, 22, 54, 6, 29, 3, 13]\n",
      "\n",
      "arr.sort()\n",
      "\n",
      "print(arr)\n",
      "\n",
      "# Output: [3, 5, 6, 13, 17, 22, 29, 41, 54]\n",
      "===============INFERENCE===============\n",
      "\n",
      "         def sort_and_print(x):\n",
      "             return x.sort()\n",
      "         print sort_and_print([1,2,3,4,5])\n",
      "\n",
      "--- 7.359999895095825 Execution time ---\n"
     ]
    }
   ],
   "source": [
    "flag = 10\n",
    "counter = 0\n",
    "metrics = {}\n",
    "# for el in iter(dataset):\n",
    "for el in iter(dataset):\n",
    "    instruction = el['instruction']\n",
    "    print('==================================================================================')\n",
    "    print(instruction)\n",
    "    print('===============EXPECTED===============')\n",
    "    print(el['output'])\n",
    "    print('===============INFERENCE===============')\n",
    "    start_time = time.time()\n",
    "    print(inference(instruction))\n",
    "    metrics[instruction] = time.time() - start_time\n",
    "    print(\"--- %s Execution time ---\" % (time.time() - start_time))\n",
    "    counter += 1\n",
    "    if counter == flag:\n",
    "        break"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "e9b7b263-e58a-4b66-903c-826bee8ea693",
   "metadata": {},
   "outputs": [],
   "source": [
    "x = list(metrics.keys())\n",
    "x[:] = (elem[:30] for elem in x)\n",
    "y = metrics.values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "1e88bffe-9af7-4d7e-9f47-06b46c8d5671",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "ax.bar(x, y, width=1, edgecolor=\"white\", linewidth=0.7)\n",
    "plt.xticks(rotation=-90)\n",
    "plt.title(\"Tiempo de generación (segundos) según entrada\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "84fb2f23-a8ad-4fc5-b6ce-8a0291e0f016",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Tiempo medio de generación: 11.544975423812867 segundos\n"
     ]
    }
   ],
   "source": [
    "print(f\"Tiempo medio de generación: {np.mean(list(y))} segundos\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3f451483-a521-47e1-b8ec-64af47c0ab49",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "7ecccc4d-73ea-4bc5-b138-1cb2f5dcdb39",
   "metadata": {},
   "outputs": [],
   "source": [
    "invalid_inputs = [\n",
    "    \"The stars looked brighter than ever in the clear night sky.\",\n",
    "    \"I never thought I’d find inspiration in the simplest things.\",\n",
    "    \"Her laughter echoed through the empty halls, filling the space with warmth.\",\n",
    "    \"Time flies when you're doing something you love.\",\n",
    "    \"The smell of fresh coffee in the morning is the best part of the day.\",\n",
    "    \"Sometimes the quiet moments are the most powerful.\",\n",
    "    \"He wandered through the forest, lost in thought.\",\n",
    "    \"Opportunities often come disguised as challenges.\",\n",
    "    \"The waves crashed against the shore, creating a rhythm of their own.\",\n",
    "    \"Life is like a book, and each day is a new page waiting to be written.\",\n",
    "    \"The sun set behind the mountains, painting the sky in shades of orange.\",\n",
    "    \"She held the key to his heart, though she never knew it.\",\n",
    "    \"In the silence of the night, only the crickets could be heard.\",\n",
    "    \"He smiled as he remembered the old days, full of adventure and laughter.\",\n",
    "    \"The rain tapped gently on the window, creating a soothing melody.\",\n",
    "    \"Every step she took felt like a new beginning.\",\n",
    "    \"The wind whispered secrets only the trees could understand.\",\n",
    "    \"He always believed that there was magic in the ordinary.\",\n",
    "    \"The city was alive with lights, yet she felt completely alone.\",\n",
    "    \"They say patience is a virtue, but sometimes it’s a challenge.\",\n",
    "    \"The road ahead was long, but she felt ready for the journey.\",\n",
    "    \"His dreams were bigger than the town he was born in.\",\n",
    "    \"She danced like no one was watching, lost in the music.\",\n",
    "    \"The library smelled of old books and forgotten stories.\",\n",
    "    \"Each morning brought new hope and endless possibilities.\",\n",
    "    \"His eyes sparkled with mischief, always ready for the next adventure.\",\n",
    "    \"The garden was in full bloom, bursting with color and life.\",\n",
    "    \"He could hear the ocean calling, inviting him to explore its depths.\",\n",
    "    \"The clock ticked away, but time seemed to stand still in that moment.\",\n",
    "    \"She knew that sometimes, the smallest gestures meant the most.\",\n",
    "    \"The world outside was cold, but the fireplace kept them warm and cozy.\",\n",
    "    \"A single candle flickered in the dark, casting shadows on the walls.\",\n",
    "    \"He chased his dreams like they were stars, always just out of reach.\",\n",
    "    \"The sound of laughter filled the room, making everything feel lighter.\",\n",
    "    \"She always saw beauty in the most unexpected places.\",\n",
    "    \"They stood in silence, watching the sunrise together.\",\n",
    "    \"His heart raced as he opened the letter, unsure of what it would say.\",\n",
    "    \"The mountain loomed ahead, but he was determined to reach the top.\",\n",
    "    \"She believed that everything happens for a reason, even if it's hard to see at first.\",\n",
    "    \"The wind howled through the trees, making the forest feel alive.\",\n",
    "    \"He never let fear stop him from chasing his dreams.\",\n",
    "    \"The stars twinkled above, a reminder of how vast the universe is.\",\n",
    "    \"She had a way of making even the ordinary feel extraordinary.\",\n",
    "    \"The old house creaked and groaned, full of stories untold.\",\n",
    "    \"He watched the rain fall, feeling peaceful and content.\",\n",
    "    \"The book was old and worn, but its story was timeless.\",\n",
    "    \"The river flowed quietly, reflecting the calmness of the day.\",\n",
    "    \"She found solace in the simple things, like a cup of tea and a good book.\",\n",
    "    \"The sky was a canvas, painted with clouds and sunlight.\",\n",
    "    \"He knew that sometimes, the best adventures were the ones you didn’t plan.\",\n",
    "    \"The waves lapped at the shore, a never-ending dance of water and sand.\",\n",
    "    \"She felt the weight of the world lift as she took a deep breath.\",\n",
    "    \"The forest was dense and dark, but he wasn’t afraid.\",\n",
    "    \"She cherished the little moments, like the sound of birds in the morning.\",\n",
    "    \"The road less traveled often leads to the most beautiful destinations.\",\n",
    "    \"He looked out the window, lost in his thoughts.\",\n",
    "    \"The old clock chimed, signaling the end of the day.\",\n",
    "    \"She felt free, as if nothing in the world could hold her back.\",\n",
    "    \"The journey of a thousand miles begins with a single step.\",\n",
    "    \"He knew that home wasn’t a place, but a feeling.\"\n",
    "]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "d65e900d-c8ae-424d-bef0-00987a00e584",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Total de entradas: 60\n",
      "Número de entradas inválidas no detectadas: 3\n",
      "Porcentaje de error: 5.0\n"
     ]
    }
   ],
   "source": [
    "counter = 0\n",
    "for invalid_input in invalid_inputs:\n",
    "    result = inference(invalid_input)\n",
    "    if \"ERROR\" not in result:\n",
    "        counter += 1\n",
    "\n",
    "print(f\"Total de entradas: {len(invalid_inputs)}\")\n",
    "print(f\"Número de entradas inválidas no detectadas: {counter}\")\n",
    "print(f\"Porcentaje de error: {(counter*100)/len(invalid_inputs)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "e089e4ff-dc9d-4d5a-a2a6-3b89817a0b52",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots()\n",
    "x = [\"Aciertos\", \"Fallos\"]\n",
    "y = [len(invalid_inputs)-counter, counter]\n",
    "ax.bar(x, y, width=1, edgecolor=\"white\", linewidth=0.7)\n",
    "ax.set(ylim=(0, len(invalid_inputs)))\n",
    "plt.title(\"Prueba de detección de contexto\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "db45907e-01ff-4aa9-923c-c010a4c4636e",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "accelerator": "GPU",
  "colab": {
   "gpuType": "T4",
   "provenance": []
  },
  "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": "3.11.7"
  },
  "widgets": {
   "application/vnd.jupyter.widget-state+json": {
    "007faa50cbc94eb2b838749d432d1940": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "09da5199390a4079915baa233aa79f4d": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HBoxModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HBoxModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HBoxView",
      "box_style": "",
      "children": [
       "IPY_MODEL_0dae19d1b7834166ac5c4071f048b4cb",
       "IPY_MODEL_7b75194bb37c4be9bae59fa595ede223",
       "IPY_MODEL_84dccd9123b64e89b06b57b926457bc1"
      ],
      "layout": "IPY_MODEL_007faa50cbc94eb2b838749d432d1940"
     }
    },
    "0dae19d1b7834166ac5c4071f048b4cb": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_9e63f4a6e13c47b5aebb4b4fa301e9ef",
      "placeholder": "​",
      "style": "IPY_MODEL_68a8fe7a564040cd9a804630e2684092",
      "value": "Loading checkpoint shards: 100%"
     }
    },
    "22f3ebdbf39547a898d1b71002885141": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "5975278a9f9f42f8bc048f019ee58ebd": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "ProgressStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "ProgressStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "bar_color": null,
      "description_width": ""
     }
    },
    "68a8fe7a564040cd9a804630e2684092": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    },
    "7b75194bb37c4be9bae59fa595ede223": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "FloatProgressModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "FloatProgressModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "ProgressView",
      "bar_style": "success",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_c0243034f60e48458ad7f65f09fc82ae",
      "max": 4,
      "min": 0,
      "orientation": "horizontal",
      "style": "IPY_MODEL_5975278a9f9f42f8bc048f019ee58ebd",
      "value": 4
     }
    },
    "84dccd9123b64e89b06b57b926457bc1": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "HTMLModel",
     "state": {
      "_dom_classes": [],
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "HTMLModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/controls",
      "_view_module_version": "1.5.0",
      "_view_name": "HTMLView",
      "description": "",
      "description_tooltip": null,
      "layout": "IPY_MODEL_22f3ebdbf39547a898d1b71002885141",
      "placeholder": "​",
      "style": "IPY_MODEL_e3543eedc47d4263aad24aae557ca84a",
      "value": " 4/4 [01:26&lt;00:00, 18.80s/it]"
     }
    },
    "9e63f4a6e13c47b5aebb4b4fa301e9ef": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "c0243034f60e48458ad7f65f09fc82ae": {
     "model_module": "@jupyter-widgets/base",
     "model_module_version": "1.2.0",
     "model_name": "LayoutModel",
     "state": {
      "_model_module": "@jupyter-widgets/base",
      "_model_module_version": "1.2.0",
      "_model_name": "LayoutModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "LayoutView",
      "align_content": null,
      "align_items": null,
      "align_self": null,
      "border": null,
      "bottom": null,
      "display": null,
      "flex": null,
      "flex_flow": null,
      "grid_area": null,
      "grid_auto_columns": null,
      "grid_auto_flow": null,
      "grid_auto_rows": null,
      "grid_column": null,
      "grid_gap": null,
      "grid_row": null,
      "grid_template_areas": null,
      "grid_template_columns": null,
      "grid_template_rows": null,
      "height": null,
      "justify_content": null,
      "justify_items": null,
      "left": null,
      "margin": null,
      "max_height": null,
      "max_width": null,
      "min_height": null,
      "min_width": null,
      "object_fit": null,
      "object_position": null,
      "order": null,
      "overflow": null,
      "overflow_x": null,
      "overflow_y": null,
      "padding": null,
      "right": null,
      "top": null,
      "visibility": null,
      "width": null
     }
    },
    "e3543eedc47d4263aad24aae557ca84a": {
     "model_module": "@jupyter-widgets/controls",
     "model_module_version": "1.5.0",
     "model_name": "DescriptionStyleModel",
     "state": {
      "_model_module": "@jupyter-widgets/controls",
      "_model_module_version": "1.5.0",
      "_model_name": "DescriptionStyleModel",
      "_view_count": null,
      "_view_module": "@jupyter-widgets/base",
      "_view_module_version": "1.2.0",
      "_view_name": "StyleView",
      "description_width": ""
     }
    }
   }
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}