noshot 0.1.7__py3-none-any.whl → 0.1.8__py3-none-any.whl

noshot/data/ML TS XAI/NLP/NLP 5/5.Forward and Backward.ipynb

@@ -1,149 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "id": "d0c4e603-f344-4cc6-bed7-870a28b98725",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "states = ['cp', 'ip']\n",
-    "ip = [1.0, 0.0]\n",
-    "stp = [[0.7, 0.3], [0.5, 0.5]]\n",
-    "op = ['lem', 'ice', 'cola']\n",
-    "opp = [[0.3, 0.1, 0.6], [0.2, 0.7, 0.1]]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "id": "674dac88-a26e-4af4-93cf-d8467f98a872",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def forward_alg():\n",
-    "    alpha = []\n",
-    "    for _ in range(len(op)+1):\n",
-    "        alpha.append([0,0])\n",
-    "\n",
-    "    '''alpha[0][0] = 1.0\n",
-    "    alpha[0][1] = 0.0\n",
-    "    #print(alpha)'''\n",
-    "    \n",
-    "    for i in range(len(alpha)):\n",
-    "        if i == 0:\n",
-    "            alpha[i][0] = ip[0]\n",
-    "            alpha[i][1] = ip[1]\n",
-    "            continue\n",
-    "        \n",
-    "        alpha[i][0] = float(float(stp[0][0]*opp[0][i-1]*alpha[i-1][0]) + float(stp[1][0]*opp[1][i-1]*alpha[i-1][1]))\n",
-    "\n",
-    "        alpha[i][1] = float(float(stp[1][1]*opp[1][i-1]*alpha[i-1][1]) + float(stp[0][1]*opp[0][i-1]*alpha[i-1][0]))\n",
-    "\n",
-    "    return alpha"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "id": "da4ffe33-f516-40ae-a721-0afee9db9f31",
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "[1.0, 0.0]\n",
-      "[0.21, 0.09]\n",
-      "[0.0462, 0.0378]\n",
-      "[0.021293999999999997, 0.010206]\n",
-      "\n",
-      "Probability of whole sequence :  0.0315\n"
-     ]
-    }
-   ],
-   "source": [
-    "alpha = forward_alg()\n",
-    "pos_alpha = 0.0\n",
-    "for i in alpha:\n",
-    "    pos_alpha = sum(i)\n",
-    "    print(i)\n",
-    "    \n",
-    "print(\"\\nProbability of whole sequence : \",pos_alpha)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 4,
-   "id": "81593ceb-34b6-4dff-82f3-d04dd379d8a5",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def backward_alg():\n",
-    "    beta = []\n",
-    "    for _ in range(len(op)+1):\n",
-    "        beta.append([0.0,0.0])\n",
-    "\n",
-    "    beta[len(op)][0] = 1.0\n",
-    "    beta[len(op)][1] = 1.0\n",
-    "    #print(beta)\n",
-    "\n",
-    "    for i in range(len(op)-1,-1,-1):\n",
-    "        \n",
-    "        beta[i][0] = float(float(stp[0][0]*opp[0][i]*beta[i+1][0]) + float(stp[0][1]*opp[0][i]*beta[i+1][1]))\n",
-    "\n",
-    "        beta[i][1] = float(float(stp[1][1]*opp[1][i]*beta[i+1][1]) + float(stp[1][0]*opp[1][i]*beta[i+1][0]))\n",
-    "\n",
-    "    return beta"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 5,
-   "id": "d982baed-e8e7-4abf-935c-f2f50ac3e4bc",
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "[0.0315, 0.029]\n",
-      "[0.045, 0.245]\n",
-      "[0.6, 0.1]\n",
-      "[1.0, 1.0]\n",
-      "\n",
-      "Probability of whole sequence :  0.0315\n"
-     ]
-    }
-   ],
-   "source": [
-    "beta = backward_alg()\n",
-    "for i in beta:\n",
-    "    print(i)\n",
-    "pos_beta = beta[0][0]*ip[0] + beta[0][1]*ip[1]\n",
-    "print(\"\\nProbability of whole sequence : \",pos_beta)"
-   ]
-  }
- ],
- "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": "3.12.4"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 5
-}

noshot/data/ML TS XAI/NLP/NLP 5/6.Viterbi.ipynb

@@ -1,111 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "id": "054ed639-31cd-489c-87d7-6f910bdff677",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "states = ['cp', 'ip']\n",
-    "ip = [1.0, 0.0]\n",
-    "stp = [[0.7, 0.3], [0.5, 0.5]]\n",
-    "op = ['lem', 'ice', 'cola']\n",
-    "opp = [[0.3, 0.1, 0.6], [0.2, 0.7, 0.1]]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "id": "11e5e731-7595-450d-866e-cdd0f488d20f",
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "def viterbi_algorithm(op):\n",
-    "    delta = []\n",
-    "    psi = []\n",
-    "\n",
-    "    # Initialize the delta and psi matrices\n",
-    "    for _ in range(len(op)):\n",
-    "        delta.append([0.0, 0.0])\n",
-    "        psi.append([0, 0])\n",
-    "\n",
-    "    # Initialization step\n",
-    "    for i in range(len(states)):\n",
-    "        delta[0][i] = ip[i] * opp[i][0]\n",
-    "\n",
-    "    # Recursion step\n",
-    "    for t in range(1, len(op)):\n",
-    "        for j in range(len(states)):\n",
-    "            max_prob = 0.0\n",
-    "            max_state = 0\n",
-    "\n",
-    "            for i in range(len(states)):\n",
-    "                prob = delta[t - 1][i] * stp[i][j] * opp[j][t]\n",
-    "                if prob > max_prob:\n",
-    "                    max_prob = prob\n",
-    "                    max_state = i\n",
-    "\n",
-    "            delta[t][j] = max_prob\n",
-    "            psi[t][j] = max_state\n",
-    "\n",
-    "    # Termination step\n",
-    "    best_path_prob = max(delta[-1])\n",
-    "    best_last_state = delta[-1].index(best_path_prob)\n",
-    "\n",
-    "    # Backtrack to find the best tag sequence\n",
-    "    best_path = [best_last_state]\n",
-    "    for t in range(len(op) - 1, 0, -1):\n",
-    "        best_last_state = psi[t][best_last_state]\n",
-    "        best_path.insert(0, best_last_state)\n",
-    "\n",
-    "    return best_path, best_path_prob"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "id": "9a2c4fc4-ba8e-4d4f-9f58-8f225919508b",
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Best Tag Sequence: ['cp', 'ip', 'cp']\n",
-      "Probability of the Best Tag Sequence: 0.0189\n"
-     ]
-    }
-   ],
-   "source": [
-    "# Call the Viterbi algorithm\n",
-    "best_tag_sequence, probability = viterbi_algorithm(op)\n",
-    "\n",
-    "# Print the best tag sequence and its probability\n",
-    "print(\"Best Tag Sequence:\", [states[i] for i in best_tag_sequence])\n",
-    "print(\"Probability of the Best Tag Sequence:\", probability)"
-   ]
-  }
- ],
- "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": "3.12.4"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 5
-}

noshot/data/ML TS XAI/NLP/NLP 5/7.PCFG Parse Tree.ipynb

@@ -1,134 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "LHS `->` RHS  \n",
-    "Element1 `|` Element2 `|` Element3  \n",
-    "`NP PP [0.4]` -- non-terminal symbol   \n",
-    "`'he' [0.6]` -- terminal symbol "
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "grammarstring = \"\"\"\n",
-    "S -> NP VP [1.0]\n",
-    "NP -> NP PP [0.4] | 'he' [0.1] | 'dessert' [0.3] | 'lunch' [0.1] | 'saw' [0.1]\n",
-    "PP -> Pre NP [1.0]\n",
-    "VP -> Verb NP [0.3] | VP PP [0.7]\n",
-    "Pre -> 'with' [0.6] | 'in' [0.4]\n",
-    "Verb -> 'ate' [0.7] | 'saw' [0.3]\n",
-    "\"\"\""
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "from nltk import PCFG, InsideChartParser \n",
-    "# Remember InsideChartParser\n",
-    "\n",
-    "grammar = PCFG.fromstring(grammarstring)\n",
-    "\n",
-    "parser = InsideChartParser(grammar=grammar)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "['he', 'saw', 'lunch', 'with', 'dessert']"
-      ]
-     },
-     "execution_count": 3,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "from nltk.tokenize import word_tokenize\n",
-    "\n",
-    "sentence = \"he saw lunch with dessert\"\n",
-    "\n",
-    "tokens = word_tokenize(sentence)\n",
-    "tokens"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 4,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "               S                       \n",
-      "  _____________|____                    \n",
-      " |                  VP                 \n",
-      " |         _________|________           \n",
-      " |        VP                 PP        \n",
-      " |    ____|____          ____|_____     \n",
-      " NP Verb       NP      Pre         NP  \n",
-      " |   |         |        |          |    \n",
-      " he saw      lunch     with     dessert\n",
-      "\n",
-      "Prob:  0.00011339999999999999\n",
-      "           S                       \n",
-      "  _________|____                    \n",
-      " |              VP                 \n",
-      " |    __________|___                \n",
-      " |   |              NP             \n",
-      " |   |      ________|____           \n",
-      " |   |     |             PP        \n",
-      " |   |     |         ____|_____     \n",
-      " NP Verb   NP      Pre         NP  \n",
-      " |   |     |        |          |    \n",
-      " he saw  lunch     with     dessert\n",
-      "\n",
-      "Prob:  6.480000000000002e-05\n"
-     ]
-    }
-   ],
-   "source": [
-    "trees = parser.parse(tokens)\n",
-    "\n",
-    "for tree in trees:\n",
-    "    tree.pretty_print() # Remember this\n",
-    "    print(\"Prob: \", tree.prob())# .prob"
-   ]
-  }
- ],
- "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": "3.12.4"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 4
-}

noshot/data/ML TS XAI/NLP/NLP 5/7.PCFG using cyk.ipynb

@@ -1,101 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "id": "fcbaa054-0d06-4ac6-86f6-817121de6640",
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Inside probability of S: 21.400000000000002\n",
-      "Inside probability of NP: 18.479999999999997\n",
-      "Inside probability of VP: 17.120000000000005\n",
-      "Inside probability of Det: 0.0\n",
-      "Inside probability of N: 0.0\n",
-      "Inside probability of V: 0.0\n",
-      "Total inside probability: 21.400000000000002\n"
-     ]
-    }
-   ],
-   "source": [
-    "from collections import defaultdict\n",
-    "from itertools import product\n",
-    "\n",
-    "def cyk_algorithm(words, pcfg_rules):\n",
-    "    n = len(words)\n",
-    "    table = [[defaultdict(float) for _ in range(n)] for _ in range(n)]\n",
-    "\n",
-    "    # Initialization\n",
-    "    for i, word in enumerate(words):\n",
-    "        for nt, (prob, terminals) in pcfg_rules.items():\n",
-    "            if word in terminals:\n",
-    "                table[i][i][nt] = prob\n",
-    "\n",
-    "    # CYK Algorithm\n",
-    "    for length in range(2, n + 1):\n",
-    "        for i in range(n - length + 1):\n",
-    "            j = i + length - 1\n",
-    "            for k in range(i, j):\n",
-    "                for A, (prob_A, _) in pcfg_rules.items():\n",
-    "                    for B, (prob_B, _) in pcfg_rules.items():\n",
-    "                        for C in table[i][k]:\n",
-    "                            for D in table[k + 1][j]:\n",
-    "                                prob = prob_A * prob_B * pcfg_rules[A][1].count(C) * pcfg_rules[B][1].count(D)\n",
-    "                                table[i][j][A] += prob\n",
-    "\n",
-    "    return table\n",
-    "\n",
-    "# Example PCFG rules (non-terminal -> (probability, [productions]))\n",
-    "pcfg_rules = {\n",
-    "    'S': (1.0, ['NP', 'VP']),\n",
-    "    'NP': (0.7, ['Det', 'N']),\n",
-    "    'VP': (0.8, ['V', 'NP']),\n",
-    "    'Det': (1.0, ['the']),\n",
-    "    'N': (0.6, ['cat', 'dog']),\n",
-    "    'V': (0.9, ['chased'])\n",
-    "}\n",
-    "\n",
-    "# Example input sentence\n",
-    "words = ['the', 'cat', 'chased', 'the', 'dog']\n",
-    "\n",
-    "# Call CYK algorithm to get inside probabilities\n",
-    "table = cyk_algorithm(words, pcfg_rules)\n",
-    "\n",
-    "# Inside probabilities for non-terminals in the top cell of the table\n",
-    "inside_probabilities = table[0][-1]\n",
-    "\n",
-    "# Print inside probabilities\n",
-    "for nt, prob in inside_probabilities.items():\n",
-    "    print(f'Inside probability of {nt}: {prob}')\n",
-    "\n",
-    "# Total inside probability (probability of the whole sentence)\n",
-    "total_probability = inside_probabilities['S']\n",
-    "print(f'Total inside probability: {total_probability}')"
-   ]
-  }
- ],
- "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": "3.12.4"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 5
-}