Skip to content

Commit

Permalink
added more details
Browse files Browse the repository at this point in the history
  • Loading branch information
@nipunbatra
nipunbatra committed Dec 19, 2023
1 parent 52b9e9e commit 368fb30
Showing 1 changed file with 172 additions and 1 deletion.
173 changes: 172 additions & 1 deletion posts/2023-Dec-18-transcript.ipynb
View file
Original file line number Diff line number Diff line change
Expand Up @@ -1548,10 +1548,181 @@
},
{
"cell_type": "code",
"execution_count": null,
"execution_count": 24,
"id": "a7b39c75",
"metadata": {},
"outputs": [],
"source": [
"llm = Ollama(model=\"mistral\", \n",
" callback_manager = CallbackManager([StreamingStdOutCallbackHandler()]))"
]
},
{
"cell_type": "code",
"execution_count": 25,
"id": "b1808481",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Please provide a bullet-point summary for the given text:\n",
"\n",
" * The text discusses various concepts related to machine learning, including definitions, rules for recognizing digits using a program, and predicting the quality or condition of tomatoes based on their visual features.\n",
"* Machine learning is defined as the ability of computers to learn without being explicitly programmed, allowing them to make decisions and predictions based on data.\n",
"* The text covers an example of creating a company that uses computer vision to classify tomatoes as good or bad based on their visual features, with the goal of scaling the process to automate the assessment of tomato quality.\n",
"* The text also discusses various concepts related to machine learning algorithms and performance measures, including entropy and information gain.\n",
"* The text encourages readers to think about how these concepts could be applied in different contexts, such as tennis or doctor decision-making.\n",
"* The text mentions the ID3 algorithm as an example of a decision tree algorithm for classification problems.\n",
"* The text also mentions the importance of handling imbalance datasets and calculating precision, recall, F score, and Matthew's correlation coefficient to evaluate machine learning models accurately. * The text discusses various concepts related to machine learning, including definitions, rules for recognizing digits using a program, and predicting the quality or condition of tomatoes based on their visual features.\n",
"* Machine learning is defined as the ability of computers to learn without being explicitly programmed, allowing them to make decisions and predictions based on data.\n",
"* The text covers an example of creating a company that uses computer vision to classify tomatoes as good or bad based on their visual features, with the goal of scaling the process to automate the assessment of tomato quality.\n",
"* The text also discusses various concepts related to machine learning algorithms and performance measures, including entropy and information gain.\n",
"* The text encourages readers to think about how these concepts could be applied in different contexts, such as tennis or doctor decision-making.\n",
"* The text mentions the ID3 algorithm as an example of a decision tree algorithm for classification problems.\n",
"* The text also mentions the importance of handling imbalance datasets and calculating precision, recall, F score, and Matthew's correlation coefficient to evaluate machine learning models accurately.\n",
"\n",
"====================================================================================================\n",
"\n",
"Summarize the following in Markdown bullets:\n",
"\n",
" * The speaker requests attendees to sign up on Google Cloud and mentions some announcements\n",
"* There will be an extra lecture on Saturday, 11th Jan at 11am in 1.101\n",
"* Attendees are encouraged to ask questions about the FAQ and projects shared on Google Docs\n",
"* Announcement of availability of video and slides from first lecture on Google Cloud\n",
"* Review of previous lecture on machine learning, focusing on definition and difference between explicit and implicit programming\n",
"* Task: write a program to recognize digits in an image dataset\n",
"* Discussion on rules for recognizing digits and the importance of considering visual features\n",
"* Introduction to the concept of machine learning as a way to learn from experience and improve performance over time\n",
"* Announcement of project for this semester: predicting the quality or condition of computer components based on visual features\n",
"* Discussion of use cases and business implications, such as scaling processes and reducing human input in grocery stores\n",
"* Introduction to concepts of training set, test set, and model evaluation using metrics such as accuracy, precision, recall, and F1 score. * The speaker requests attendees to sign up on Google Cloud and mentions some announcements\n",
"* There will be an extra lecture on Saturday, 11th Jan at 11am in 1.101\n",
"* Attendees are encouraged to ask questions about the FAQ and projects shared on Google Docs\n",
"* Announcement of availability of video and slides from first lecture on Google Cloud\n",
"* Review of previous lecture on machine learning, focusing on definition and difference between explicit and implicit programming\n",
"* Task: write a program to recognize digits in an image dataset\n",
"* Discussion on rules for recognizing digits and the importance of considering visual features\n",
"* Introduction to the concept of machine learning as a way to learn from experience and improve performance over time\n",
"* Announcement of project for this semester: predicting the quality or condition of computer components based on visual features\n",
"* Discussion of use cases and business implications, such as scaling processes and reducing human input in grocery stores\n",
"* Introduction to concepts of training set, test set, and model evaluation using metrics such as accuracy, precision, recall, and F1 score.\n",
"\n",
"====================================================================================================\n",
"\n",
"Highlight the important topics and subtopics in the given lecture:\n",
"\n",
" Important topics and subtopics in the given lecture:\n",
"\n",
"1. Machine Learning: Definition, Explicit vs Implicit programming\n",
"2. Digit recognition using machine learning: Rules, constraints\n",
"3. Machine Learning Algorithms: Decision Trees, Information Gain, Entropy\n",
"4. Decision Tree Construction: Root node, recursive property, best attribute\n",
"5. Performance Measure: Entropy, Information Gain, Greedy algorithm\n",
"6. IDP Algorithm: Examples target attribute and arguments, creating root node, recursive partitioning\n",
"7. Evaluating the performance of a decision tree: Accuracy, Precision, Recall, F1-score, Matthew's correlation coefficient.\n",
"\n",
"Additional topics discussed but not explicitly stated as subtopics:\n",
"\n",
"* Google Cloud, sign up and announcements\n",
"* Extra lecture on Saturday, 11th Jan at 11am in 1.101\n",
"* FAQ and projects\n",
"* Previous lecture recap and machine learning history\n",
"* Data rules and traditional programming vs machine learning\n",
"* Computer vision with tomato quality prediction\n",
"* Energy consumption prediction for IT on campus\n",
"* Performance measures in regression tasks: Mean squared error, Mean absolute error, Mean error. Important topics and subtopics in the given lecture:\n",
"\n",
"1. Machine Learning: Definition, Explicit vs Implicit programming\n",
"2. Digit recognition using machine learning: Rules, constraints\n",
"3. Machine Learning Algorithms: Decision Trees, Information Gain, Entropy\n",
"4. Decision Tree Construction: Root node, recursive property, best attribute\n",
"5. Performance Measure: Entropy, Information Gain, Greedy algorithm\n",
"6. IDP Algorithm: Examples target attribute and arguments, creating root node, recursive partitioning\n",
"7. Evaluating the performance of a decision tree: Accuracy, Precision, Recall, F1-score, Matthew's correlation coefficient.\n",
"\n",
"Additional topics discussed but not explicitly stated as subtopics:\n",
"\n",
"* Google Cloud, sign up and announcements\n",
"* Extra lecture on Saturday, 11th Jan at 11am in 1.101\n",
"* FAQ and projects\n",
"* Previous lecture recap and machine learning history\n",
"* Data rules and traditional programming vs machine learning\n",
"* Computer vision with tomato quality prediction\n",
"* Energy consumption prediction for IT on campus\n",
"* Performance measures in regression tasks: Mean squared error, Mean absolute error, Mean error.\n",
"\n",
"====================================================================================================\n",
"\n",
"Give us some question for a quiz based on the following text:\n",
"\n",
" 1. What is the purpose of signing up on Google Cloud mentioned in the text?\n",
"2. In what location will there be an extra lecture taking place on January 11th at 11am?\n",
"3. What are the rules for recognizing the digit '4' from the dataset provided?\n",
"4. Why do the slides and video from the first lecture not yet exist on code translate?\n",
"5. What is the definition of machine learning according to Arthur Sandler, who first used the term in 1959?\n",
"6. Can a program itself learn? Or is it always explicitly programmed?\n",
"7. In what way does adopting an executive program differ from traditional programming?\n",
"8. How can one recognize if a digit is '4' based on its shape?\n",
"9. What is the meaning of \"slides\" in this context?\n",
"10. What are some other rules for recognizing digits besides the ones mentioned?\n",
"11. In addition to color, size, and texture, what other features might be useful for predicting the condition (quality or ripeness) of tomatoes?\n",
"12. How can the sample number be a useful feature when predicting the condition of tomatoes?\n",
"13. What is the training set in machine learning and how does it differ from the test set?\n",
"14. What are some examples of regression problems in machine learning?\n",
"15. Can you explain the difference between precision and recall in machine learning?\n",
"16. Given a confusion matrix, how can precision and recall be calculated?\n",
"17. What is the main squared error and what role does it play in evaluating machine learning models?\n",
"18. What is the advantage of using decision trees over other more complex machine learning algorithms for certain tasks?\n",
"19. How does the ID3 algorithm work for building decision trees from a dataset?\n",
"20. What is entropy and how can it be used to select the best attribute for splitting a dataset in a decision tree? 1. What is the purpose of signing up on Google Cloud mentioned in the text?\n",
"2. In what location will there be an extra lecture taking place on January 11th at 11am?\n",
"3. What are the rules for recognizing the digit '4' from the dataset provided?\n",
"4. Why do the slides and video from the first lecture not yet exist on code translate?\n",
"5. What is the definition of machine learning according to Arthur Sandler, who first used the term in 1959?\n",
"6. Can a program itself learn? Or is it always explicitly programmed?\n",
"7. In what way does adopting an executive program differ from traditional programming?\n",
"8. How can one recognize if a digit is '4' based on its shape?\n",
"9. What is the meaning of \"slides\" in this context?\n",
"10. What are some other rules for recognizing digits besides the ones mentioned?\n",
"11. In addition to color, size, and texture, what other features might be useful for predicting the condition (quality or ripeness) of tomatoes?\n",
"12. How can the sample number be a useful feature when predicting the condition of tomatoes?\n",
"13. What is the training set in machine learning and how does it differ from the test set?\n",
"14. What are some examples of regression problems in machine learning?\n",
"15. Can you explain the difference between precision and recall in machine learning?\n",
"16. Given a confusion matrix, how can precision and recall be calculated?\n",
"17. What is the main squared error and what role does it play in evaluating machine learning models?\n",
"18. What is the advantage of using decision trees over other more complex machine learning algorithms for certain tasks?\n",
"19. How does the ID3 algorithm work for building decision trees from a dataset?\n",
"20. What is entropy and how can it be used to select the best attribute for splitting a dataset in a decision tree?\n",
"\n",
"====================================================================================================\n",
"\n"
]
}
],
"source": [
"prompt_qs = [\"Please provide a bullet-point summary for the given text:\",\n",
" \"Summarize the following in Markdown bullets:\",\n",
" \"Highlight the important topics and subtopics in the given lecture:\",\n",
" \"Give us some question for a quiz based on the following text:\"]\n",
"\n",
"prompts = [q + \"\\n\" + transcription[\"text\"] for q in prompt_qs]\n",
"\n",
"for prompt, prompt_qs in zip(prompts, prompt_qs):\n",
" print(prompt_qs, end=\"\\n\\n\")\n",
" output = llm(prompt)\n",
" print(output, end=\"\\n\\n\")\n",
" print(\"==\"*50, end=\"\\n\\n\")"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "94c6bcdf",
"metadata": {},
"outputs": [],
"source": []
}
],
Expand Down

0 comments on commit 368fb30

Please sign in to comment.