endolith

*.json
*checkpoint.ipynb

endolith

Written by ChatGPT and me.

I also had to enable temperhum as a non-root user and set up a cron job to call this script every once in a while.

endolith

FINALLY.

Instructions:

1. Install Qalculate
2. Install AutoHotkey
3. Add qalc.exe to your PATH. I use Rapid Environment Editor for this.
4. Run qalc.ahk

To use it:

endolith

"""
Break up a long recording to fit within the Whisper API's limits, with some
overlap, so no words are missed, and then feed to OpenAI Whisper API to
transcribe it to .txt file.  Written by endolith and ChatGPT-4.
"""

import openai
import math
import os
import subprocess

endolith

import numpy as np
import matplotlib.pyplot as plt
from scipy.ndimage import gaussian_filter

rng = np.random.default_rng()

# shape = (1024, )
# shape = (512, 512)
shape = (128, 128, 128)

endolith

"""
Written by GPT3 from this English prompt (with one tweak to the prompt, to get 
the timestamp parameter to not error out, and PEP8 fixes):

----

1. Open a sqlite database with a name like `2022-07-17 17.48.19.db` or
    `2021-04-11 08.44.16.db`
2. Inside is a table called `locations`, with headers [LOCTYPE INTEGER,
    TIMESTAMP BIGINT, LATITUDE REAL, LONGITUDE REAL,ALTITUDE REAL, ACCURACY

endolith

"""
Confirm the formula for estimating the variance of an estimated probability.

Created on Fri May 27 11:06:43 2022
"""
import numpy as np


def func(x):
    """

endolith

My first keras experiments:

1. simplest.py — A single neuron with no activation function and no bias (so just input→weight→output, can learn y=mx functions only). Show the function for a few random weights. (Doesn't learn anything.)
2. linear with bias.py — A single neuron with no activation function but with bias (so it can learn y=mx+b functions, such as relationship between Celsius and Fahrenheit).
3. single_neuron_buffer.py — A single neuron with sigmoid activation and no bias, trained to learn a digital logic buffer.
4. single_neuron_inverter.py — A single neuron with sigmoid activation and no bias, trained to learn an inverting analog amplifier.

endolith

"""
Train a neural network to learn an FIR filter.

Created on Fri Aug  3 15:00:40 2018
"""
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense
from tensorflow.keras.callbacks import Callback
import numpy as np
from scipy import signal

endolith

python 3.8.*
scikit-learn ==0.24.1