search

kochkov92 / Tictactoe

Simplistic Tic-Tac-Toe and an AI for it
1 stars 1 forks source link

Might not be able to tomorrow #4

Open j-chou opened 8 years ago

j-chou commented 8 years ago

Hey guys, I'm studying for a final with a friend tomorrow morning, so I might not be able to make it this Sunday. Please continue without me.

tranchieuminh commented 8 years ago

Hey,

Since two of us are not here tmr, may be we should skip.

Minh

On Sat, Dec 12, 2015 at 2:48 PM, j-chou notifications@github.com wrote:

Hey guys, I'm studying for a final with a friend tomorrow morning, so I might not be able to make it this Sunday. Please continue without me.

— Reply to this email directly or view it on GitHub https://github.com/kochkov92/Tictactoe/issues/4.

kochkov92 commented 8 years ago

Hey,

Yeah, I guess that makes sense, I also got to finish the projects that are due coming Tuesday…

Dima

On Dec 12, 2015, at 9:50 PM, tranchieuminh notifications@github.com wrote:

Hey,

Since two of us are not here tmr, may be we should skip.

Minh

On Sat, Dec 12, 2015 at 2:48 PM, j-chou notifications@github.com wrote:

Hey guys, I'm studying for a final with a friend tomorrow morning, so I might not be able to make it this Sunday. Please continue without me.

— Reply to this email directly or view it on GitHub https://github.com/kochkov92/Tictactoe/issues/4.

— Reply to this email directly or view it on GitHub https://github.com/kochkov92/Tictactoe/issues/4#issuecomment-164218541.

jdove2 commented 8 years ago

Had some spare time on shift so I added a bit. Mostly put a skeleton for the NeuralNetwork class and a specific TTTNetwork class. As of now I can make a network with an arbitrary number of hidden layers and arbitrary number of nodes per layer (must be equal for each layer) and it makes a complete network with random weights. Also added a print statement to the neuron class so it can print out in a readable format

I noticed the constructors for the Neurons dont include a layerIndex so I made a new constructor that does, although we probably want to go back and change or remove the old constructors since the ordering of parameters might get confusing. Could probably make the neural network more elegantly decide which activation function to use instead of having it hard coded. Havent tested the computeMove method yet.

Jason

On Sun, Dec 13, 2015 at 8:04 AM, Dima notifications@github.com wrote:

Hey,

Yeah, I guess that makes sense, I also got to finish the projects that are due coming Tuesday…

Dima

On Dec 12, 2015, at 9:50 PM, tranchieuminh notifications@github.com wrote:

Hey,

Since two of us are not here tmr, may be we should skip.

Minh

On Sat, Dec 12, 2015 at 2:48 PM, j-chou notifications@github.com wrote:

Hey guys, I'm studying for a final with a friend tomorrow morning, so I might not be able to make it this Sunday. Please continue without me.

— Reply to this email directly or view it on GitHub https://github.com/kochkov92/Tictactoe/issues/4.

— Reply to this email directly or view it on GitHub < https://github.com/kochkov92/Tictactoe/issues/4#issuecomment-164218541>.

— Reply to this email directly or view it on GitHub https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_kochkov92_Tictactoe_issues_4-23issuecomment-2D164261840&d=BQMFaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=drJyf_ptmyjtcnhwdvosmeUbQ0JOdEtdCmPOI-BoePM&m=gZi-K9rw9hDD8tRQelogq60U2M9YQZj2-pVb4iRxyik&s=STOAoBOjdg526dVQ2EEUORw9HPi7wcKlJevxuittljw&e= .