有無Machine Learning高手

如果UG major cs未grad 又想行machine learning條path
我應該學好啲乜？

Machine Learning 唔係CS科目咩？定係DataScience ?

得兩個elective course係講AI ML

其實好多時上堂學到嘅嘢都好theoretical. 應用上都係要自己去學。 打個比喻，喺學校上堂學到嘅嘢就好似學到車裏面部engine點 work 有啲乜嘢原理。但係應用上你只要識揸部車就得啦。你有時間有興趣先至打開個車頭蓋睇吓裏面啲零件。好可惜係學校學到嘅嘢，由一粒螺絲開始，慢慢你起個engine。將一啲普通嘅idea 複雜化, 所以好多人都唔知點樣入手ML.

有興趣喺 ML 加返啲人性入去
有好多行業同情況唔需要去到 big data ML
Good enough is good enough
應該會係下個熱潮

ML newbie
近年Asia stat arb 點?
好似唔係好work?

MeanReversion 好死。Momentum 有得執下。打個和啦。要永恆working ststarb trade EU la.

有stat底已經贏在起跑線。我應為加小小Linear Algebra. Basic Python, 之後 focus on http://Keras.io Tensorflow. 無敵。

BTW ML 可以好多範。好多industries 都要人材做ML. 想清楚自己興趣，之後專攻一兩樣，如果唔係，好難學得哂。

而家好多人加residual入去network, alexnet唔多人用了

我見到而家興deep residual network
同埋最近開始講hinton嗰個capsule network
有冇ching已經用緊CaspNet試嘢？

有stat底已經贏在起跑線。我應為加小小Linear Algebra. Basic Python, 之後 focus on http://Keras.io Tensorflow. 無敵。

BTW ML 可以好多範。好多industries 都要人材做ML. 想清楚自己興趣，之後專攻一兩樣，如果唔係，好難學得哂。

而家好多人加residual入去network, alexnet唔多人用了

我見到而家興deep residual network
同埋最近開始講hinton嗰個capsule network
有冇ching已經用緊CaspNet試嘢？

多謝Tips ！未玩過，放入ToDo list. 依家去研究。

有stat底已經贏在起跑線。我應為加小小Linear Algebra. Basic Python, 之後 focus on http://Keras.io Tensorflow. 無敵。

BTW ML 可以好多範。好多industries 都要人材做ML. 想清楚自己興趣，之後專攻一兩樣，如果唔係，好難學得哂。

而家好多人加residual入去network, alexnet唔多人用了

我見到而家興deep residual network
同埋最近開始講hinton嗰個capsule network
有冇ching已經用緊CaspNet試嘢？

多謝Tips ！未玩過，放入ToDo list. 依家去研究。

唔駛客氣
呢個network原意係想用喺computer vision度
如果大家有興趣可以去睇呢條youtube片，條片係解釋返caspnet嘅原理，連Hinton本人都喺reddit讚佢解釋得好
https://youtu.be/pPN8d0E3900


https://www.reddit.com/r/MachineLearning/comments/7ew7ba/d_capsule_networks_capsnets_tutorial/

如果UG major cs未grad 又想行machine learning條path
我應該學好啲乜？

Machine Learning 唔係CS科目咩？定係DataScience ?

得兩個elective course係講AI ML

其實好多時上堂學到嘅嘢都好theoretical. 應用上都係要自己去學。 打個比喻，喺學校上堂學到嘅嘢就好似學到車裏面部engine點 work 有啲乜嘢原理。但係應用上你只要識揸部車就得啦。你有時間有興趣先至打開個車頭蓋睇吓裏面啲零件。好可惜係學校學到嘅嘢，由一粒螺絲開始，慢慢你起個engine。將一啲普通嘅idea 複雜化, 所以好多人都唔知點樣入手ML.

應該要點入手

做車手先。是旦down 一兩架車嚟玩(toolkit: Keras With TensorFlow backend, or something quite interesting PyRO), 當你setup又run啲sample之後,你會覺得好神奇, 你就會慢慢打開車頭蓋，慢慢研究。你所有問題Google上面都有答案，一啲都唔似我哋啲死人PropTrade QuantTrade, 下下都Blackbox.

要玩ML就係而家. 因為有太多corporation opensource晒佢哋packages.

巴打 我對algo trading幾有興趣, 應該點開始？

大學讀過基本Python, major Finance 依加讀緊網上Udemy既 Python for Financial Analysis and Algorithmic Trading

推介andrew ng 既online course,
另外呢個fb grp好搞笑，
Machine Learning Memes for Convolutional Teens

如果UG major cs未grad 又想行machine learning條path
我應該學好啲乜？

Machine Learning 唔係CS科目咩？定係DataScience ?

得兩個elective course係講AI ML

其實好多時上堂學到嘅嘢都好theoretical. 應用上都係要自己去學。 打個比喻，喺學校上堂學到嘅嘢就好似學到車裏面部engine點 work 有啲乜嘢原理。但係應用上你只要識揸部車就得啦。你有時間有興趣先至打開個車頭蓋睇吓裏面啲零件。好可惜係學校學到嘅嘢，由一粒螺絲開始，慢慢你起個engine。將一啲普通嘅idea 複雜化, 所以好多人都唔知點樣入手ML.

應該要點入手

做車手先。是旦down 一兩架車嚟玩(toolkit: Keras With TensorFlow backend, or something quite interesting PyRO), 當你setup又run啲sample之後,你會覺得好神奇, 你就會慢慢打開車頭蓋，慢慢研究。你所有問題Google上面都有答案，一啲都唔似我哋啲死人PropTrade QuantTrade, 下下都Blackbox.

要玩ML就係而家. 因為有太多corporation opensource晒佢哋packages.

Pyro係做probabilistic modelling, Bayesian inference喎，Uber新出package, 要識用PyTorch

留名學嘢
其實而家做Algorithm Trading有幾準？

利申：用咗好多年Matlab，感覺同樓主背景有啲似

Lm

留名學嘢

有stat底已經贏在起跑線。我應為加小小Linear Algebra. Basic Python, 之後 focus on http://Keras.io Tensorflow. 無敵。

BTW ML 可以好多範。好多industries 都要人材做ML. 想清楚自己興趣，之後專攻一兩樣，如果唔係，好難學得哂。

而家好多人加residual入去network, alexnet唔多人用了

我見到而家興deep residual network
同埋最近開始講hinton嗰個capsule network
有冇ching已經用緊CaspNet試嘢？

多謝Tips ！未玩過，放入ToDo list. 依家去研究。

唔駛客氣
呢個network原意係想用喺computer vision度
如果大家有興趣可以去睇呢條youtube片，條片係解釋返caspnet嘅原理，連Hinton本人都喺reddit讚佢解釋得好
https://youtu.be/pPN8d0E3900


https://www.reddit.com/r/MachineLearning/comments/7ew7ba/d_capsule_networks_capsnets_tutorial/

好撚深

如果UG major cs未grad 又想行machine learning條path
我應該學好啲乜？

Machine Learning 唔係CS科目咩？定係DataScience ?

得兩個elective course係講AI ML

其實好多時上堂學到嘅嘢都好theoretical. 應用上都係要自己去學。 打個比喻，喺學校上堂學到嘅嘢就好似學到車裏面部engine點 work 有啲乜嘢原理。但係應用上你只要識揸部車就得啦。你有時間有興趣先至打開個車頭蓋睇吓裏面啲零件。好可惜係學校學到嘅嘢，由一粒螺絲開始，慢慢你起個engine。將一啲普通嘅idea 複雜化, 所以好多人都唔知點樣入手ML.

應該要點入手

做車手先。是旦down 一兩架車嚟玩(toolkit: Keras With TensorFlow backend, or something quite interesting PyRO), 當你setup又run啲sample之後,你會覺得好神奇, 你就會慢慢打開車頭蓋，慢慢研究。你所有問題Google上面都有答案，一啲都唔似我哋啲死人PropTrade QuantTrade, 下下都Blackbox.

要玩ML就係而家. 因為有太多corporation opensource晒佢哋packages.

Pyro係做probabilistic modelling, Bayesian inference喎，Uber新出package, 要識用PyTorch

係呀 ! ML 主要係去 forecast 一個 outcome. 大部份人依家主要都係focus 在 neural network. 好多人忽視左 Bayesian Inference centric 嘅 model. 當你手上無Big Data. Bayesian Inference is quite effective. 所以我 suggest 學兩派，NN on Keras. BPmodeling on PyRO ( pretty new to the scene ).

你有什麼經驗可以分享下？

Machine Learning 唔係CS科目咩？定係DataScience ?

得兩個elective course係講AI ML

其實好多時上堂學到嘅嘢都好theoretical. 應用上都係要自己去學。 打個比喻，喺學校上堂學到嘅嘢就好似學到車裏面部engine點 work 有啲乜嘢原理。但係應用上你只要識揸部車就得啦。你有時間有興趣先至打開個車頭蓋睇吓裏面啲零件。好可惜係學校學到嘅嘢，由一粒螺絲開始，慢慢你起個engine。將一啲普通嘅idea 複雜化, 所以好多人都唔知點樣入手ML.

應該要點入手

做車手先。是旦down 一兩架車嚟玩(toolkit: Keras With TensorFlow backend, or something quite interesting PyRO), 當你setup又run啲sample之後,你會覺得好神奇, 你就會慢慢打開車頭蓋，慢慢研究。你所有問題Google上面都有答案，一啲都唔似我哋啲死人PropTrade QuantTrade, 下下都Blackbox.

要玩ML就係而家. 因為有太多corporation opensource晒佢哋packages.

Pyro係做probabilistic modelling, Bayesian inference喎，Uber新出package, 要識用PyTorch

係呀 ! ML 主要係去 forecast 一個 outcome. 大部份人依家主要都係focus 在 neural network. 好多人忽視左 Bayesian Inference centric 嘅 model. 當你手上無Big Data. Bayesian Inference is quite effective. 所以我 suggest 學兩派，NN on Keras. BPmodeling on PyRO ( pretty new to the scene ).

你有什麼經驗可以分享下？

巴打 我對algo trading幾有興趣, 應該點開始？

大學讀過基本Python, major Finance 依加讀緊網上Udemy既 Python for Financial Analysis and Algorithmic Trading

首先，AlgoTrading , PropQuantTrade 係好唔同。好多人都以為係一樣。

AlgoTrading mainly means using algothrim to help facilitate trading orders. I.e. Vwap Twap Iceberg 之類。賺嘅錢係睇吓你 execution cost 個slippage. 跟住就係charge client 的commission. 所有order都係initiated by clients.

PropQuantTrade 就自由啲。即係啲人所講嘅Blackbox trading. All orders are generated by the strategy ( we don’t call this algorithm ). 賺嘅錢係你嘅PNL.

回歸正題，AlgoTrading ML 嘅 input 係clients orders details and time stamp , output 係 cost saving.
因為clients orders 都好有pattern ， 無乜noise ， ML 可以設計出一款可以減低成本嘅execution algorithm.

PropTrade 就難一啲。Input 係market data ， output 係 pnl. ML 要設計一款strategy maximize pnl and sharpe.

We can discuss more if you are interested in trading; I try to share as much as I could.

得兩個elective course係講AI ML

其實好多時上堂學到嘅嘢都好theoretical. 應用上都係要自己去學。 打個比喻，喺學校上堂學到嘅嘢就好似學到車裏面部engine點 work 有啲乜嘢原理。但係應用上你只要識揸部車就得啦。你有時間有興趣先至打開個車頭蓋睇吓裏面啲零件。好可惜係學校學到嘅嘢，由一粒螺絲開始，慢慢你起個engine。將一啲普通嘅idea 複雜化, 所以好多人都唔知點樣入手ML.

應該要點入手

做車手先。是旦down 一兩架車嚟玩(toolkit: Keras With TensorFlow backend, or something quite interesting PyRO), 當你setup又run啲sample之後,你會覺得好神奇, 你就會慢慢打開車頭蓋，慢慢研究。你所有問題Google上面都有答案，一啲都唔似我哋啲死人PropTrade QuantTrade, 下下都Blackbox.

要玩ML就係而家. 因為有太多corporation opensource晒佢哋packages.

Pyro係做probabilistic modelling, Bayesian inference喎，Uber新出package, 要識用PyTorch

係呀 ! ML 主要係去 forecast 一個 outcome. 大部份人依家主要都係focus 在 neural network. 好多人忽視左 Bayesian Inference centric 嘅 model. 當你手上無Big Data. Bayesian Inference is quite effective. 所以我 suggest 學兩派，NN on Keras. BPmodeling on PyRO ( pretty new to the scene ).

你有什麼經驗可以分享下？

我都係從BI 開始。因為我嗰field data 太小同noisy. 我玩過PyMC3 但太難。冇辦法，轉左玩RNN.

有stat底已經贏在起跑線。我應為加小小Linear Algebra. Basic Python, 之後 focus on http://Keras.io Tensorflow. 無敵。

BTW ML 可以好多範。好多industries 都要人材做ML. 想清楚自己興趣，之後專攻一兩樣，如果唔係，好難學得哂。

而家好多人加residual入去network, alexnet唔多人用了

我見到而家興deep residual network
同埋最近開始講hinton嗰個capsule network
有冇ching已經用緊CaspNet試嘢？

多謝Tips ！未玩過，放入ToDo list. 依家去研究。

唔駛客氣
呢個network原意係想用喺computer vision度
如果大家有興趣可以去睇呢條youtube片，條片係解釋返caspnet嘅原理，連Hinton本人都喺reddit讚佢解釋得好
https://youtu.be/pPN8d0E3900


https://www.reddit.com/r/MachineLearning/comments/7ew7ba/d_capsule_networks_capsnets_tutorial/

好撚深

兩星期前的 project 。雖然paper 係2011. ML 真係日日新鮮日日金。

留名，新手。但覺得Alphago 套觀棋model 可以套用落trading，近似人類睇股市既陰陽燭。但trading成功資金管理佔重比例，要搵其他model 去做模擬

有stat底已經贏在起跑線。我應為加小小Linear Algebra. Basic Python, 之後 focus on http://Keras.io Tensorflow. 無敵。

BTW ML 可以好多範。好多industries 都要人材做ML. 想清楚自己興趣，之後專攻一兩樣，如果唔係，好難學得哂。

而家好多人加residual入去network, alexnet唔多人用了

我見到而家興deep residual network
同埋最近開始講hinton嗰個capsule network
有冇ching已經用緊CaspNet試嘢？

多謝Tips ！未玩過，放入ToDo list. 依家去研究。

唔駛客氣
呢個network原意係想用喺computer vision度
如果大家有興趣可以去睇呢條youtube片，條片係解釋返caspnet嘅原理，連Hinton本人都喺reddit讚佢解釋得好
https://youtu.be/pPN8d0E3900


https://www.reddit.com/r/MachineLearning/comments/7ew7ba/d_capsule_networks_capsnets_tutorial/

好撚深

兩星期前的 project 。雖然paper 係2011. ML 真係日日新鮮日日金。

已經有keras tf pytorch

呢個network做左我之前想做既3D connection,同埋spatial invariant 問題

但係computational requirement好似都幾高下

而家好多人加residual入去network, alexnet唔多人用了

我見到而家興deep residual network
同埋最近開始講hinton嗰個capsule network
有冇ching已經用緊CaspNet試嘢？

多謝Tips ！未玩過，放入ToDo list. 依家去研究。

唔駛客氣
呢個network原意係想用喺computer vision度
如果大家有興趣可以去睇呢條youtube片，條片係解釋返caspnet嘅原理，連Hinton本人都喺reddit讚佢解釋得好
https://youtu.be/pPN8d0E3900


https://www.reddit.com/r/MachineLearning/comments/7ew7ba/d_capsule_networks_capsnets_tutorial/

好撚深

兩星期前的 project 。雖然paper 係2011. ML 真係日日新鮮日日金。

已經有keras tf pytorch

呢個network做左我之前想做既3D connection,同埋spatial invariant 問題

但係computational requirement好似都幾高下

佢仲有篇新既soft decision tree