参考自邹博的博客!
LFM:将评分矩阵分解为 item-feature 和 user-feature矩阵,feature数量事先人工确定,但是这两个矩阵参数未知,首先随机选取参数,再以此梯度下降迭代即可得到。
<code class="hljs python has-numbering" style="display: block; padding: 0px; color: inherit; box-sizing: border-box; font-family: 'Source Code Pro', monospace;font-size:undefined; white-space: pre; border-radius: 0px; word-wrap: normal; background: transparent;"><span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">import</span> pandas <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">as</span> pd <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">import</span> numpy <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">as</span> np <span class="hljs-function" style="box-sizing: border-box;"><span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">def</span> <span class="hljs-title" style="box-sizing: border-box;">lfm</span><span class="hljs-params" style="color: rgb(102, 0, 102); box-sizing: border-box;">(user_item,k,alpha = <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0.01</span>,lamda = <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0.01</span>)</span>:</span> <span class="hljs-string" style="color: rgb(0, 136, 0); box-sizing: border-box;">"""user_item is matrix of user item,k is the number of latent number"""</span> <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">if</span> <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">not</span> isinstance(user_item,list): <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">raise</span>(<span class="hljs-string" style="color: rgb(0, 136, 0); box-sizing: border-box;">"user item:{user_item} is not a matrix list!"</span>.format(user_item=user_item)) mat = np.array(user_item) user_number,item_number = mat.shape <span class="hljs-comment" style="color: rgb(136, 0, 0); box-sizing: border-box;"># init the user and item latent matrix</span> u = np.random.rand(user_number,k) v = np.random.rand(item_number,k) <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">for</span> it <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">in</span> range(<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">1000</span>): <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">for</span> i <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">in</span> range(user_number): <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">for</span> j <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">in</span> range(item_number): <span class="hljs-comment" style="color: rgb(136, 0, 0); box-sizing: border-box;"># err</span> err = user_item[i][j] - np.dot(u[i],v[j]) <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">for</span> r <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">in</span> range(k): gu = err * v[j][r] + lamda * u[i][r] gv = err * u[i][r] + lamda * v[j][r] u[i][r] += alpha * gu v[j][r] += alpha * gv <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">return</span> u,v <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">if</span> __name__ == <span class="hljs-string" style="color: rgb(0, 136, 0); box-sizing: border-box;">"__main__"</span>: mat = [[<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>],[<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">3</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">4</span>],[<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">3</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">4</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">3</span>],[<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">3</span>],[<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">4</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">4</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>],[<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">4</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>]] u,v = lfm(mat,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">3</span>) <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">print</span> u <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">print</span> v <span class="hljs-keyword" style="color: rgb(0, 0, 136); box-sizing: border-box;">print</span> np.dot(u,v.T) </code><ul class="pre-numbering" style="box-sizing: border-box; position: absolute; width: 50px; top: 0px; left: 0px; margin: 0px; padding: 6px 0px 40px; border-right-width: 1px; border-right-style: solid; border-right-color: rgb(221, 221, 221); list-style: none; text-align: right; background-color: rgb(238, 238, 238);"><li style="box-sizing: border-box; padding: 0px 5px;">1</li><li style="box-sizing: border-box; padding: 0px 5px;">2</li><li style="box-sizing: border-box; padding: 0px 5px;">3</li><li style="box-sizing: border-box; padding: 0px 5px;">4</li><li style="box-sizing: border-box; padding: 0px 5px;">5</li><li style="box-sizing: border-box; padding: 0px 5px;">6</li><li style="box-sizing: border-box; padding: 0px 5px;">7</li><li style="box-sizing: border-box; padding: 0px 5px;">8</li><li style="box-sizing: border-box; padding: 0px 5px;">9</li><li style="box-sizing: border-box; padding: 0px 5px;">10</li><li style="box-sizing: border-box; padding: 0px 5px;">11</li><li style="box-sizing: border-box; padding: 0px 5px;">12</li><li style="box-sizing: border-box; padding: 0px 5px;">13</li><li style="box-sizing: border-box; padding: 0px 5px;">14</li><li style="box-sizing: border-box; padding: 0px 5px;">15</li><li style="box-sizing: border-box; padding: 0px 5px;">16</li><li style="box-sizing: border-box; padding: 0px 5px;">17</li><li style="box-sizing: border-box; padding: 0px 5px;">18</li><li style="box-sizing: border-box; padding: 0px 5px;">19</li><li style="box-sizing: border-box; padding: 0px 5px;">20</li><li style="box-sizing: border-box; padding: 0px 5px;">21</li><li style="box-sizing: border-box; padding: 0px 5px;">22</li><li style="box-sizing: border-box; padding: 0px 5px;">23</li><li style="box-sizing: border-box; padding: 0px 5px;">24</li><li style="box-sizing: border-box; padding: 0px 5px;">25</li><li style="box-sizing: border-box; padding: 0px 5px;">26</li><li style="box-sizing: border-box; padding: 0px 5px;">27</li><li style="box-sizing: border-box; padding: 0px 5px;">28</li><li style="box-sizing: border-box; padding: 0px 5px;">29</li><li style="box-sizing: border-box; padding: 0px 5px;">30</li></ul>
初始评分矩阵
<code class="hljs json has-numbering" style="display: block; padding: 0px; color: inherit; box-sizing: border-box; font-family: 'Source Code Pro', monospace;font-size:undefined; white-space: pre; border-radius: 0px; word-wrap: normal; background: transparent;">[[<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>],[<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">3</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">4</span>],[<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">3</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">4</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">3</span>],[<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">3</span>],[<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">4</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">4</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>],[<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">4</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>,<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5</span>]]</code><ul class="pre-numbering" style="box-sizing: border-box; position: absolute; width: 50px; top: 0px; left: 0px; margin: 0px; padding: 6px 0px 40px; border-right-width: 1px; border-right-style: solid; border-right-color: rgb(221, 221, 221); list-style: none; text-align: right; background-color: rgb(238, 238, 238);"><li style="box-sizing: border-box; padding: 0px 5px;">1</li></ul>
迭代后的u v 矩阵
<code class="hljs json has-numbering" style="display: block; padding: 0px; color: inherit; box-sizing: border-box; font-family: 'Source Code Pro', monospace;font-size:undefined; white-space: pre; border-radius: 0px; word-wrap: normal; background: transparent;"> [[ <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0.18447051</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">2.09168616</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0.8831042</span> ] [ <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">1.7912757</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">1.24549537</span> -<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0.48677593</span>] [-<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0.1002325</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">1.29114151</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0.86229546</span>] [ <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">1.38787106</span> -<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0.70947744</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0.76619629</span>] [ <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">1.26487691</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">1.27983092</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">1.02112708</span>] [ <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">1.48243041</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">1.08592233</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">1.13930052</span>]] [[ <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">1.30296976</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">2.33056037</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0.22352664</span>] [-<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0.38291845</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">1.51851172</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">2.39872099</span>] [ <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">2.63593771</span> -<span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0.74171661</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">1.47001405</span>] [ <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">1.59681827</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">1.36696209</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">1.48520738</span>]] </code><ul class="pre-numbering" style="box-sizing: border-box; position: absolute; width: 50px; top: 0px; left: 0px; margin: 0px; padding: 6px 0px 40px; border-right-width: 1px; border-right-style: solid; border-right-color: rgb(221, 221, 221); list-style: none; text-align: right; background-color: rgb(238, 238, 238);"><li style="box-sizing: border-box; padding: 0px 5px;">1</li><li style="box-sizing: border-box; padding: 0px 5px;">2</li><li style="box-sizing: border-box; padding: 0px 5px;">3</li><li style="box-sizing: border-box; padding: 0px 5px;">4</li><li style="box-sizing: border-box; padding: 0px 5px;">5</li><li style="box-sizing: border-box; padding: 0px 5px;">6</li><li style="box-sizing: border-box; padding: 0px 5px;">7</li><li style="box-sizing: border-box; padding: 0px 5px;">8</li><li style="box-sizing: border-box; padding: 0px 5px;">9</li><li style="box-sizing: border-box; padding: 0px 5px;">10</li><li style="box-sizing: border-box; padding: 0px 5px;">11</li><li style="box-sizing: border-box; padding: 0px 5px;">12</li><li style="box-sizing: border-box; padding: 0px 5px;">13</li></ul>
预测的结果
<code class="hljs json has-numbering" style="display: block; padding: 0px; color: inherit; box-sizing: border-box; font-family: 'Source Code Pro', monospace;font-size:undefined; white-space: pre; border-radius: 0px; word-wrap: normal; background: transparent;"> [[ <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5.31255767</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5.22393338</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0.23298999</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">4.46541444</span>] [ <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5.12787282</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0.03774717</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">3.08231911</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">3.83992351</span>] [ <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">3.07122934</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">4.06740062</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0.04571872</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">2.88557601</span>] [ <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0.3261393</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">0.22909989</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5.31089216</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">2.38430948</span>] [ <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">4.85906868</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">3.9084925</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">3.88593604</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5.28584437</span>] [ <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">4.71703356</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">3.81419992</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">4.77693537</span> <span class="hljs-number" style="color: rgb(0, 102, 102); box-sizing: border-box;">5.54368417</span>]] </code><ul class="pre-numbering" style="box-sizing: border-box; position: absolute; width: 50px; top: 0px; left: 0px; margin: 0px; padding: 6px 0px 40px; border-right-width: 1px; border-right-style: solid; border-right-color: rgb(221, 221, 221); list-style: none; text-align: right; background-color: rgb(238, 238, 238);"><li style="box-sizing: border-box; padding: 0px 5px;">1</li><li style="box-sizing: border-box; padding: 0px 5px;">2</li><li style="box-sizing: border-box; padding: 0px 5px;">3</li><li style="box-sizing: border-box; padding: 0px 5px;">4</li><li style="box-sizing: border-box; padding: 0px 5px;">5</li><li style="box-sizing: border-box; padding: 0px 5px;">6</li><li style="box-sizing: border-box; padding: 0px 5px;">7</li><li style="box-sizing: border-box; padding: 0px 5px;">8</li></ul><div class="save_code tracking-ad" data-mod="popu_249" style="position: absolute; top: 5px; visibility: hidden; box-sizing: border-box; height: 60px; right: 30px; color: rgb(255, 255, 255); cursor: pointer; z-index: 2;"><a target=_blank style="color: rgb(51, 102, 153); box-sizing: border-box;"><img src="http://static.blog.csdn.net/images/save_snippets.png" style="border: none; box-sizing: border-box; max-width: 100%;" alt="" /></a></div>
结果接近,效果还行。
如果来一个未对item1评分的user1,但是user1和item1的feature我们知道即上面的u1和v1,只需要u1 * v1.T即可得到5.31. 这样就可以有针对性的进行推荐了。