v

is there a reason you sampled from an uniform distribution (as opposed to a normal distribution)? just wondering if this was intentional or not!

a couple of updates to luaL_reg and luaL_openlib

@tc thanks for this awesome implementation... including jar which should be helpful to beginners like me!

included method to remove stopwords from text. here is an example with the blob text: ``` >>> blob = TextBlob(text) >>> blob.remove_stopwords ``` ... should produce the following output ```...

includes local caching to eliminate multiple (and unnecessary) socrata calls

thanks for this awesome implementation. fyi, the input gate `i = tf.sigmoid(tf.matmul(x,U[0]) + tf.matmul(st_1,W[0]))` typically uses tanh as the activation function (unlike other gates, which **should** use sigmoid).

include additional comments

1

@felixangell nice example. ive include comments to clarify difference between call stack and data stack.

graph algorithms

7

theres a rich body of graph algorithsm eg topological sort... if you think it makes sense ill implement and submit a pr :)

wrt: https://github.com/danielsabinasz/TensorSlow/blob/eb787eedb7929868c58f2fd3750011c8d6f30847/tensorslow/session.py#L45 for completeness sake, i think tensorflow performs topological sort rather than postorder traversal - happy to modify if you think it's necessary :-)

heres a cleaner way to calculate discounted rewards: `np.array([sum([gamma**t*r for t, r in enumerate(rewards[i:])]) for i in range(len(rewards))])`