First cut of geometric asian pricer #82
Conversation
|
Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA). View this failed invocation of the CLA check for more information. For the most up to date status, view the checks section at the bottom of the pull request. |
| t2 = tf.reduce_sum(tf.map_fn(lambda x: tf.minimum(*tf.meshgrid(x, tf.transpose(x))), | ||
| tf.transpose(sampling_times)), [1,2]) / sample_count**2 |
There was a problem hiding this comment.
here use tf.vectorized_map and set fallback_to_while_loop to False. This should automatically vectorize what you want to do and provide significant speed up on a GPU device
There was a problem hiding this comment.
also, could you please add a comment explaining what is achieved here?
There was a problem hiding this comment.
Moved to vectorized_map
I've added an explanation comment above - let me know if this works or if you have a standard way of formatting math comments.
First cut of geometric asian pricer Code review updates
jackgillett101
force-pushed
the
master
branch
from
0b6190a
to
f9886f7
Compare
|
Thanks @jackgillett101 This is now submitted! |
Geometric asian option pricer under black scholes, this pricer transforms the asian problem into an equivalent vanilla problem and uses the vanilla pricer to price.
Affects this issue: #56