markdown
stringlengths 0
37k
| code
stringlengths 1
33.3k
| path
stringlengths 8
215
| repo_name
stringlengths 6
77
| license
stringclasses 15
values | hash
stringlengths 32
32
|
---|---|---|---|---|---|
GenBank files don't have any per-letter annotations: | record.letter_annotations | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 12ccff5d1ecf84b9a96054244cab62a1 |
Most of the annotations information gets recorded in the \verb|annotations| dictionary, for example: | len(record.annotations)
record.annotations["source"] | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 1fb4f49da34cd06056571e960f19a6aa |
The dbxrefs list gets populated from any PROJECT or DBLINK lines: | record.dbxrefs | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 5958c66254ef3fa448dcdbc3c1559f0c |
Finally, and perhaps most interestingly, all the entries in the features table (e.g. the genes or CDS features) get recorded as SeqFeature objects in the features list. | len(record.features) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 3e17e84aaeb9e6ec28c5c0144fa8c5d6 |
Feature, location and position objects
SeqFeature objects
Sequence features are an essential part of describing a sequence. Once you get beyond the sequence itself, you need some way to organize and easily get at the more 'abstract' information that is known about the sequence. While it is probably impossible to develop a general sequence feature class that will cover everything, the Biopython SeqFeature class attempts to encapsulate as much of the information about the sequence as possible. The design is heavily based on the GenBank/EMBL feature tables, so if you understand how they look, you'll probably have an easier time grasping the structure of the Biopython classes.
The key idea about each SeqFeature object is to describe a region on a parent sequence, typically a SeqRecord object. That region is described with a location object, typically a range between two positions (see below).
The SeqFeature class has a number of attributes, so first we'll list them and their general features, and then later in the chapter work through examples to show how this applies to a real life example. The attributes of a SeqFeature are:
.type] - This is a textual description of the type of feature (for instance, this will be something like 'CDS' or 'gene').
.location - The location of the SeqFeature on the sequence
that you are dealing with. The
SeqFeature delegates much of its functionality to the location
object, and includes a number of shortcut attributes for properties
of the location:
.ref - shorthand for .location.ref - any (different)
reference sequence the location is referring to. Usually just None.
.ref_db - shorthand for .location.ref_db - specifies
the database any identifier in .ref refers to. Usually just None.
.strand - shorthand for .location.strand - the strand on
the sequence that the feature is located on. For double stranded nucleotide
sequence this may either be 1 for the top strand, -1 for the bottom
strand, 0 if the strand is important but is unknown, or None
if it doesn't matter. This is None for proteins, or single stranded sequences.
.qualifiers - This is a Python dictionary of additional information about the feature. The key is some kind of terse one-word description of what the information contained in the value is about, and the value is the actual information. For example, a common key for a qualifier might be 'evidence' and the value might be 'computational (non-experimental). This is just a way to let the person who is looking at the feature know that it has not be experimentally (i.e. in a wet lab) confirmed. Note that other the value will be a list of strings (even when there is only one string). This is a reflection of the feature tables in GenBank/EMBL files.
.sub_features - This used to be used to represent features with complicated locations like 'joins' in GenBank/EMBL files. This has been deprecated with the introduction of the CompoundLocation object, and should now be ignored.
Positions and locations
The key idea about each SeqFeature object is to describe a
region on a parent sequence, for which we use a location object,
typically describing a range between two positions. Two try to
clarify the terminology we're using:
position - This refers to a single position on a sequence,
which may be fuzzy or not. For instance, 5, 20, <100 and >200 are all positions.
location - A location is region of sequence bounded by
some positions. For instance 5..20 (i.e. 5 to 20) is a location.
I just mention this because sometimes I get confused between the two.
FeatureLocation object
Unless you work with eukaryotic genes, most SeqFeature locations are
extremely simple - you just need start and end coordinates and a strand.
That's essentially all the basic FeatureLocation object does.
In practise of course, things can be more complicated. First of all
we have to handle compound locations made up of several regions.
Secondly, the positions themselves may be fuzzy (inexact).
CompoundLocation object
Biopython 1.62 introduced the CompoundLocation as part of
a restructuring of how complex locations made up of multiple regions
are represented.
The main usage is for handling `join' locations in EMBL/GenBank files.
Fuzzy Positions
So far we've only used simple positions. One complication in dealing
with feature locations comes in the positions themselves.
In biology many times things aren't entirely certain
(as much as us wet lab biologists try to make them certain!). For
instance, you might do a dinucleotide priming experiment and discover
that the start of mRNA transcript starts at one of two sites. This
is very useful information, but the complication comes in how to
represent this as a position. To help us deal with this, we have
the concept of fuzzy positions. Basically there are several types
of fuzzy positions, so we have five classes do deal with them:
ExactPosition - As its name suggests, this class represents a position which is specified as exact along the sequence. This is represented as just a number, and you can get the position by looking at the position attribute of the object.
BeforePosition - This class represents a fuzzy position
that occurs prior to some specified site. In GenBank/EMBL notation,
this is represented as something like <13, signifying that
the real position is located somewhere less than 13. To get
the specified upper boundary, look at the position
attribute of the object.
AfterPosition - Contrary to BeforePosition, this
class represents a position that occurs after some specified site.
This is represented in GenBank as >13, and like
BeforePosition, you get the boundary number by looking
at the position attribute of the object.
WithinPosition - Occasionally used for GenBank/EMBL locations,
this class models a position which occurs somewhere between two
specified nucleotides. In GenBank/EMBL notation, this would be
represented as (1.5), to represent that the position is somewhere
within the range 1 to 5. To get the information in this class you
have to look at two attributes. The position attribute
specifies the lower boundary of the range we are looking at, so in
our example case this would be one. The extension attribute
specifies the range to the higher boundary, so in this case it
would be 4. So object.position is the lower boundary and
object.position + object.extension is the upper boundary.
OneOfPosition - Occasionally used for GenBank/EMBL locations,
this class deals with a position where several possible values exist,
for instance you could use this if the start codon was unclear and
there where two candidates for the start of the gene. Alternatively,
that might be handled explicitly as two related gene features.
UnknownPosition - This class deals with a position of unknown
location. This is not used in GenBank/EMBL, but corresponds to the `?'
feature coordinate used in UniProt.
Here's an example where we create a location with fuzzy end points: | from Bio import SeqFeature
start_pos = SeqFeature.AfterPosition(5)
end_pos = SeqFeature.BetweenPosition(9, left=8, right=9)
my_location = SeqFeature.FeatureLocation(start_pos, end_pos) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 4fecc3bac3a4b2ad967c6e4831e1f22a |
Note that the details of some of the fuzzy-locations changed in Biopython 1.59,
in particular for BetweenPosition and WithinPosition you must now make it explicit
which integer position should be used for slicing etc. For a start position this
is generally the lower (left) value, while for an end position this would generally
be the higher (right) value.
If you print out a FeatureLocation object, you can get a nice representation of the information: | print(my_location) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 34bd8b30ff9b97e1ea5c220394373264 |
We can access the fuzzy start and end positions using the start and end attributes of the location: | my_location.start
print(my_location.start)
my_location.end
print(my_location.end) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 18f0994d890a7f17d3019bb8c33c9599 |
If you don't want to deal with fuzzy positions and just want numbers,
they are actually subclasses of integers so should work like integers: | int(my_location.start)
int(my_location.end) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 0bcf8f24ed5bf5e73f63d85dbc9933e7 |
For compatibility with older versions of Biopython you can ask for the
\verb|nofuzzy_start| and \verb|nofuzzy_end| attributes of the location
which are plain integers: | my_location.nofuzzy_start
my_location.nofuzzy_end | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | cfba2f457e384c6d15e94dd866e72ab9 |
Notice that this just gives you back the position attributes of the fuzzy locations.
Similarly, to make it easy to create a position without worrying about fuzzy positions, you can just pass in numbers to the FeaturePosition constructors, and you'll get back out ExactPosition objects: | exact_location = SeqFeature.FeatureLocation(5, 9)
print(exact_location)
exact_location.start
print(int(exact_location.start))
exact_location.nofuzzy_start | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | ff41121428cad8ef5157038c4691521d |
That is most of the nitty gritty about dealing with fuzzy positions in Biopython.
It has been designed so that dealing with fuzziness is not that much more
complicated than dealing with exact positions, and hopefully you find that true!
Location testing
You can use the Python keyword in with a SeqFeature or location
object to see if the base/residue for a parent coordinate is within the
feature/location or not.
For example, suppose you have a SNP of interest and you want to know which
features this SNP is within, and lets suppose this SNP is at index 4350
(Python counting!). Here is a simple brute force solution where we just
check all the features one by one in a loop: | my_snp = 4350
record = SeqIO.read("data/NC_005816.gb", "genbank")
for feature in record.features:
if my_snp in feature:
print("%s %s" % (feature.type, feature.qualifiers.get('db_xref'))) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | a2b5929cce91d5b99d3e962faba5d636 |
Note that gene and CDS features from GenBank or EMBL files defined with joins
are the union of the exons -- they do not cover any introns.
Sequence described by a feature or location
A SeqFeature or location object doesn't directly contain a sequence, instead the location describes how to get this from the parent sequence. For example consider a (short) gene sequence with location 5:18 on the reverse strand, which in GenBank/EMBL notation using 1-based counting would be complement(6..18), like this: | from Bio.SeqFeature import SeqFeature, FeatureLocation
seq = Seq("ACCGAGACGGCAAAGGCTAGCATAGGTATGAGACTTCCTTCCTGCCAGTGCTGAGGAACTGGGAGCCTAC")
feature = SeqFeature(FeatureLocation(5, 18), type="gene", strand=-1) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | b870adf1a6356cc691ae353ca3648f0e |
You could take the parent sequence, slice it to extract 5:18, and then take the reverse complement.
If you are using Biopython 1.59 or later, the feature location's start and end are integer like so this works: | feature_seq = seq[feature.location.start:feature.location.end].reverse_complement()
print(feature_seq) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | d60eff22c69d83833d2e28ff4d376c06 |
This is a simple example so this isn't too bad -- however once you have to deal with compound features (joins) this is rather messy. Instead, the SeqFeature object has an extract method to take care of all this (and since Biopython 1.78 can handle trans-splicing by supplying a dictionary of referenced sequences): | feature_seq = feature.extract(seq)
print(feature_seq) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 82d6e12d4651410a42e06341b0a8af85 |
The length of a SeqFeature or location matches
that of the region of sequence it describes. | print(len(feature_seq))
print(len(feature))
print(len(feature.location)) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | beb6b93bb51be9876f9a688378ff9f58 |
For simple FeatureLocation objects the length is just the difference between the start and end positions. However, for a CompoundLocation the length is the sum of the constituent regions.
Comparison
The SeqRecord mobjects can be very complex, but hereโs a simple example: | from Bio.SeqRecord import SeqRecord
record1 = SeqRecord(Seq("ACGT"), id="test")
record2 = SeqRecord(Seq("ACGT"), id="test") | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 664923499a09c3887447bb23e328ee38 |
What happens when you try to compare these โidenticalโ records? | record1 == record2 | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | cdfdc3b1795fdc065902f2949e3866c1 |
Perhaps surprisingly older versions of Biopython would use Pythonโs default object comparison for theSeqRecord, meaning record1 == record2 would only return True if these variables pointed at the same object in memory. In this example, record1 == record2 would have returned False here! | record1 == record2 # on old versions of Biopython! | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 22a21493396dd5689556b01424624021 |
False
As of Biopython 1.67, SeqRecord comparison like record1 == record2 will instead raise an explicit error to avoid people being caught out by this: | record1 == record2 | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 989b38259b877b2ad0206c19582ed1db |
Instead you should check the attributes you are interested in, for example the identifier and the sequence: | record1.id == record2.id
record1.seq == record2.seq | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 175e4374f37df8cac4f845364f7bd1d8 |
Beware that comparing complex objects quickly gets complicated.
References
Another common annotation related to a sequence is a reference to a journal or other published work dealing with the sequence. We have a fairly simple way of representing a Reference in Biopython -- we have a Bio.SeqFeature.Reference class that stores the relevant information about a reference as attributes of an object.
The attributes include things that you would expect to see in a reference like journal, title and authors. Additionally, it also can hold the medline_id and pubmed_id and a comment about the reference. These are all accessed simply as attributes of the object.
A reference also has a location object so that it can specify a particular location on the sequence that the reference refers to. For instance, you might have a journal that is dealing with a particular gene located on a BAC, and want to specify that it only refers to this position exactly. The location is a potentially fuzzy location.
Any reference objects are stored as a list in the SeqRecord object's annotations dictionary under the key 'references'.
That's all there is too it. References are meant to be easy to deal with, and hopefully general enough to cover lots of usage cases.
The format method
The format method of the SeqRecord class gives a string
containing your record formatted using one of the output file formats
supported by Bio.SeqIO, such as FASTA: | record = SeqRecord(
Seq(
"MMYQQGCFAGGTVLRLAKDLAENNRGARVLVVCSEITAVTFRGPSETHLDSMVGQALFGD"
"GAGAVIVGSDPDLSVERPLYELVWTGATLLPDSEGAIDGHLREVGLTFHLLKDVPGLISK"
"NIEKSLKEAFTPLGISDWNSTFWIAHPGGPAILDQVEAKLGLKEEKMRATREVLSEYGNM"
"SSAC"
),
id="gi|14150838|gb|AAK54648.1|AF376133_1",
description="chalcone synthase [Cucumis sativus]",
)
print(record.format("fasta")) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 426c653c59a6d35dcf1c33e8480f01cc |
This format method takes a single mandatory argument, a lower case string which is
supported by Bio.SeqIO as an output format.
However, some of the file formats Bio.SeqIO can write to require more than
one record (typically the case for multiple sequence alignment formats), and thus won't
work via this format() method.
Slicing a SeqRecord
You can slice a SeqRecord, to give you a new SeqRecord covering just
part of the sequence. What is important
here is that any per-letter annotations are also sliced, and any features which fall
completely within the new sequence are preserved (with their locations adjusted).
For example, taking the same GenBank file used earlier: | record = SeqIO.read("data/NC_005816.gb", "genbank")
print(record)
len(record)
len(record.features) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 85bb342dfd1df1fd2f87819e61871346 |
For this example we're going to focus in on the pim gene, YP_pPCP05.
If you have a look at the GenBank file directly you'll find this gene/CDS has
location string 4343..4780, or in Python counting 4342:4780.
From looking at the file you can work out that these are the twelfth and
thirteenth entries in the file, so in Python zero-based counting they are
entries 11 and 12 in the features list: | print(record.features[20])
print(record.features[21]) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 0c4c7a35f3f1d4f0e07c1b2dfd03efcb |
Let's slice this parent record from 4300 to 4800 (enough to include the pim
gene/CDS), and see how many features we get: | sub_record = record[4300:4800]
sub_record
len(sub_record)
len(sub_record.features) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 0a079704bf8a2e00a27fc76021d1e228 |
Our sub-record just has two features, the gene and CDS entries for YP_pPCP05: | print(sub_record.features[0])
print(sub_record.features[1]) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | e50b9d2db665c3b4f14aaa0d3f1bbb38 |
Notice that their locations have been adjusted to reflect the new parent sequence!
While Biopython has done something sensible and hopefully intuitive with the features
(and any per-letter annotation), for the other annotation it is impossible to know if
this still applies to the sub-sequence or not. To avoid guessing, the annotations
and dbxrefs are omitted from the sub-record, and it is up to you to transfer
any relevant information as appropriate. | print(sub_record.annotations)
print(sub_record.dbxrefs) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 457174c51c956ed382979371b5ef3965 |
The same point could be made about the record id, name
and description, but for practicality these are preserved: | print(sub_record.id)
print(sub_record.name)
print(sub_record.description) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | c2280e8ec682fafa0c11791f0e2f8966 |
This illustrates the problem nicely though, our new sub-record is
not the complete sequence of the plasmid, so the description is wrong!
Let's fix this and then view the sub-record as a reduced FASTA file using
the format method described above: | sub_record.description ="Yersinia pestis biovar Microtus str. 91001 plasmid pPCP1, partial."
print(sub_record.format("fasta")) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 986970a35f4108d38c7e9de7ff5643b9 |
Adding SeqRecord objects
You can add SeqRecord objects together, giving a new SeqRecord.
What is important here is that any common
per-letter annotations are also added, all the features are preserved (with their
locations adjusted), and any other common annotation is also kept (like the id, name
and description).
For an example with per-letter annotation, we'll use the first record in a
FASTQ file. | record = next(SeqIO.parse("data/example.fastq", "fastq"))
print(len(record))
print(record.seq)
print(record.letter_annotations["phred_quality"]) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | b165b5c084a140dc93dee1335abbff05 |
Let's suppose this was Roche 454 data, and that from other information
you think the TTT should be only TT. We can make a new edited
record by first slicing the SeqRecord before and after the 'extra'
third T: | left = record[:20]
print(left.seq)
print(left.letter_annotations["phred_quality"])
right = record[21:]
print(right.seq)
print(right.letter_annotations["phred_quality"]) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 45cd8fb662b14e327d905cce5cb78101 |
Now add the two parts together: | edited = left + right
print(len(edited))
print(edited.seq)
print(edited.letter_annotations["phred_quality"]) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | bb3d2041172f7e4f0dc17d83e3c36ed1 |
Easy and intuitive? We hope so! You can make this shorter with just: | edited = record[:20] + record[21:] | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | b4c798f63f4d20083bc7a39bd311138b |
Now, for an example with features, we'll use a GenBnak file.
Suppose you have a circular genome: | record = SeqIO.read("data/NC_005816.gb", "genbank")
print(record) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 76d1eaa535bbe928b85c3f8da7ed8755 |
You can shift the origin like this: | print(len(record))
print(len(record.features))
print(record.dbxrefs)
print(record.annotations.keys()) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 1d201740f8aeafcd7fd64dba5cecec8d |
You can shift the origin like this: | shifted = record[2000:] + record[:2000]
print(shifted)
print(len(shifted)) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 477f4ad9dc86165b56fd0125cdf83f47 |
Note that this isn't perfect in that some annotation like the database cross references
and one of the features (the source feature) have been lost: | print(len(shifted.features))
print(shifted.dbxrefs)
print(shifted.annotations.keys()) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 0e2b783c0fbab732ae75b86bd2f933da |
This is because the SeqRecord slicing step is cautious in what annotation
it preserves (erroneously propagating annotation can cause major problems). If
you want to keep the database cross references or the annotations dictionary,
this must be done explicitly: | shifted.dbxrefs = record.dbxrefs[:]
shifted.annotations = record.annotations.copy()
print(shifted.dbxrefs)
print(shifted.annotations.keys()) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | ae947ca763ecc63c9f3ded0d6b4e1ed5 |
Also note that in an example like this, you should probably change the record
identifiers since the NCBI references refer to the original unmodified
sequence.
Reverse-complementing SeqRecord objects
One of the new features in Biopython 1.57 was the SeqRecord object's
reverse_complement method. This tries to balance easy of use with worries
about what to do with the annotation in the reverse complemented record.
For the sequence, this uses the Seq object's reverse complement method. Any
features are transferred with the location and strand recalculated. Likewise
any per-letter-annotation is also copied but reversed (which makes sense for
typical examples like quality scores). However, transfer of most annotation
is problematical.
For instance, if the record ID was an accession, that accession should not really
apply to the reverse complemented sequence, and transferring the identifier by
default could easily cause subtle data corruption in downstream analysis.
Therefore by default, the SeqRecord's id, name, description, annotations
and database cross references are all not transferred by default.
The SeqRecord object's reverse_complement method takes a number
of optional arguments corresponding to properties of the record. Setting these
arguments to True means copy the old values, while False means
drop the old values and use the default value. You can alternatively provide
the new desired value instead.
Consider this example record: | record = SeqIO.read("data/NC_005816.gb", "genbank")
print("%s %i %i %i %i" % (record.id, len(record), len(record.features), len(record.dbxrefs), len(record.annotations))) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 67d1c067de465788123aedff4c600e38 |
Here we take the reverse complement and specify a new identifier - but notice
how most of the annotation is dropped (but not the features): | rc = record.reverse_complement(id="TESTING")
print("%s %i %i %i %i" % (rc.id, len(rc), len(rc.features), len(rc.dbxrefs), len(rc.annotations))) | notebooks/04 - Sequence Annotation objects.ipynb | tiagoantao/biopython-notebook | mit | 0b2ee7a7e20f83cbc5fa72548fd1ec24 |
Banana-shaped target distribution | dtarget = lambda x: exp( (-x[0]**2)/200. - 0.5*(x[1]+(0.05*x[0]**2) - 100.*0.05)**2)
x1 = np.linspace(-20, 20, 101)
x2 = np.linspace(-15, 10, 101)
X, Y = np.meshgrid(x1, x2)
Z = np.array(map(dtarget, zip(X.flat, Y.flat))).reshape(101, 101)
plt.figure(figsize=(10,7))
plt.contour(X, Y, Z)
plt.show()
start = np.array([[2., 5.]
for i in xrange(4)])
chains = HMC(dtarget, start, Eps=0.5, L=200, m=0.5, N=5000)
plt.figure(figsize=(10,7))
plt.contour(X, Y, Z)
plt.plot(chains[0][:, 0], chains[0][:, 1], alpha=0.8)
plt.plot(chains[1][:, 0], chains[1][:, 1], alpha=0.8)
plt.plot(chains[2][:, 0], chains[2][:, 1], alpha=0.8)
plt.plot(chains[3][:, 0], chains[3][:, 1], alpha=0.8)
plt.show()
plt.subplot(211)
plt.title(Gelman(chains)[0])
for i in xrange(chains.shape[0]):
plt.plot(chains[i,:,0])
plt.ylabel('x1')
plt.subplot(212)
for i in xrange(chains.shape[0]):
plt.plot(chains[i,:,1])
plt.ylabel('x2')
plt.tight_layout()
plt.show() | Hamiltonian MCMC (HMC).ipynb | erickpeirson/statistical-computing | cc0-1.0 | a640b64584f4c84dbdef7c557c5c343c |
Retrieving training and test data
The MNIST data set already contains both training and test data. There are 55,000 data points of training data, and 10,000 points of test data.
Each MNIST data point has:
1. an image of a handwritten digit and
2. a corresponding label (a number 0-9 that identifies the image)
We'll call the images, which will be the input to our neural network, X and their corresponding labels Y.
We're going to want our labels as one-hot vectors, which are vectors that holds mostly 0's and one 1. It's easiest to see this in a example. As a one-hot vector, the number 0 is represented as [1, 0, 0, 0, 0, 0, 0, 0, 0, 0], and 4 is represented as [0, 0, 0, 0, 1, 0, 0, 0, 0, 0].
Flattened data
For this example, we'll be using flattened data or a representation of MNIST images in one dimension rather than two. So, each handwritten number image, which is 28x28 pixels, will be represented as a one dimensional array of 784 pixel values.
Flattening the data throws away information about the 2D structure of the image, but it simplifies our data so that all of the training data can be contained in one array whose shape is [55000, 784]; the first dimension is the number of training images and the second dimension is the number of pixels in each image. This is the kind of data that is easy to analyze using a simple neural network. | # Retrieve the training and test data
trainX, trainY, testX, testY = mnist.load_data(one_hot=True)
trainX[0] | tutorials/intro-to-tflearn/TFLearn_Digit_Recognition.ipynb | wbbeyourself/cn-deep-learning | mit | eb06859998dde796c55a2bf9a6e5a17c |
Visualize the training data
Provided below is a function that will help you visualize the MNIST data. By passing in the index of a training example, the function show_digit will display that training image along with it's corresponding label in the title. | # Visualizing the data
import matplotlib.pyplot as plt
%matplotlib inline
# Function for displaying a training image by it's index in the MNIST set
def show_digit(index):
label = trainY[index].argmax(axis=0)
# Reshape 784 array into 28x28 image
image = trainX[index].reshape([28,28])
plt.title('Training data, index: %d, Label: %d' % (index, label))
plt.imshow(image, cmap='gray_r')
plt.show()
# Display the first (index 0) training image
show_digit(3) | tutorials/intro-to-tflearn/TFLearn_Digit_Recognition.ipynb | wbbeyourself/cn-deep-learning | mit | 0546289dbeeec636b377ec0cbbef5bac |
Building the network
TFLearn lets you build the network by defining the layers in that network.
For this example, you'll define:
The input layer, which tells the network the number of inputs it should expect for each piece of MNIST data.
Hidden layers, which recognize patterns in data and connect the input to the output layer, and
The output layer, which defines how the network learns and outputs a label for a given image.
Let's start with the input layer; to define the input layer, you'll define the type of data that the network expects. For example,
net = tflearn.input_data([None, 100])
would create a network with 100 inputs. The number of inputs to your network needs to match the size of your data. For this example, we're using 784 element long vectors to encode our input data, so we need 784 input units.
Adding layers
To add new hidden layers, you use
net = tflearn.fully_connected(net, n_units, activation='ReLU')
This adds a fully connected layer where every unit (or node) in the previous layer is connected to every unit in this layer. The first argument net is the network you created in the tflearn.input_data call, it designates the input to the hidden layer. You can set the number of units in the layer with n_units, and set the activation function with the activation keyword. You can keep adding layers to your network by repeated calling tflearn.fully_connected(net, n_units).
Then, to set how you train the network, use:
net = tflearn.regression(net, optimizer='sgd', learning_rate=0.1, loss='categorical_crossentropy')
Again, this is passing in the network you've been building. The keywords:
optimizer sets the training method, here stochastic gradient descent
learning_rate is the learning rate
loss determines how the network error is calculated. In this example, with categorical cross-entropy.
Finally, you put all this together to create the model with tflearn.DNN(net).
Exercise: Below in the build_model() function, you'll put together the network using TFLearn. You get to choose how many layers to use, how many hidden units, etc.
Hint: The final output layer must have 10 output nodes (one for each digit 0-9). It's also recommended to use a softmax activation layer as your final output layer. | # Define the neural network
def build_model():
# This resets all parameters and variables, leave this here
tf.reset_default_graph()
#### Your code ####
# Include the input layer, hidden layer(s), and set how you want to train the model
net = tflearn.input_data([None, 784])
net = tflearn.fully_connected(net, n_units=200, activation='ReLU')
net = tflearn.fully_connected(net, n_units=30, activation='ReLU')
net = tflearn.fully_connected(net, n_units=10, activation='softmax')
net = tflearn.regression(net, optimizer='sgd', learning_rate=0.1, loss='categorical_crossentropy')
# This model assumes that your network is named "net"
model = tflearn.DNN(net)
return model
# Build the model
model = build_model() | tutorials/intro-to-tflearn/TFLearn_Digit_Recognition.ipynb | wbbeyourself/cn-deep-learning | mit | 563e116090830340a48d088824ca4588 |
'Hello World!' is a data structure called a string. | type('Hello World')
4+5
type(9.)
4 * 5
4**5 # exponentiation
# naming things and storing them in memory for later use
x = 2**10
print(x)
whos
# with explanation
print('The value of x is {:,}.'.format(x)) # you can change the formatting of x inside the brackets
type(3.14159)
print('The value of pi is approximately {0:.2f}.'.format(3.14159)) # 0 is the argument, .2 means two past . | notebooks/introduction_to_python.ipynb | jwjohnson314/data-801 | mit | 0347aa878f94c3fb817f63ccba6ff563 |
Lists
Lists are a commonly used python data structure. | x = [1, 2, 3]
type(x)
whos
x.append(4)
x
# throws an error
x.prepend(0)
y = [0]
x+y
y+x
whos
# didn't save it - let's do it again
y = y+x
y
# Exercise: there is a more efficient way - find the reference in the docs for the insert command.
# insert the value 2.5 into the list into the appropriate spot
# your code here:
y.insert(3, 2.5)
print(y)
# a bigger list - list is a function too
z = list(range(100)) # range is a special type in Python
print(z)
# getting help
?range
# try shift+tab when calling unfamiliar function for quick access to docstring
range()
# Exercise: get the docstring for the 'open' function
# your code here:
open()
# Exercise: get the docstring for the 'list' function
# your code here:
list()
# often we need to get extract elements from a list. Python uses zero-based indexing
print(z[0])
print(z[5])
# ranges/slices
print(z[4:5]) # 4 is included
print(z[4:]) # 4 is included
print(z[:4]) # 4 is not included
z[2:4] + z[7:9]
# Exercise: write a list consisting of the entries in z whose first digit is a prime number
# your code here:
# from the end of the list
z[-10:]
# by step size other than 1
z[10:20:2] # start:stop:step
# when you're going all the way to the end
z[10::2] # stop omitted
# exercise: can you write a single operation to return z in reversed order?
# your code here:
z[::-1]
# removing values
z.remove(2)
print(z[:10])
# strings are a lot like lists
string = 'This is A poOrly cAPitAlized string.'
string[:4]
type(string[:4])
string[::2]
string[-1]
print(string.lower())
print(string.upper())
print(string.split('A'))
type(string.split('A'))
address = 'http://www.wikiart.org/en/jan-van-eyck/the-birth-of-john-the-baptist-1422'
artist, painting = address.split('/')[4:]
print(artist)
print(painting)
# digression - unicode
ord('.') # encoding
chr(97)
# homework reading: http://www.joelonsoftware.com/articles/Unicode.html
# string arithmetic
x = 'Hello '
y = 'World'
print(x+y)
x[:5] + '_' + y
x.replace(' ', '_')
x.append('_')
x
x.replace(' ', '_') + y
# strings are not exactly lists!
x*5 + y.append(' ')*3 | notebooks/introduction_to_python.ipynb | jwjohnson314/data-801 | mit | 743664e22d5989487d7f51453eca0ce1 |
Exercise: take a few minutes to read the docs for text strings here: https://docs.python.org/3/library/stdtypes.html#textseq
Immutable means 'can't be changed'
So if you want to change a string, you need to make a copy of some sort. | x * 5 + (y + str(' ')) * 3 | notebooks/introduction_to_python.ipynb | jwjohnson314/data-801 | mit | 3045f20202c7438744aeb084da3730f3 |
Tuples
Exercise: Find the doc page for the tuples datatype. What is the difference between a tuple and a list? | # Exercise: write a tuple consisting of the first five letters of the alphabet (lower-case) in reversed order
# your code here
tup = ('z', 'y', 'x', 'w', 'v')
type(tup)
tup[3] | notebooks/introduction_to_python.ipynb | jwjohnson314/data-801 | mit | 0e0449442d64e97335a2a942d72deaee |
Dicts
The dictionary data structure consists of key-value pairs. This shows up a lot; for instance, when reading JSON files (http://www.json.org/) | x = ['Bob', 'Amy', 'Fred']
y = [32, 27, 19]
z = dict(zip(x, y))
type(z)
z
z[1]
z['Bob']
z.keys()
z.values()
detailed ={'amy': {'age': 32, 'school': 'UNH', 'GPA':4.0}, 'bob': {'age': 27, 'school': 'UNC', 'GPA':3.4}}
detailed['amy']['school']
# less trivial example
# library imports; ignore for now
from urllib.request import urlopen
import json
url = 'http://www.wikiart.org/en/App/Painting/' + \
'PaintingsByArtist?artistUrl=' + \
'pablo-picasso' + '&json=2'
raw = urlopen(url).read().decode('utf8')
d = json.loads(raw)
type(d)
d
type(d[0])
d[0].keys() | notebooks/introduction_to_python.ipynb | jwjohnson314/data-801 | mit | d220a996f895860a318fba55984dc4d7 |
Control structures: the 'for' loop | # indents matter in Python
for i in range(20):
print('%s: %s' % (d[i]['title'], d[i]['completitionYear']))
# exercises: print the sizes and titles of the last ten paintings in this list.
# The statement should print as 'title: width pixels x height pixels'
# your code here: | notebooks/introduction_to_python.ipynb | jwjohnson314/data-801 | mit | 463620e2de50f4f83a878327b5f73038 |
The 'if-then' statement | data = [1.2, 2.4, 23.3, 4.5]
new_data = []
for i in range(len(data)):
if round(data[i]) % 2 == 0: # modular arithmetic, remainder of 0
new_data.append(round(data[i]))
else:
new_data.append(0)
print(new_data)
| notebooks/introduction_to_python.ipynb | jwjohnson314/data-801 | mit | 7d26af2272581ea05a575612f9841182 |
Digression - list comprehensions
Rather than a for loop, in a situation like that above, Python has a method called a list comprehension for creating lists. Sometimes this is more efficient. It's often nicer syntactically, as long as the number of conditions is not too large (<= 2 is a good guideline). | print(data)
new_new_data = [round(i) if round(i) % 2 == 0 else 0 for i in data]
print(new_new_data)
data = list(range(20))
for i in data:
if i % 2 == 0:
print(i)
elif i >= 10:
print('wow, that\'s a big odd number - still no fun')
else:
print('odd num no fun')
| notebooks/introduction_to_python.ipynb | jwjohnson314/data-801 | mit | 2a79d2256a4e8cd17c4646aadf565258 |
The 'while' loop | # beware loops that don't terminate
counter = 0
tmp = 2
while counter < 10:
tmp = tmp**2
counter += 1
print('{:,}'.format(tmp))
print('tmp is %d digits long, that\'s huge!' % len(str(tmp)))
# the 'pass' command
for i in range(10):
if i % 2 == 0:
print(i)
else:
pass
# the continue command
for letter in 'Python':
if letter == 'h':
continue
print('Current Letter :', letter)
# the pass command
for letter in 'Python':
if letter == 'h':
pass
print('Current Letter :', letter)
# the break command
for letter in 'Python':
if letter == 'h':
break
print('Current Letter :', letter) | notebooks/introduction_to_python.ipynb | jwjohnson314/data-801 | mit | 5ac591133d4368e5e98475467d49e5b0 |
Functions
Functions take in inputs and produce outputs. | def square(x):
'''input: a numerical value x
output: the square of x
'''
return x**2
square(3.14)
# Exercise: write a function called 'reverse' to take in a string and reverse it
# your code here:
# test
reverse('Hi, my name is Joan Jett')
def raise_to_power(x, n=2): # 2 is the default for n
return x**n
raise_to_power(3)
raise_to_power(3,4)
def write_to_file(filepath, string):
'''make sure the file doesn\'t exist; this will overwrite'''
with open(filepath, 'w+') as f:
f.writelines(string)
write_to_file('test.txt', 'fred was here')
! cat test.txt
with open('test.txt') as f:
content = f.read()
print(content)
write_to_file('test.txt', 'goodbye for now\n') # \n is the newline character
! cat test.txt
# Exercise: what are the modes for editing a file? | notebooks/introduction_to_python.ipynb | jwjohnson314/data-801 | mit | 74f9a9e062fa3413cf551d5d65f51ee2 |
Make a PMF of <tt>numkdhh</tt>, the number of children under 18 in the respondent's household. | numkdhh = thinkstats2.Pmf(resp.numkdhh)
numkdhh | code/chap03ex.ipynb | goodwordalchemy/thinkstats_notes_and_exercises | gpl-3.0 | 724ccf4fb44948493b66ce4f5dfb199f |
Display the PMF. | thinkplot.Hist(numkdhh, label='actual')
thinkplot.Config(title="PMF of num children under 18",
xlabel="number of children under 18",
ylabel="probability") | code/chap03ex.ipynb | goodwordalchemy/thinkstats_notes_and_exercises | gpl-3.0 | 96547581da40b0193b91a8cd2521af3c |
Make a the biased Pmf of children in the household, as observed if you surveyed the children instead of the respondents. | biased_pmf = BiasPmf(numkdhh, label='biased')
thinkplot.Hist(biased_pmf)
thinkplot.Config(title="PMF of num children under 18",
xlabel="number of children under 18",
ylabel="probability") | code/chap03ex.ipynb | goodwordalchemy/thinkstats_notes_and_exercises | gpl-3.0 | 65bc74150d58a0c5f50e8772cd6f3f0b |
Display the actual Pmf and the biased Pmf on the same axes. | width = 0.45
thinkplot.PrePlot(2)
thinkplot.Hist(biased_pmf, align="right", label="biased", width=width)
thinkplot.Hist(numkdhh, align="left", label="actual", width=width)
thinkplot.Config(title="PMFs of children under 18 in a household",
xlabel='number of children',
ylabel='probability') | code/chap03ex.ipynb | goodwordalchemy/thinkstats_notes_and_exercises | gpl-3.0 | 563959976b4ee1aff75351cc69b2c6e3 |
Compute the means of the two Pmfs. | print "actual mean:", numkdhh.Mean()
print "biased mean:", biased_pmf.Mean() | code/chap03ex.ipynb | goodwordalchemy/thinkstats_notes_and_exercises | gpl-3.0 | adf9ee36546bd6abb00f78fed2eed682 |
Verification of the FUSED-Wind wrapper
common inputs | v80 = wt.WindTurbine('Vestas v80 2MW offshore','V80_2MW_offshore.dat',70,40)
HR1 = wf.WindFarm('Horns Rev 1','HR_coordinates.dat',v80)
WD = range(0,360,1) | examples/Script.ipynb | rethore/FUSED-Wake | agpl-3.0 | 4761e86ee711a075e6ca660682fbe786 |
The following figure shows the distribution of the sum of three dice, pmf_3d6, and the distribution of the best three out of four, pmf_best3. | pmf_3d6.plot(label='sum of 3 dice')
pmf_best3.plot(label='best 3 of 4', style='--')
decorate_dice('Distribution of attributes') | notebooks/chap07.ipynb | AllenDowney/ThinkBayes2 | mit | 5a19d9d2e5d7c3a08c53592df6663c81 |
Most characters have at least one attribute greater than 12; almost 10% of them have an 18.
The following figure shows the CDFs for the three distributions we have computed. | import matplotlib.pyplot as plt
cdf_3d6 = pmf_3d6.make_cdf()
cdf_3d6.plot(label='sum of 3 dice')
cdf_best3 = pmf_best3.make_cdf()
cdf_best3.plot(label='best 3 of 4 dice', style='--')
cdf_max6.plot(label='max of 6 attributes', style=':')
decorate_dice('Distribution of attributes')
plt.ylabel('CDF'); | notebooks/chap07.ipynb | AllenDowney/ThinkBayes2 | mit | 1e9bc657e594e0bee5d6e0fa3310a8f6 |
Here's what it looks like, along with the distribution of the maximum. | cdf_min6.plot(color='C4', label='minimum of 6')
cdf_max6.plot(color='C2', label='maximum of 6', style=':')
decorate_dice('Minimum and maximum of six attributes')
plt.ylabel('CDF'); | notebooks/chap07.ipynb | AllenDowney/ThinkBayes2 | mit | d7a4b5087dcdff01f4fde2d21e7a1f60 |
We can compare it to the distribution of attributes you get by rolling four dice at adding up the best three. | cdf_best3.plot(label='best 3 of 4', color='C1', style='--')
cdf_standard.step(label='standard set', color='C7')
decorate_dice('Distribution of attributes')
plt.ylabel('CDF'); | notebooks/chap07.ipynb | AllenDowney/ThinkBayes2 | mit | 965d934722daaf891966d39dea6b0878 |
I plotted cdf_standard as a step function to show more clearly that it contains only a few quantities. | # Solution goes here
# Solution goes here
# Solution goes here
# Solution goes here
# Solution goes here
# Solution goes here | notebooks/chap07.ipynb | AllenDowney/ThinkBayes2 | mit | 5a48a3f60a3a414e10ad08224816b206 |
Exercise: Suppose you are fighting three monsters:
One is armed with a short sword that causes one 6-sided die of damage,
One is armed with a battle axe that causes one 8-sided die of damage, and
One is armed with a bastard sword that causes one 10-sided die of damage.
One of the monsters, chosen at random, attacks you and does 1 point of damage.
Which monster do you think it was? Compute the posterior probability that each monster was the attacker.
If the same monster attacks you again, what is the probability that you suffer 6 points of damage?
Hint: Compute a posterior distribution as we have done before and pass it as one of the arguments to make_mixture. | # Solution goes here
# Solution goes here
# Solution goes here
# Solution goes here | notebooks/chap07.ipynb | AllenDowney/ThinkBayes2 | mit | ca01d82f4820eec7f1aca15664838460 |
Exercise: Henri Poincarรฉ was a French mathematician who taught at the Sorbonne around 1900. The following anecdote about him is probably fiction, but it makes an interesting probability problem.
Supposedly Poincarรฉ suspected that his local bakery was selling loaves of bread that were lighter than the advertised weight of 1 kg, so every day for a year he bought a loaf of bread, brought it home and weighed it. At the end of the year, he plotted the distribution of his measurements and showed that it fit a normal distribution with mean 950 g and standard deviation 50 g. He brought this evidence to the bread police, who gave the baker a warning.
For the next year, Poincarรฉ continued to weigh his bread every day. At the end of the year, he found that the average weight was 1000 g, just as it should be, but again he complained to the bread police, and this time they fined the baker.
Why? Because the shape of the new distribution was asymmetric. Unlike the normal distribution, it was skewed to the right, which is consistent with the hypothesis that the baker was still making 950 g loaves, but deliberately giving Poincarรฉ the heavier ones.
To see whether this anecdote is plausible, let's suppose that when the baker sees Poincarรฉ coming, he hefts n loaves of bread and gives Poincarรฉ the heaviest one. How many loaves would the baker have to heft to make the average of the maximum 1000 g?
To get you started, I'll generate a year's worth of data from a normal distribution with the given parameters. | mean = 950
std = 50
np.random.seed(17)
sample = np.random.normal(mean, std, size=365)
# Solution goes here
# Solution goes here | notebooks/chap07.ipynb | AllenDowney/ThinkBayes2 | mit | a0a63606888308299df9833be12a85ad |
In the meanwhile we are trying to have more information about pandas. In the following sections we are using the value_counts method to have more information about each feature values. This method specify number of different values for given feature. | housing['total_rooms'].value_counts()
housing['ocean_proximity'].value_counts() | ml/housing/Housing.ipynb | 1995parham/Learning | gpl-2.0 | 4f3cb937292a322b36c4d7f64f4142b3 |
See the difference between loc and iloc methods in a simple pandas DataFrame. | pd.DataFrame([{'a': 1, 'b': '1'}, {'a': 2, 'b': 1}, {'a': 3, 'b': 1}]).iloc[1]
pd.DataFrame([{'a': 1, 'b': '1'}, {'a': 2, 'b': 1}, {'a': 3, 'b': 1}]).loc[1]
pd.DataFrame([{'a': 1, 'b': '1'}, {'a': 2, 'b': 1}, {'a': 3, 'b': 1}]).loc[1, ['b']]
pd.DataFrame([{'a': 1, 'b': '1'}, {'a': 2, 'b': 1}, {'a': 3, 'b': 1}]).loc[[True, True, False]] | ml/housing/Housing.ipynb | 1995parham/Learning | gpl-2.0 | 37a339ea299b1723110c8482f276104f |
Here we want to see the apply function of pandas for an specific feature. | pd.DataFrame([{'a': 1, 'b': '1'}, {'a': 2, 'b': 1}, {'a': 3, 'b': 1}])['a'].apply(lambda a: a > 10) | ml/housing/Housing.ipynb | 1995parham/Learning | gpl-2.0 | a796658720caf4b6ae474f34a617f7ea |
The following function helps to split the given dataset into test and train sets. | from zlib import crc32
import numpy as np
def test_set_check(identifier, test_ratio):
return crc32(np.int64(identifier)) & 0xffffffff < test_ratio * 2**32
def split_train_test_by_id(data, test_ratio, id_column):
ids = data[id_column]
in_test_set = ids.apply(lambda _id: test_set_check(_id, test_ratio))
return data.loc[~in_test_set], data.loc[in_test_set]
housing_with_id = housing.reset_index() # adds an "index" column
train_set, test_set = split_train_test_by_id(housing_with_id, 0.2, 'index')
housing = train_set.copy()
housing.plot(kind="scatter", x="longitude", y="latitude", alpha=0.1)
import matplotlib.pyplot as plt
housing.plot(kind='scatter', x='longitude', y='latitude',
alpha=0.4, s=housing['population']/100, label='population',
c='median_house_value', cmap=plt.get_cmap('jet'), colorbar=True,
) | ml/housing/Housing.ipynb | 1995parham/Learning | gpl-2.0 | a8ae4aae596afb02673a09a1cf3adc1a |
Below is a plot of the signal. | plt.figure(figsize=(figWidth, 4))
plt.plot(signalTime, signalSamples)
plt.xlabel("t")
plt.ylabel("Amplitude")
plt.suptitle('Source Signal')
plt.show() | src/articles/PDMPlayground/index.ipynb | bradhowes/keystrokecountdown | mit | 1f3992c0a329c1e681168e6a79b33589 |
To verify that the signal really has only two frequency components, here is the output of the FFT for it. | fftFreqs = np.arange(bandwidth)
fftValues = (np.fft.fft(signalSamples) / sampleFrequency)[:int(bandwidth)]
plt.plot(fftFreqs, np.absolute(fftValues))
plt.xlim(0, bandwidth)
plt.ylim(0, 0.3)
plt.xlabel("Frequency")
plt.ylabel("Magnitude")
plt.suptitle("Source Signal Frequency Components")
plt.show() | src/articles/PDMPlayground/index.ipynb | bradhowes/keystrokecountdown | mit | 59b8608f7e76d3bfdd69b0bfec8a7530 |
PDM Modulation
Now that we have a signal to work with, next step is to generate a pulse train from it. The code below is a simple hack that generates 64 samples for every one in the original signal. Normally, this would involve interpolation so that the 63 additional samples vary linearly from the previous sample to the current one. This lack will introduce some noise due to discontinuities.
The setting pdmFreq is the number of samples to create for each element in signalSamples. | pdmFreq = 64
pdmPulses = np.empty(sampleFrequency * pdmFreq)
pdmTime = np.arange(0, pdmPulses.size)
pdmIndex = 0
signalIndex = 0
quantizationError = 0
while pdmIndex < pdmPulses.size:
sample = signalSamples[signalIndex]
signalIndex += 1
for tmp in range(pdmFreq):
if sample >= quantizationError:
bit = 1
else:
bit = -1
quantizationError = bit - sample + quantizationError
pdmPulses[pdmIndex] = bit
pdmIndex += 1
print(pdmIndex, signalIndex, pdmPulses.size, signalSamples.size) | src/articles/PDMPlayground/index.ipynb | bradhowes/keystrokecountdown | mit | 3a6bcc19f2400f146de0d3c9089997c8 |
Visualize the first 4K PDM samples. We should be able to clearly see the pulsing. | span = 1024
plt.figure(figsize=(16, 6))
counter = 1
for pos in range(0, pdmIndex, span):
from matplotlib.ticker import MultipleLocator
plt.subplot(4, 1, counter)
counter += 1
# Generate a set of time values that correspond to pulses with +1 values. Remove the rest
# and plot.
plt.vlines(np.delete(pdmTime[pos:pos + span], np.nonzero(pdmPulses[pos:pos + span] > 0.0)[0]), 0, 1, 'g')
plt.ylim(0, 1)
plt.xlim(pos, pos + span)
plt.tick_params(axis='both', which='major', labelsize=8)
ca = plt.gca()
axes = ca.axes
axes.yaxis.set_visible(False)
# axes.yaxis.set_ticklabels([])
axes.xaxis.set_ticks_position('bottom')
# axes.xaxis.set_ticks(np.arange(pos, pos + span, 64))
axes.xaxis.set_major_locator(MultipleLocator(64))
spines = axes.spines
for tag in ('top', 'bottom'):
spines[tag].set_visible(False)
if counter == 5:
break
plt.show() | src/articles/PDMPlayground/index.ipynb | bradhowes/keystrokecountdown | mit | 3b436f25cc750a409c777aad6d631611 |
Low-pass Filter
A fundamental nature of high-frequency sampling for PCM is that the noise from the quantization resulting from the PCM modulator is also of high-frequency (in a real system, there is also low-freq noise from clock jitter, heat, etc). When we decimate the signal, we do not want to bring the noise into the lower frequencies so we need to filter the samples before incorporating them into the new, lower-frequency signal. Our low-pass filter is a finite impulse response (FIR) type, with tap values
taken from the TFilter web application.
Our filter is designed to operate at 2 x sampleFrequency so that it will cover our original bandwidth (512 Hz) in the pass-band and heavily attenuate everything else above.
LowPassFilter.py source | import LowPassFilter
lpf = LowPassFilter.LowPassFilter() | src/articles/PDMPlayground/index.ipynb | bradhowes/keystrokecountdown | mit | 758f66c238d006a3e1534d7500ffb99d |
PDM Decimation
Our PDM signal has a sampling frequency of 64 × sampleFrequency or 65.536 kHz. To get to our original sampleFrequency we need to ultimately use one sample out of every 64 we see in the PDM pulse train.
Since we want to filter out high-frequency noise, and our filter is tuned for 2 × sampleFrequency (2.048 kHz), will take every 32nd sample and send each to our filter, but with will only use every other filtered sample.
NOTE: the reconstruction here of a sample value from PDM values is not what would really take place. In particular, the code below obtains an average of the +/- unity values in the chain, where a real implementation would count bits and convert into a sample value. | derivedSamples = []
pdmIndex = 0
while pdmIndex < pdmPulses.size:
lpf(pdmPulses[int(pdmIndex)])
pdmIndex += pdmFreq / 2
filtered = lpf(pdmPulses[int(pdmIndex)])
pdmIndex += pdmFreq / 2
derivedSamples.append(filtered)
derivedSamples = np.array(derivedSamples)
signalSamples.size, derivedSamples.size | src/articles/PDMPlayground/index.ipynb | bradhowes/keystrokecountdown | mit | a93f0c1328565453193f6e7282e2fa88 |
Now plots of the resulting signal in both time and frequency domains | plt.figure(figsize=(figWidth, 4))
plt.plot(signalTime, derivedSamples)
plt.xlabel("t")
plt.ylabel("Amplitude")
plt.suptitle('Derived Signal')
plt.show()
fftFreqs = np.arange(bandwidth)
fftValues = (np.fft.fft(derivedSamples) / sampleFrequency)[:int(bandwidth)]
plt.plot(fftFreqs, np.absolute(fftValues))
plt.xlim(0, bandwidth)
plt.ylim(0, 0.3)
plt.xlabel("Frequency")
plt.ylabel("Magnitude")
plt.suptitle("Derived Signal Frequency Components")
plt.show() | src/articles/PDMPlayground/index.ipynb | bradhowes/keystrokecountdown | mit | 0bc062ddb4ad07415655f872faa5866e |
Filtering Test
Let's redo the PCM modulation / decimation steps but this time while injecting a high-frequency (32.767 kHz) signal with 30% intensity during the modulation. Hopefully, we will not see this noise appear in the final result. | pdmFreq = 64
pdmPulses = np.empty(sampleFrequency * pdmFreq)
pdmTime = np.arange(0, pdmPulses.size)
pdmIndex = 0
signalIndex = 0
quantizationError = 0
noiseFreq = 32767 # Hz
noiseAmplitude = .30
noiseSampleDuration = 1.0 / (sampleFrequency * pdmFreq)
noiseTime = np.arange(0, 1, noiseSampleDuration)
noiseSamples = np.sin(2.0 * np.pi * noiseFreq * noiseTime) * noiseAmplitude
while pdmIndex < pdmPulses.size:
sample = signalSamples[signalIndex] + noiseSamples[pdmIndex]
signalIndex += 1
for tmp in range(pdmFreq):
if sample >= quantizationError:
bit = 1
else:
bit = -1
quantizationError = bit - sample + quantizationError
pdmPulses[pdmIndex] = bit
pdmIndex += 1
print(pdmIndex, signalIndex, pdmPulses.size, signalSamples.size, noiseSamples.size)
derivedSamples = []
pdmIndex = 0
lpf.reset()
while pdmIndex < pdmPulses.size:
lpf(pdmPulses[int(pdmIndex)])
pdmIndex += pdmFreq / 2
filtered = lpf(pdmPulses[int(pdmIndex)])
pdmIndex += pdmFreq / 2
derivedSamples.append(filtered)
derivedSamples = np.array(derivedSamples)
plt.figure(figsize=(figWidth, 4))
plt.plot(signalTime, derivedSamples)
plt.xlabel("t")
plt.ylabel("Amplitude")
plt.suptitle('Derived Signal')
plt.show()
fftFreqs = np.arange(bandwidth)
fftValues = (np.fft.fft(derivedSamples) / sampleFrequency)[:int(bandwidth)]
plt.plot(fftFreqs, np.absolute(fftValues))
plt.xlim(0, bandwidth)
plt.ylim(0, 0.3)
plt.xlabel("Frequency")
plt.ylabel("Magnitude")
plt.suptitle("Derived Signal Frequency Components")
plt.show() | src/articles/PDMPlayground/index.ipynb | bradhowes/keystrokecountdown | mit | 82881e8e9ec862017432072bcf0ad25d |
ไฝฟ็จsklearnๅฎ็ฐk-means่็ฑป
sklearn.cluster.KMeansๆไพไบไธไธช็จไบๅk-means่็ฑป็ๆฅๅฃ. | from sklearn.cluster import KMeans
import numpy as np
X = np.array([[1, 2], [1, 4], [1, 0],[4, 2], [4, 4], [4, 0]])
kmeans = KMeans(n_clusters=2, random_state=0).fit(X) | ipynbs/unsupervised/Kmeans.ipynb | NLP-Deeplearning-Club/Classic-ML-Methods-Algo | mit | 288ab7735abc659fb2b31f684ce2a9e4 |
ๆฅ็ๆจกๅ่ฎญ็ป็ปๆๅๅไธชๅ้็ๆ ็ญพ | kmeans.labels_ | ipynbs/unsupervised/Kmeans.ipynb | NLP-Deeplearning-Club/Classic-ML-Methods-Algo | mit | f1e78ab17fb5ef02dc445553aac7a5c2 |
ๆจกๅ่ฎญ็ป็ปๆๅ็จไบ้ขๆตๅ้็ๆ ็ญพ | kmeans.predict([[0, 0], [4, 4]]) | ipynbs/unsupervised/Kmeans.ipynb | NLP-Deeplearning-Club/Classic-ML-Methods-Algo | mit | 8b8b345174fe07b9933f5a2c9973b1ba |
ๆจกๅ่ฎญ็ป็ปๆๅ็ๅไธช็ฐ็ไธญๅฟ็น | kmeans.cluster_centers_ | ipynbs/unsupervised/Kmeans.ipynb | NLP-Deeplearning-Club/Classic-ML-Methods-Algo | mit | 67742029afd1a6e84e66676bfc4698c9 |
Process MEG data | data_path = sample.data_path()
raw_fname = data_path + '/MEG/sample/sample_audvis_filt-0-40_raw.fif'
raw = mne.io.read_raw_fif(raw_fname)
raw.set_eeg_reference() # set EEG average reference
events = mne.find_events(raw, stim_channel='STI 014')
event_id = dict(aud_r=1) # event trigger and conditions
tmin = -0.2 # start of each epoch (200ms before the trigger)
tmax = 0.5 # end of each epoch (500ms after the trigger)
raw.info['bads'] = ['MEG 2443', 'EEG 053']
picks = mne.pick_types(raw.info, meg=True, eeg=False, eog=True,
exclude='bads')
baseline = (None, 0) # means from the first instant to t = 0
reject = dict(grad=4000e-13, mag=4e-12, eog=150e-6)
epochs = mne.Epochs(raw, events, event_id, tmin, tmax, proj=True, picks=picks,
baseline=baseline, reject=reject) | 0.14/_downloads/plot_mne_dspm_source_localization.ipynb | mne-tools/mne-tools.github.io | bsd-3-clause | 1406797736209ff96da519814f5f39ef |
This is an alternative way of calculating the capacity by approximating the integral using the Gauss-Hermite Quadrature (https://en.wikipedia.org/wiki/Gauss%E2%80%93Hermite_quadrature). The Gauss-Hermite quadrature states that
\begin{equation}
\int_{-\infty}^\infty e^{-x^2}f(x)\mathrm{d}x \approx \sum_{i=1}^nw_if(x_i)
\end{equation}
where $w_i$ and $x_i$ are the respective weights and roots that are given by the Hermite polynomials.
We have to rearrange the integral $I = \int_{-\infty}^\infty f_Y(y)\log_2(f_Y(y))\mathrm{d}y$ a little bit to put it into a form suitable for the Gauss-Hermite quadrature
\begin{align}
I &= \frac{1}{2}\sum_{x\in{\pm 1}}\int_{-\infty}^\infty f_{Y|X}(y|X=x)\log_2(f_Y(y))\mathrm{d}y \
&= \frac{1}{2}\sum_{x\in{\pm 1}}\int_{-\infty}^\infty \frac{1}{\sqrt{2\pi}\sigma_n}e^{-\frac{(y-x)^2}{2\sigma_n^2}}\log_2(f_Y(y))\mathrm{d}y \
&\stackrel{(a)}{=} \frac{1}{2}\sum_{x\in{\pm 1}}\int_{-\infty}^\infty \frac{1}{\sqrt{\pi}}e^{-z^2}\log_2(f_Y(\sqrt{2}\sigma_n z + x))\mathrm{d}z \
&\approx \frac{1}{2\sqrt{\pi}}\sum_{x\in{\pm 1}} \sum_{i=1}^nw_i \log_2(f_Y(\sqrt{2}\sigma_n x_i + x))
\end{align}
where in $(a)$, we substitute $z = \frac{y-x}{\sqrt{2}\sigma}$ | # alternative method using Gauss-Hermite Quadrature (see https://en.wikipedia.org/wiki/Gauss%E2%80%93Hermite_quadrature)
# use 40 components to approximate the integral, should be sufficiently exact
x_GH, w_GH = np.polynomial.hermite.hermgauss(40)
print(w_GH)
def C_BIAWGN_GH(sigman):
integral_xplus1 = np.sum(w_GH * [np.log2(f_Y(np.sqrt(2)*sigman*xi + 1, sigman)) for xi in x_GH])
integral_xminus1 = np.sum(w_GH * [np.log2(f_Y(np.sqrt(2)*sigman*xi - 1, sigman)) for xi in x_GH])
integral = (integral_xplus1 + integral_xminus1)/2/np.sqrt(np.pi)
return -integral - 0.5*np.log2(2*np.pi*np.exp(1)*sigman**2)
| SC468/BIAWGN_Capacity.ipynb | kit-cel/wt | gpl-2.0 | 31b4cf1ea61ead167f0f002a2825824e |
Plot the capacity curves as a function of $E_s/N_0$ (in dB) and $E_b/N_0$ (in dB). In order to calculate $E_b/N_0$, we recall from the lecture that
\begin{equation}
\frac{E_s}{N_0} = r\cdot \frac{E_b}{N_0}\qquad\Rightarrow\qquad\frac{E_b}{N_0} = \frac{1}{r}\cdot \frac{E_s}{N_0}
\end{equation}
Next, we know that the best rate that can be achieved is the capacity, i.e., $r=C$. Hence, we get $\frac{E_b}{N_0}=\frac{1}{C}\cdot\frac{E_s}{N_0}$. Converting to decibels yields
\begin{align}
\frac{E_b}{N_0}\bigg|{\textrm{dB}} &= 10\cdot\log{10}\left(\frac{1}{C}\cdot\frac{E_s}{N_0}\right) \
&= 10\cdot\log_{10}\left(\frac{1}{C}\right) + 10\cdot\log_{10}\left(\frac{E_s}{N_0}\right) \
&= \frac{E_s}{N_0}\bigg|{\textrm{dB}} - 10\cdot\log{10}(C)
\end{align} | fig = plt.figure(1,figsize=(15,7))
plt.subplot(121)
plt.plot(esno_dB_range, capacity_AWGN)
plt.plot(esno_dB_range, capacity_BIAWGN)
plt.xlim((-10,10))
plt.ylim((0,2))
plt.xlabel('$E_s/N_0$ (dB)',fontsize=16)
plt.ylabel('Capacity (bit/channel use)',fontsize=16)
plt.grid(True)
plt.legend(['AWGN','BI-AWGN'],fontsize=14)
# plot Eb/N0 . Note that in this case, the rate that is used for calculating Eb/N0 is the capcity
# Eb/N0 = 1/r (Es/N0)
plt.subplot(122)
plt.plot(esno_dB_range - 10*np.log10(capacity_AWGN), capacity_AWGN)
plt.plot(esno_dB_range - 10*np.log10(capacity_BIAWGN), capacity_BIAWGN)
plt.xlim((-2,10))
plt.ylim((0,2))
plt.xlabel('$E_b/N_0$ (dB)',fontsize=16)
plt.ylabel('Capacity (bit/channel use)',fontsize=16)
plt.grid(True)
from scipy.stats import norm
# first compute the BSC error probability
# the Q function (1-CDF) is also often called survival function (sf)
delta_range = [norm.sf(1/sigman) for sigman in sigman_range]
capacity_BIAWGN_hard = [1+delta*np.log2(delta)+(1-delta)*np.log2(1-delta) for delta in delta_range]
fig = plt.figure(1,figsize=(15,7))
plt.subplot(121)
plt.plot(esno_dB_range, capacity_AWGN)
plt.plot(esno_dB_range, capacity_BIAWGN)
plt.plot(esno_dB_range, capacity_BIAWGN_hard)
plt.xlim((-10,10))
plt.ylim((0,2))
plt.xlabel('$E_s/N_0$ (dB)',fontsize=16)
plt.ylabel('Capacity (bit/channel use)',fontsize=16)
plt.grid(True)
plt.legend(['AWGN','BI-AWGN', 'Hard BI-AWGN'],fontsize=14)
# plot Eb/N0 . Note that in this case, the rate that is used for calculating Eb/N0 is the capcity
# Eb/N0 = 1/r (Es/N0)
plt.subplot(122)
plt.plot(esno_dB_range - 10*np.log10(capacity_AWGN), capacity_AWGN)
plt.plot(esno_dB_range - 10*np.log10(capacity_BIAWGN), capacity_BIAWGN)
plt.plot(esno_dB_range - 10*np.log10(capacity_BIAWGN_hard), capacity_BIAWGN_hard)
plt.xlim((-2,10))
plt.ylim((0,2))
plt.xlabel('$E_b/N_0$ (dB)',fontsize=16)
plt.ylabel('Capacity (bit/channel use)',fontsize=16)
plt.grid(True)
W = 4
| SC468/BIAWGN_Capacity.ipynb | kit-cel/wt | gpl-2.0 | d943c38f8c5ef14a4ef6596de80ce0e4 |
Time evolution of Spin Squuezing Parameter $\xi^2= \frac{N \langle\Delta J_y^2\rangle}{\langle J_z\rangle^2}$ | #set initial state for spins (Dicke basis)
nt = 1001
td0 = 1/(N*Lambda)
tmax = 10 * td0
t = np.linspace(0, tmax, nt)
excited = dicke(N, N/2, N/2)
load_file = False
if load_file == False:
# cycle over all states in Dicke space
xi2_1_list = []
xi2_2_list = []
xi2_1_min_list = []
xi2_2_min_list = []
for j in j_vals(N):
#for m in m_vals(j):
m = j
rho0 = dicke(N, j, m)
#solve using qutip (Dicke basis)
# Dissipative dynamics: Only collective emission
result = mesolve(liouv, rho0, t, [],
e_ops = [jz, jy, jy**2,jz**2, jx],
options = Options(store_states=True))
rhot = result.states
jz_t = result.expect[0]
jy_t = result.expect[1]
jy2_t = result.expect[2]
jz2_t = result.expect[3]
jx_t = result.expect[4]
Delta_jy = jy2_t - jy_t**2
xi2_1 = N * Delta_jy / (jz_t**2+jx_t**2)
# Dissipative dynamics: Only local emission
result2 = mesolve(liouv2, rho0, t, [],
e_ops = [jz, jy, jy**2,jz**2, jx],
options = Options(store_states=True))
rhot2 = result2.states
jz_t2 = result2.expect[0]
jy_t2 = result2.expect[1]
jy2_t2 = result2.expect[2]
jz2_t2 = result2.expect[3]
jx_t2 = result2.expect[4]
Delta_jy2 = jy2_t2 - jy_t2**2
xi2_2 = N * Delta_jy2 / (jz_t2**2+jx_t2**2)
xi2_1_min = np.min(xi2_1)
xi2_2_min = np.min(xi2_2)
xi2_1_list.append(xi2_1)
xi2_2_list.append(xi2_2)
xi2_1_min_list.append(xi2_1_min)
xi2_2_min_list.append(xi2_2_min)
print("|j, m> = ",j,m) | examples/piqs-spin-squeezing-noise.ipynb | qutip/qutip-notebooks | lgpl-3.0 | 016242e8c882e0a1167a3b663939dad8 |
Visualization | label_size2 = 20
lw = 3
texplot = False
# if texplot == True:
# plt.rc('text', usetex = True)
# plt.rc('xtick', labelsize=label_size)
# plt.rc('ytick', labelsize=label_size)
fig1 = plt.figure(figsize = (10,6))
for xi2_1 in xi2_1_list:
plt.plot(t*(N*Lambda), xi2_1, '-', label = r' $\gamma_\Downarrow=0.2$', linewidth = lw)
for xi2_2 in xi2_2_list:
plt.plot(t*(N*Lambda), xi2_2, '-.', label = r'$\gamma_\downarrow=0.2$')
plt.plot(t*(N*Lambda), 1+0*t, '--k')
plt.xlim([0,3])
plt.ylim([0,8000.5])
plt.ylim([0,2.5])
plt.xlabel(r'$ N \Lambda t$', fontsize = label_size2)
plt.ylabel(r'$\xi^2$', fontsize = label_size2)
#plt.legend(fontsize = label_size2*0.8)
plt.title(r'Spin Squeezing Parameter, $N={}$'.format(N), fontsize = label_size2)
plt.show()
plt.close()
## Here we find for how long the spin-squeezing parameter, xi2,
## is less than 1 (non-classical or "quantum" condition), in the two dynamics
dt_quantum_xi1_list = []
dt_quantum_xi2_list = []
dt1_jm =[]
dt2_jm =[]
ds = dicke_space(N)
i = 0
for j in j_vals(N):
#for m in m_vals(j):
m = j
rho0 = dicke(N, j, m)
quantum_xi1 = xi2_1_list[i][xi2_1_list[i] < 1.0]
quantum_xi2 = xi2_2_list[i][xi2_2_list[i] < 1.0]
# first ensemble
if len(quantum_xi1)>0:
dt_quantum_xi1 = len(quantum_xi1)
dt1_jm.append((dt_quantum_xi1, j, m))
else:
dt_quantum_xi1 = 0.0
# second ensemble
if len(quantum_xi2)>0:
dt_quantum_xi2 = len(quantum_xi2)
dt2_jm.append((dt_quantum_xi2, j, m))
else:
dt_quantum_xi2 = 0.0
dt_quantum_xi1_list.append(dt_quantum_xi1)
dt_quantum_xi2_list.append(dt_quantum_xi2)
i = i+1
print("collective emission: (squeezing time, j, m)")
print(dt1_jm)
print("local emission: (squeezing time, j, m)")
print(dt2_jm) | examples/piqs-spin-squeezing-noise.ipynb | qutip/qutip-notebooks | lgpl-3.0 | 4c9a249f867b9a4a9de7bd673580fc1d |
Visualization | plt.rc('text', usetex = True)
label_size = 20
label_size2 = 20
label_size3 = 20
plt.rc('xtick', labelsize=label_size)
plt.rc('ytick', labelsize=label_size)
lw = 3
i0 = -3
i0s=2
fig1 = plt.figure(figsize = (8,5))
# excited state spin squeezing
plt.plot(t*(N*Lambda), xi2_1_list[-1], 'k-',
label = r'$|\frac{N}{2},\frac{N}{2}\rangle$, $\gamma_\Downarrow=0.2\Lambda$',
linewidth = 0.8)
plt.plot(t*(N*Lambda), xi2_2_list[-1], 'r--',
label = r'$|\frac{N}{2},\frac{N}{2}\rangle$, $\gamma_\downarrow=0.2\Lambda$',
linewidth = 0.8)
# state with max time of spin squeezing
plt.plot(t*(N*Lambda), xi2_1_list[i0], 'k-',
label = r'$|j,j\rangle$, $\gamma_\Downarrow=0.2\Lambda$',
linewidth = 0.8+0.4*i0s*lw)
plt.plot(t*(N*Lambda), xi2_2_list[i0], 'r--',
label = r'$|j,j\rangle$, $\gamma_\downarrow=0.2\Lambda$',
linewidth = 0.8+0.4*i0s*lw)
plt.plot(t*(N*Lambda), 1+0*t, '--k')
plt.xlim([0,2.5])
plt.yticks([0,1,2])
plt.ylim([-1,2.])
plt.xlabel(r'$ N \Lambda t$', fontsize = label_size3)
plt.ylabel(r'$\xi^2$', fontsize = label_size3)
plt.legend(fontsize = label_size2*0.8, ncol=2)
fname = 'figures/spin_squeezing_N_{}_states.pdf'.format(N)
plt.title(r'Spin Squeezing Parameter, $N={}$'.format(N), fontsize = label_size2)
plt.show()
plt.close() | examples/piqs-spin-squeezing-noise.ipynb | qutip/qutip-notebooks | lgpl-3.0 | be3d17a2952e9b6c5055e00dcc9c6c7b |
The plot shows the spin squeezing parameter for two different dynamics -- only collective de-excitation, black curves; only local de-excitation, red curves -- and for two different inital states, the maximally excited state (thin curves) and another Dicke state with longer squeezing time (thick curves). This study, performed in Refs. [5,6] for the maximally excited state has been extended to any Dicke state in Ref. [7]. | # plot the dt matrix in the Dicke space
plt.rc('text', usetex = True)
label_size = 20
label_size2 = 20
label_size3 = 20
plt.rc('xtick', labelsize=label_size)
plt.rc('ytick', labelsize=label_size)
lw = 3
i0 = 7
i0s=2
ratio_squeezing_local = 3
fig1 = plt.figure(figsize = (6,8))
ds = dicke_space(N)
value_excited = 3
ds[0,0]=value_excited
ds[int(N/2-i0),int(N/2-i0)]=value_excited * ratio_squeezing_local
plt.imshow(ds, cmap="inferno_r")
plt.xticks([])
plt.yticks([])
plt.xlabel(r"$j$", fontsize = label_size3)
plt.ylabel(r"$m$", fontsize = label_size3)
plt.title(r"Dicke space $(j,m)$ for $N={}$".format(N), fontsize = label_size3)
plt.show()
plt.close() | examples/piqs-spin-squeezing-noise.ipynb | qutip/qutip-notebooks | lgpl-3.0 | d40bf4c8a6a0475751b2e1e2a63736d1 |
The Plot above shows the two initial states (darker dots) $|\frac{N}{2},\frac{N}{2}\rangle$ (top edge of the Dicke triangle, red dot) and $|j,j\rangle$, with $j=\frac{N}{2}-3=7$ (black dot). A study of the Dicke triangle (dark yellow space) and state engineering is performed in Ref. [8] for different initial state.
References
[1] D. J. Wineland, J. J. Bollinger, W. M. Itano, F. L. Moore, and D. J. Heinzen, Spin squeezing and reduced quantum noise in spectroscopy, Phys. Rev. A 46, R6797 (1992)
[2] M. Kitagawa and M. Ueda, Squeezed spin states, Phys. Rev. A 47, 5138 (1993)
[3] J. Ma, X. Wang, C.-P. Sun, and F. Nori, Quantum spin squeezing, Physics Reports 509, 89 (2011)
[4] L. Pezzeฬ, A. Smerzi, M. K. Oberthaler, R. Schmied, and P. Treutlein, Quantum metrology with nonclassical states of atomic ensembles, Reviews of Modern Physics, in press (2018)
[5] B. A. Chase and J. Geremia, Collective processes of an ensemble of spin-1 particles, Phys. Rev. A 78,0521012 (2008)
[6] B. Q. Baragiola, B. A. Chase, and J. Geremia, Collective uncertainty in partially polarized and partially deco- hered spin-1 systems, Phys. Rev. A 81, 032104 (2010)
[7] N. Shammah, S. Ahmed, N. Lambert, S. De Liberato, and F. Nori,
Open quantum systems with local and collective incoherent processes: Efficient numerical simulation using permutational invariance https://arxiv.org/abs/1805.05129
[8] N. Shammah, N. Lambert, F. Nori, and S. De Liberato, Superradiance with local phase-breaking effects, Phys. Rev. A 96, 023863 (2017). | qutip.about() | examples/piqs-spin-squeezing-noise.ipynb | qutip/qutip-notebooks | lgpl-3.0 | 7a5cd4876d945e290d8c8b5a5363f40c |
2) Create classes/bins
Instead of having a range of values you can discretize in classes/bins. Make use of pandas' qcut: Discretize variable into equal-sized buckets. | data['height'].hist(bins=100)
plt.title('Height population distribution')
plt.xlabel('cm')
plt.ylabel('freq') | course/class2/01-clean/examples/00-kill.ipynb | hershaw/data-science-101 | mit | 87c1f88aa74f5f0af89f49eee38a38a5 |
Step 1: Fit the Initial Random Forest
Just fit every feature with equal weights per the usual random forest code e.g. DecisionForestClassifier in scikit-learn | # Load the iris data
iris = load_iris()
# Create the train-test datasets
X_train, X_test, y_train, y_test = train_test_split(iris.data, iris.target)
np.random.seed(1039)
# Just fit a simple random forest classifier with 2 decision trees
rf = RandomForestClassifier(n_estimators = 2)
rf.fit(X = X_train, y = y_train)
# Now plot the trees individually
for idx, dtree in enumerate(rf.estimators_):
print(idx)
utils.draw_tree(inp_tree = dtree)
#utils.draw_tree(inp_tree = rf.estimators_[1]) | jupyter/backup_deprecated_nbs/06_explore_binary_decision_tree.ipynb | Yu-Group/scikit-learn-sandbox | mit | 3c358e9f57094279fd25e68bc6a74d3e |
Get the second Decision tree to use for testing | estimator = rf.estimators_[1]
from sklearn.tree import _tree
estimator.tree_.node_count
estimator.tree_.children_left[0]
estimator.tree_.children_right[0]
_tree.TREE_LEAF | jupyter/backup_deprecated_nbs/06_explore_binary_decision_tree.ipynb | Yu-Group/scikit-learn-sandbox | mit | 0681696cb092ed1874668e7a8221e7d1 |
Write down an efficient Binary Tree Traversal Function | # Now plot the trees individually
utils.draw_tree(inp_tree = estimator)
def binaryTreePaths(dtree, root_node_id = 0):
# Use these lists to parse the tree structure
children_left = dtree.tree_.children_left
children_right = dtree.tree_.children_right
if root_node_id is None:
paths = []
if root_node_id == _tree.TREE_LEAF:
raise ValueError("Invalid node_id %s" % _tree.TREE_LEAF)
# if left/right is None we'll get empty list anyway
if children_left[root_node_id] != _tree.TREE_LEAF:
paths = [str(root_node_id) + '->' + str(l)
for l in binaryTreePaths(dtree, children_left[root_node_id]) +
binaryTreePaths(dtree, children_right[root_node_id])]
else:
paths = [root_node_id]
return paths
x1 = binaryTreePaths(rf.estimators_[1], root_node_id = 0)
x1
def binaryTreePaths2(dtree, root_node_id = 0):
# Use these lists to parse the tree structure
children_left = dtree.tree_.children_left
children_right = dtree.tree_.children_right
if root_node_id is None:
paths = []
if root_node_id == _tree.TREE_LEAF:
raise ValueError("Invalid node_id %s" % _tree.TREE_LEAF)
# if left/right is None we'll get empty list anyway
if children_left[root_node_id] != _tree.TREE_LEAF:
paths = [np.append(root_node_id, l)
for l in binaryTreePaths2(dtree, children_left[root_node_id]) +
binaryTreePaths2(dtree, children_right[root_node_id])]
else:
paths = [root_node_id]
return paths
x = binaryTreePaths2(rf.estimators_[1], root_node_id = 0)
x
leaf_nodes = [y[-1] for y in x]
leaf_nodes
n_node_samples = estimator.tree_.n_node_samples
num_samples = [n_node_samples[y].astype(int) for y in leaf_nodes]
print(n_node_samples)
print(len(n_node_samples))
num_samples
print(num_samples)
print(sum(num_samples))
print(sum(n_node_samples))
X_train.shape
value = estimator.tree_.value
values = [value[y].astype(int) for y in leaf_nodes]
print(values)
# This should match the number of rows in the training feature set
print(sum(values).sum())
values
feature_names = ["X" + str(i) for i in range(X_train.shape[1])]
np.asarray(feature_names)
print(type(feature_names))
print(feature_names[0])
print(feature_names[-2])
feature = estimator.tree_.feature
z = [feature[y].astype(int) for y in x]
z
#[feature_names[i] for i in z]
max_dpth = estimator.tree_.max_depth
max_dpth
max_n_class = estimator.tree_.max_n_classes
max_n_class
print("nodes", np.asarray(a = nodes, dtype = "int64"), sep = ":\n")
print("node_depth", node_depth, sep = ":\n")
print("leaf_node", is_leaves, sep = ":\n")
print("feature_names", used_feature_names, sep = ":\n")
print("feature", feature, sep = ":\n") | jupyter/backup_deprecated_nbs/06_explore_binary_decision_tree.ipynb | Yu-Group/scikit-learn-sandbox | mit | 787eb8afd3af7108b3e0be16642b3f05 |
Options | ## Retrieve the bounding box of the specified county - if no county is specified, the bounding boxes for all NM counties will be requested
countyBBOXlink = "http://gstore.unm.edu/apps/epscor/search/nm_counties.json?limit=100&query=" + county_name ## define the request URL
print countyBBOXlink ## print the request URL for verification
print
bboxFile = urllib.urlopen(countyBBOXlink) ## request the bounding box information from the server
bboxData = json.load(bboxFile)
# print bboxData
# Get data for BBOX defined by specified county(ies)
myCounties = []
for countyBBOX in bboxData["results"]:
minx,miny,maxx,maxy = countyBBOX[u'box']
myDownloadLink = "http://waterservices.usgs.gov/nwis/iv/?bBox=%f,%f,%f,%f&format=json&period=P7D¶meterCd=00060" % (minx,miny,maxx,maxy) # retrieve data for the specified BBOX for the last 7 days as JSON
print myDownloadLink
myCounty = {u'name':countyBBOX[u'text'],u'minx':minx,u'miny':miny,u'maxx':maxx,u'maxy':maxy,u'downloadLink':myDownloadLink}
myCounties.append(myCounty)
#countySubset = [myCounties[0]]
#print countySubset
valueList = []
for county in myCounties:
print "processing: %s" % county["downloadLink"]
try:
datafile = urllib.urlopen(county["downloadLink"])
data = json.load(datafile)
values = data["value"]["timeSeries"][0]["values"]
for item in values:
for valueItem in item["value"]:
#print json.dumps(item["value"], sort_keys=True, indent=4)
myValue = {"dateTime":valueItem["dateTime"].replace("T"," ").replace(".000-06:00",""),"value":valueItem["value"], "county":county["name"]}
#print myValue
valueList.append(myValue)
#print valueList
except:
print "\tfailed for this one ..."
#print json.dumps(values, sort_keys=True, indent=4)
df = pandas.DataFrame(valueList)
df['dateTime'] = pandas.to_datetime(df["dateTime"])
df['value'] = df['value'].astype(float).fillna(-1)
print df.shape
print df.dtypes
print "column names"
print "------------"
for colName in df.columns:
print colName
print
print df.head()
%matplotlib inline
fig,ax = plt.subplots(figsize=(10,8))
ax.width = 1
ax.height = .5
plt.xkcd()
#plt.ylim(-25,30)
ax.plot_date(df['dateTime'], df['value'], '.', label="Discharge (cf/sec)", color="0.2")
fig.autofmt_xdate()
plt.legend(loc=2, bbox_to_anchor=(1.0,1))
plt.title("15-minute Discharge - cubic feet per second")
plt.ylabel("Discharge")
plt.xlabel("Date")
plt.show() | presentations/2014-04-CI-day/examples/notebook_02-Copy1.ipynb | karlbenedict/karlbenedict.github.io | mit | bacd15b85a32cf9f9f1473070e8faa54 |
<p style="text-align: right; direction: rtl; float: right; clear: both;">
ืืชืืฆืื ืืื ืจืฉืืื ืฉื ืื ืืคืขืืืืช ืฉืืคืฉืจ ืืืคืขืื ืขื <i>str</i>.<br>
ืืฉืื ืืื, ืืืืืฅ ืืื ืืืชืขืื ืืคืขืืืืช ืืจืฉืืื ืฉืฉืื ืืชืืื ืืงื ืชืืชืื.
</p>
<p style="text-align: right; direction: rtl; float: right; clear: both;">
ืืจืืง ื ืืกืฃ ืฉืื ืจืื ื ืื ืืืชืจ, ืืืื ืืกืืืืืช ืคืืชืื ืจืืืช ืฉืืฆื ืืื ืืขืืื ืืื.<br>
ืืืจืืง ืืื ืฆืืื ืกืื ืื ืชืื ืื ืืืฉืชื ื ืฉืืชื ืขืืืืื ืขืืื, ืืกืืื "ื ืงืืื" ืืื ืืืืฆื ืขื <kbd dir="ltr" style="direction: ltr">โน TAB</kbd>.
</p> | # ืืงืื ืืช ืืกืื ืืืจื ืื ืงืืื, ืืื ืืืฆื ืขื ืืืงืฉ "ืืื" ืืืงืืืช
str.
# ื ืืชื ืื ืื:
"Hello".
# ืื ืื:
s = "Hello"
s. | week02/6_Documentation.ipynb | PythonFreeCourse/Notebooks | mit | 58068d3bbd9ad95953a15303a584ff01 |
<span style="text-align: right; direction: rtl; float: right; clear: both;">ืชืืขืื ืขื ืคืขืืื ืื ืขื ืคืื ืงืฆืื</span>
<p style="text-align: right; direction: rtl; float: right; clear: both;">
ืืืงืจื ืฉื ืจืฆื ืืืคืฉ ืคืจืืื ื ืืกืคืื ืขื ืืืช ืืคืื ืงืฆืืืช ืื ืืคืขืืืืช (ื ื ืื <code>len</code>, ืื <code dir="ltr" style="direction: ltr">str.upper()</code>), ืืชืืขืื ืฉื ืคืืืชืื ืืื ืืงืืจ ืืืืข ื ืืืจ ืืื.<br>
ืื ืื ืื ื ื ืืฆืืื ืืชืื ืืืืืจืช, ืืฉ ืืจืืง ื ืืื ืืงืื ืืืง ืืืชืืขืื ืืื ืืฆืืจื ืืืืจื โ ืคืฉืื ื ืจืฉืื ืืชื ืงืื ืืช ืฉื ืืคืื ืงืฆืื, ืืืืจืื ืกืืื ืฉืืื:
</p> | len? | week02/6_Documentation.ipynb | PythonFreeCourse/Notebooks | mit | 5004ef48614754e80203dc87d1629c1b |
<p style="text-align: right; direction: rtl; float: right; clear: both;">
ืืจืืข ืฉื ืจืืฅ ืืช ืืชื, ืชืงืคืืฅ ืื ื ืืืื ืืช ืขื ืืืืข ื ืืกืฃ ืขื ืืคืื ืงืฆืื.<br>
ืื ืื ืื ื ืจืืฆืื ืืงืื ืืืืข ืขื ืคืขืืื, ื ืืชืื ืืช ืกืื ืืขืจื ืฉืขืืื ืื ืื ื ืจืืฆืื ืืืฆืข ืืืชื (ื ื ืื, str):
</p> | # str - ืืฉื ืฉื ืืืคืืก ืื ืชืื ืื (ืืกืื ืฉื ืืขืจื)
# . - ืื ืงืืื ืืื ืกืืืื ืฉืืคืขืืื ืฉืืชืื ื ืืืจืื ืฉืืืืช ืืกืื ืฉืืชืื ื ืืคื ืื
# upper - ืืฉื ืฉื ืืคืขืืื ืฉืขืืื ืจืืฆืื ืืงืื ืขืืจื
# ? - ืืืงืฉ ืืช ืืืืืข ืขื ืืคืขืืื
str.upper? | week02/6_Documentation.ipynb | PythonFreeCourse/Notebooks | mit | 093e689814e1fd1e82d3be8c679e7757 |
<div class="align-center" style="display: flex; text-align: right; direction: rtl;">
<div style="display: flex; width: 10%; float: right; ">
<img src="images/warning.png" style="height: 50px !important;" alt="ืืืืจื!">
</div>
<div style="width: 90%">
<p style="text-align: right; direction: rtl;">
ืงืจืืื ืืคืื ืงืฆืื, ืงืจื ืืืกืคืช ืืชืืืื <code>()</code> ืืคื ื ืกืืื ืืฉืืื, ืชืคืขืื ืืช ืืคืื ืงืฆืื ืื ืืคืขืืื ืืืงืื ืืชืช ืืื ืขืืจื.
</p>
</div>
</div>
<p style="text-align: right; direction: rtl; float: right; clear: both;">
ืืชืื ืืืื ืืช ืืขืืจื ืฉืชืืคืชื ืืืืืจืช ื ืจืื ืฉืืจืืช ืืืืืืืช ืคืจืืื ืืขื ืืื ืื:
</p>
<ul style="text-align: right; direction: rtl; float: right; clear: both;">
<li><dfn>Signature</dfn> โ ืืชืืืช ืืคืขืืื ืื ืืคืื ืงืฆืื, ืืืืืืช ืืช ืืฉื ืฉืื ืืืช ืืคืจืืืจืื ืฉืื.</li>
<li><dfn>Docstring</dfn> โ ืืื ืืืืื ืฉืืชืืจืืช ืืืื ืื ืืคืื ืงืฆืื ืขืืฉื, ืืืขืืชืื ื ืืชื ืืช ืืืืข ื ืืกืฃ ืขื ืืคืจืืืจืื.</li>
</ul>
<p style="text-align: right; direction: rtl; float: right; clear: both;">
ืืขืช ืขืชื, ื ืชืขืื ืืืจืืืืื <em>self</em>, <em>*</em> ืื <em>/</em> ืฉืืืคืืขื ืืื ืคืขื ืืฉืื Signature.
</p>
<p style="text-align: right; direction: rtl; float: right; clear: both;">ืืฉืืื ืขืืจื ื ืืกืคืื</p>
<p style="text-align: right; direction: rtl; float: right; clear: both;">
ืขืืื ืืชืื ืืช ืืื ืืืืจ ืืืืืื, ืืงืืืืื ืืฉืืืื ื ืืืจืื ืฉืืืจืชื ืืขืืืจ ืืืชืื ืช.<br>
ืืคื ืืื ืืื ืืืคืืคืืืจืืื ืฉืืื:
</p>
<ul style="text-align: right; direction: rtl; float: right; clear: both;">
<li><a href="https://google.com">Google</a> โ ืืคืฉื ืืืื ืืช ืืฉืืื ืฉืืื ืึพGoogle. ืืชืื ืช ืืื ืขืืฉื ืืช ืื ืคืขืืื ืจืืืช ืืืื. ืงืจืื ืืืืืื ืฉืืืฉืื ืืขืืื ืืืจ ื ืชืงื ืืขืืจ ืืืขืื ืฉืืื.</li>
<li><a href="https://docs.python.org/3">ืืชืืขืื ืฉื ืคืืืชืื</a> โ ืืืื ืืจืื ืืืืข, ืืืขืืชืื ืืืืืืืช ืืืขืืืืช.</li>
<li><a href="https://stackoverflow.com">Stack Overflow</a> โ ืืื ืืืชืจืื ืืื ืืืืจืื ืืขืืื ืืคืืชืื, ืืืืื ืืขืจืืช ืฉืืืืช ืืชืฉืืืืช ืขื ืืืจืื ืื ืืืข ืืื ืื ืฉืงืฉืืจ ืืชืื ืืช.</li>
<li><a href="https://github.com">GitHub</a> โ ืืชืจ ืฉืื ืื ืฉืื ืื ืืืื ืืช ืืงืื ืฉืืื ืืืฉืชืคืื ืืืชื ืขื ืืืจืื. ืืฉ ืื ืฉืืจืช ืืืคืืฉ, ืืืื ืืขืืื ืืฆืืจื ืืฆืืืช ืืืืืืืช ืืฉืืืืฉ ืืงืื.</li>
</ul>
<p style="text-align: right; direction: rtl; float: right; clear: both;">ืชืจืืื</p>
<p style="text-align: right; direction: rtl; float: right; clear: both;">
ืืชืจืืื ืื ืืฉืชืืฉื ืืืืืช ืืฆืืจื ืืชืืขืื ืฉื ืคืืืชืื ืืื ืืืืืช ืคืขืืืืช ืฉืื ืืืื ื ืขืืืื.
</p>
<p style="text-align: right; direction: rtl; float: right; clear: both;">ืกืืืืจ ืจืฉืืื</p>
<p style="text-align: right; direction: rtl; float: right; clear: both;">
ืืคื ืืื ืจืฉืืืช ืืืกืคืจืื ืืืืขืืื ืึพ1 ืขื 10 ืืกืืจ ืืืืืื.<br>
ืืื ืชืืืื ืืกืืจ ืืืชื ืืฉืืจื ืืืช ืฉื ืงืื, ืืืืืคืืก ืืืชื ืืฉืืจื ืืืช ื ืืกืคืช?<br>
ืืคืื ืฉืืืืคืก ืขื ืืืกื ืฆืจืื ืืืืืช: <samp dir="ltr" style="direction: ltr;">[1, 2, 3, 4, 5, 6, 7, 8, 9, 10]</samp>.
</p> | numbers = [2, 9, 10, 8, 7, 4, 3, 5, 6, 1] | week02/6_Documentation.ipynb | PythonFreeCourse/Notebooks | mit | 859b050acda475572faeefa72b1c55e8 |
In this example, it is True that our variable m is larger than zero, and therefore, the print call ('if code' in the above figure) is executed. Now, what if the condition were not True? Well... | n = -5
if n > 0:
print("Larger than zero.") | docs/mpg-if_error_continue/examples/e-02-2_conditionals.ipynb | marburg-open-courseware/gmoc | mit | e471f34d53921be13326235a9315e69e |