markdown
stringlengths 0
1.02M
| code
stringlengths 0
832k
| output
stringlengths 0
1.02M
| license
stringlengths 3
36
| path
stringlengths 6
265
| repo_name
stringlengths 6
127
|
|---|---|---|---|---|---|
Mapping QTL in BXD mice using R/qtl2[Karl Broman](https://kbroman.org)[](https://orcid.org/0000-0002-4914-6671),[Department of Biostatistics & Medical Informatics](https://www.biostat.wisc.edu), [University of Wisconsin–Madison](https://www.wisc.edu)Our aim in this tutorial is to demonstrate how to map quantitative trait loci (QTL) in the BXD mouse recombinant inbred lines using the [R/qtl2](https://kbroman.org/qtl2) software. We will first show how to download BXD phenotypes from [GeneNetwork2](http://gn2.genenetwork.org) using its API, via the R package [R/GNapi](https://github.com/rqtl/GNapi). At the end, we will use the [R/qtl2browse](https://github.com/rqtl/qtl2browse) package to display genome scan results using the [Genetics Genome Browser](https://github.com/chfi/purescript-genome-browser). Acquiring phenotypes with the GeneNetwork APIWe will first use the [GeneNetwork2](http://gn2.genenetwork.org) API to acquire BXD phenotypes to use for mapping. We will use the R package [R/GNapi](https://github.com/rqtl/GNapi). We first need to install the package, which is not available on [CRAN](https://cran.r-project.org), but is available via a private repository.```rinstall.packages("GNapi", repos="http://rqtl.org/qtl2cran")```We then load the package using `library()`. | install.packages("GNapi", repos="http://rqtl.org/qtl2cran")
library(GNapi) | _____no_output_____ | CC0-1.0 | CTC2019_tutorial.ipynb | genenetwork/Teaching_CTC2019 |
The [R/GNapi](https://github.com/kbroman/GNapi) has a variety of functions. For an overview, see [its vignette](http://kbroman.org/GNapi/GNapi.html). Here we will just do one thing: use the function `get_pheno()` to grab BXD phenotype data. You provide a data set and a phenotype. Phenotype 10038 concerns "habituation", measured as a difference in locomotor activity between day 1 and day 3 in a 5 minute test trial. | phe <- get_pheno("BXD", "10038")
head(phe) | _____no_output_____ | CC0-1.0 | CTC2019_tutorial.ipynb | genenetwork/Teaching_CTC2019 |
We will use just the column "value", but we need to include the strain names so that R/qtl2 can line up these phenotypes with the genotypes. | pheno <- setNames(phe$value, phe$sample_name)
head(pheno) | _____no_output_____ | CC0-1.0 | CTC2019_tutorial.ipynb | genenetwork/Teaching_CTC2019 |
Acquire genotype data with R/qtl2We now want to get genotype data for the BXD panel. We first need to install the [R/qtl2](https://kbroman.org/qtl2) package. As with R/GNapi, it is not available on CRAN, but rather is distributed via a private repository.```rinstall.packages("qtl2", repos="http://rqtl.org/qtl2cran")```We then load the package with `library()`. | install.packages("qtl2", repos="http://rqtl.org/qtl2cran")
library(qtl2) | _____no_output_____ | CC0-1.0 | CTC2019_tutorial.ipynb | genenetwork/Teaching_CTC2019 |
R/qtl2 uses a special file format for QTL data ([described here](https://kbroman.org/qtl2/assets/vignettes/input_files.html)). There are a variety of sample datasets [on Github](https://github.com/rqtl/qtl2data), including genotypes for the [mouse BXD lines](https://github.com/rqtl/qtl2data/tree/master/BXD), taken from [GeneNetwork2](http://gn2.genenetwork.org). We'll load those data directly into R using the function `read_cross2()`. | bxd_file <- "https://raw.githubusercontent.com/rqtl/qtl2data/master/BXD/bxd.zip"
bxd <- read_cross2(bxd_file) | Warning message in recode_geno(sheet, genotypes):
โ117497 genotypes treated as missing: "H"โ | CC0-1.0 | CTC2019_tutorial.ipynb | genenetwork/Teaching_CTC2019 |
We get a warning message about heterozygous genotypes being omitted. A number of the newer BXD lines have considerable heterozygosity. But these lines weren't among those phenotyped in the data we downloaded above, and so we don't need to worry about it here.The data are read into the object `bxd`, which has class `"cross2"`. It contains the genotypes and well as genetic and physical marker maps. There are also phenotype data (which we will ignore).We can get a quick summary of the dataset with `summary()`. For reasons that I don't understand, it gets printed as a big mess within this Jupyter notebook, and so here we need to surround it with `print()` to get the intended output. | print( summary(bxd) ) | Object of class cross2 (crosstype "risib")
Total individuals 198
No. genotyped individuals 198
No. phenotyped individuals 198
No. with both geno & pheno 198
No. phenotypes 5806
No. covariates 0
No. phenotype covariates 1
No. chromosomes 20
Total markers 7320
No. markers by chr:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 X
636 583 431 460 470 449 437 319 447 317 375 308 244 281 247 272 291 250 310 193
| CC0-1.0 | CTC2019_tutorial.ipynb | genenetwork/Teaching_CTC2019 |
The first step in QTL analysis is to calculate genotype probabilities at putative QTL positions across the genome, conditional on the observed marker data. This allows us that consider positions between the genotyped markers and to allow for the presence of genotyping errors.First, we need to define the positions that we will consider. We will take the observed marker positions and insert a set of "pseudomarkers" (marker-like positions that are not actually markers). We do this with the function `insert_pseudomarkers()`. We pull the genetic map (`gmap`) out of the `bxd` data as our basic map; `step=0.2` and `stepwidth="max"` mean to insert pseudomarkers so that no two adjacent markers or pseudomarkers are more than 0.2 cM apart. That is, in any marker interval that is greater than 0.2 cM, we will insert one or more evenly spaced pseudomarkers, so that the intervals between markers and pseudomarkers are no more than 0.2 cM. | gmap <- insert_pseudomarkers(bxd$gmap, step=0.2, stepwidth="max") | _____no_output_____ | CC0-1.0 | CTC2019_tutorial.ipynb | genenetwork/Teaching_CTC2019 |
We will be interested in results with respect to the physical map (in Mbp), and so we need to create a corresponding map that includes the pseudomarker positions. We do this with the function `interp_map()`, which uses linear interpolation to get estimated positions for the inserted pseudomarkers. | pmap <- interp_map(gmap, bxd$gmap, bxd$pmap) | _____no_output_____ | CC0-1.0 | CTC2019_tutorial.ipynb | genenetwork/Teaching_CTC2019 |
We can now proceed with calculating genotype probabilities for all BXD strains at all markers and pseudomarkers, conditional on the observed marker genotypes and assuming a 0.2% genotyping error rate. We use the [Carter-Falconer](https://doi.org/10.1007/BF02996226) map function to convert between cM and recombination fractions; it assumes a high degree of crossover interference, appropriate for the mouse. | pr <- calc_genoprob(bxd, gmap, error_prob=0.002, map_function="c-f") | _____no_output_____ | CC0-1.0 | CTC2019_tutorial.ipynb | genenetwork/Teaching_CTC2019 |
In the QTL analysis, we will fit a linear mixed model to account for polygenic background effects. We will use the "leave one chromosome out" (LOCO) method for this. When we scan a chromosome for a QTL, we include a polygenic term with a kinship matrix derived from all other chromosomes. We first need to calculate this set of kinship matrices, which we do with the function `calc_kinship()`. The second argument, `"loco"`, indicates that we want to calculate a vector of kinship matrices, each derived from the genotype probabilities but leaving one chromosome out. | k <- calc_kinship(pr, "loco") | _____no_output_____ | CC0-1.0 | CTC2019_tutorial.ipynb | genenetwork/Teaching_CTC2019 |
Now, finally, we're ready to perform the genome scan, which we do with the function `scan1()`. It takes the genotype probabilities and a set of phenotypes (here, just one phenotype). If kinship matrices are provided (here, as `k`), the scan is performed using a linear mixed model. To make the calculations faster, the residual polygenic variance is first estimated without including any QTL effect and is then taking to be fixed and known during the scan. | out <- scan1(pr, pheno, k) | _____no_output_____ | CC0-1.0 | CTC2019_tutorial.ipynb | genenetwork/Teaching_CTC2019 |
The output of `scan1()` is a matrix of LOD scores; the rows are marker/pseudomarker positions and the columns are phenotypes. We can plot the results using `plot.scan1()`, and we can just use `plot()` because it uses the class of its input to determine what plot to make.Here I'm using the package [repr](https://cran.r-project.org/package=repr) to control the height and width of the plot that's created. I installed it with `install.packages("repr")`. You can ignore that part, if you want. | library(repr)
options(repr.plot.height=4, repr.plot.width=8)
par(mar=c(5.1, 4.1, 0.6, 0.6))
plot(out, pmap) | _____no_output_____ | CC0-1.0 | CTC2019_tutorial.ipynb | genenetwork/Teaching_CTC2019 |
There's a clear QTL on chromosome 8. We can make a plot of just that chromosome with the argument `chr=15`. | plot(out, pmap, chr=15) | _____no_output_____ | CC0-1.0 | CTC2019_tutorial.ipynb | genenetwork/Teaching_CTC2019 |
Let's create a plot of the phenotype vs the genotype at the inferred QTL. We first need to identify the QTL location, which we can do using `max()`. We then use `maxmarg()` to get inferred genotypes at the inferred QTL. | mx <- max(out, pmap)
g_imp <- maxmarg(pr, pmap, chr=mx$chr, pos=mx$pos, return_char=TRUE) | _____no_output_____ | CC0-1.0 | CTC2019_tutorial.ipynb | genenetwork/Teaching_CTC2019 |
We can use `plot_pxg()` to plot the phenotype as a function of QTL genotype. We use `swap_axes=TRUE` to have the phenotype on the x-axis and the genotype on the y-axis, rather than the other way around. Here we see that the BB and DD genotypes are completely separated, phenotypically. | par(mar=c(5.1, 4.1, 0.6, 0.6))
plot_pxg(g_imp, pheno, swap_axes=TRUE, xlab="Habituation phenotype") | _____no_output_____ | CC0-1.0 | CTC2019_tutorial.ipynb | genenetwork/Teaching_CTC2019 |
Browse genome scan results with Genetics Genome BrowserThe [Genetics Genome Browser](https://github.com/chfi/purescript-genome-browser) is a fast, lightweight, [purescript]-based genome browser developed for browsing GWAS or QTL analysis results. We'll use the R package [R/qtl2browse](https://github.com/rqtl/qtl2browse) to view our QTL mapping results in the GGB.We first need to install the R/qtl2browse package, again from a private [CRAN](https://cran.r-project.org)-like repository.```rinstall.packages("qtl2browse", repos="http://rqtl.org/qtl2cran")```We then load the package and use its one function, `browse()`, which takes the `scan1()` output and corresponding physical map (in Mbp). This will open the Genetics Genome Browser in a separate tab in your web browser. | library(qtl2browse)
browse(out, pmap) | _____no_output_____ | CC0-1.0 | CTC2019_tutorial.ipynb | genenetwork/Teaching_CTC2019 |
Structured and time series data This notebook contains an implementation of the third place result in the Rossman Kaggle competition as detailed in Guo/Berkhahn's [Entity Embeddings of Categorical Variables](https://arxiv.org/abs/1604.06737).The motivation behind exploring this architecture is it's relevance to real-world application. Most data used for decision making day-to-day in industry is structured and/or time-series data. Here we explore the end-to-end process of using neural networks with practical structured data problems. | %matplotlib inline
%reload_ext autoreload
%autoreload 2
from fastai.structured import *
from fastai.column_data import *
np.set_printoptions(threshold=50, edgeitems=20)
PATH='data/rossmann/' | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
Create datasets In addition to the provided data, we will be using external datasets put together by participants in the Kaggle competition. You can download all of them [here](http://files.fast.ai/part2/lesson14/rossmann.tgz).For completeness, the implementation used to put them together is included below. | def concat_csvs(dirname):
path = f'{PATH}{dirname}'
filenames=glob(f"{PATH}/*.csv")
wrote_header = False
with open(f"{path}.csv","w") as outputfile:
for filename in filenames:
name = filename.split(".")[0]
with open(filename) as f:
line = f.readline()
if not wrote_header:
wrote_header = True
outputfile.write("file,"+line)
for line in f:
outputfile.write(name + "," + line)
outputfile.write("\n")
# concat_csvs('googletrend')
# concat_csvs('weather') | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
Feature Space:* train: Training set provided by competition* store: List of stores* store_states: mapping of store to the German state they are in* List of German state names* googletrend: trend of certain google keywords over time, found by users to correlate well w/ given data* weather: weather* test: testing set | table_names = ['train', 'store', 'store_states', 'state_names',
'googletrend', 'weather', 'test'] | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
We'll be using the popular data manipulation framework `pandas`. Among other things, pandas allows you to manipulate tables/data frames in python as one would in a database.We're going to go ahead and load all of our csv's as dataframes into the list `tables`. | tables = [pd.read_csv(f'{PATH}{fname}.csv', low_memory=False) for fname in table_names]
from IPython.display import HTML, display | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
We can use `head()` to get a quick look at the contents of each table:* train: Contains store information on a daily basis, tracks things like sales, customers, whether that day was a holdiay, etc.* store: general info about the store including competition, etc.* store_states: maps store to state it is in* state_names: Maps state abbreviations to names* googletrend: trend data for particular week/state* weather: weather conditions for each state* test: Same as training table, w/o sales and customers | for t in tables: display(t.head()) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
This is very representative of a typical industry dataset.The following returns summarized aggregate information to each table accross each field. | for t in tables: display(DataFrameSummary(t).summary()) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
Data Cleaning / Feature Engineering As a structured data problem, we necessarily have to go through all the cleaning and feature engineering, even though we're using a neural network. | train, store, store_states, state_names, googletrend, weather, test = tables
len(train),len(test) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
We turn state Holidays to booleans, to make them more convenient for modeling. We can do calculations on pandas fields using notation very similar (often identical) to numpy. | train.StateHoliday = train.StateHoliday!='0'
test.StateHoliday = test.StateHoliday!='0' | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
`join_df` is a function for joining tables on specific fields. By default, we'll be doing a left outer join of `right` on the `left` argument using the given fields for each table.Pandas does joins using the `merge` method. The `suffixes` argument describes the naming convention for duplicate fields. We've elected to leave the duplicate field names on the left untouched, and append a "\_y" to those on the right. | def join_df(left, right, left_on, right_on=None, suffix='_y'):
if right_on is None: right_on = left_on
return left.merge(right, how='left', left_on=left_on, right_on=right_on,
suffixes=("", suffix)) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
Join weather/state names. | weather = join_df(weather, state_names, "file", "StateName") | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
In pandas you can add new columns to a dataframe by simply defining it. We'll do this for googletrends by extracting dates and state names from the given data and adding those columns.We're also going to replace all instances of state name 'NI' to match the usage in the rest of the data: 'HB,NI'. This is a good opportunity to highlight pandas indexing. We can use `.loc[rows, cols]` to select a list of rows and a list of columns from the dataframe. In this case, we're selecting rows w/ statename 'NI' by using a boolean list `googletrend.State=='NI'` and selecting "State". | googletrend['Date'] = googletrend.week.str.split(' - ', expand=True)[0]
googletrend['State'] = googletrend.file.str.split('_', expand=True)[2]
googletrend.loc[googletrend.State=='NI', "State"] = 'HB,NI' | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
The following extracts particular date fields from a complete datetime for the purpose of constructing categoricals.You should *always* consider this feature extraction step when working with date-time. Without expanding your date-time into these additional fields, you can't capture any trend/cyclical behavior as a function of time at any of these granularities. We'll add to every table with a date field. | add_datepart(weather, "Date", drop=False)
add_datepart(googletrend, "Date", drop=False)
add_datepart(train, "Date", drop=False)
add_datepart(test, "Date", drop=False) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
The Google trends data has a special category for the whole of the Germany - we'll pull that out so we can use it explicitly. | trend_de = googletrend[googletrend.file == 'Rossmann_DE'] | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
Now we can outer join all of our data into a single dataframe. Recall that in outer joins everytime a value in the joining field on the left table does not have a corresponding value on the right table, the corresponding row in the new table has Null values for all right table fields. One way to check that all records are consistent and complete is to check for Null values post-join, as we do here.*Aside*: Why note just do an inner join?If you are assuming that all records are complete and match on the field you desire, an inner join will do the same thing as an outer join. However, in the event you are wrong or a mistake is made, an outer join followed by a null-check will catch it. (Comparing before/after of rows for inner join is equivalent, but requires keeping track of before/after row 's. Outer join is easier.) | store = join_df(store, store_states, "Store")
len(store[store.State.isnull()])
joined = join_df(train, store, "Store")
joined_test = join_df(test, store, "Store")
len(joined[joined.StoreType.isnull()]),len(joined_test[joined_test.StoreType.isnull()])
joined = join_df(joined, googletrend, ["State","Year", "Week"])
joined_test = join_df(joined_test, googletrend, ["State","Year", "Week"])
len(joined[joined.trend.isnull()]),len(joined_test[joined_test.trend.isnull()])
joined = joined.merge(trend_de, 'left', ["Year", "Week"], suffixes=('', '_DE'))
joined_test = joined_test.merge(trend_de, 'left', ["Year", "Week"], suffixes=('', '_DE'))
len(joined[joined.trend_DE.isnull()]),len(joined_test[joined_test.trend_DE.isnull()])
joined = join_df(joined, weather, ["State","Date"])
joined_test = join_df(joined_test, weather, ["State","Date"])
len(joined[joined.Mean_TemperatureC.isnull()]),len(joined_test[joined_test.Mean_TemperatureC.isnull()])
for df in (joined, joined_test):
for c in df.columns:
if c.endswith('_y'):
if c in df.columns: df.drop(c, inplace=True, axis=1) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
Next we'll fill in missing values to avoid complications with `NA`'s. `NA` (not available) is how Pandas indicates missing values; many models have problems when missing values are present, so it's always important to think about how to deal with them. In these cases, we are picking an arbitrary *signal value* that doesn't otherwise appear in the data. | for df in (joined,joined_test):
df['CompetitionOpenSinceYear'] = df.CompetitionOpenSinceYear.fillna(1900).astype(np.int32)
df['CompetitionOpenSinceMonth'] = df.CompetitionOpenSinceMonth.fillna(1).astype(np.int32)
df['Promo2SinceYear'] = df.Promo2SinceYear.fillna(1900).astype(np.int32)
df['Promo2SinceWeek'] = df.Promo2SinceWeek.fillna(1).astype(np.int32) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
Next we'll extract features "CompetitionOpenSince" and "CompetitionDaysOpen". Note the use of `apply()` in mapping a function across dataframe values. | for df in (joined,joined_test):
df["CompetitionOpenSince"] = pd.to_datetime(dict(year=df.CompetitionOpenSinceYear,
month=df.CompetitionOpenSinceMonth, day=15))
df["CompetitionDaysOpen"] = df.Date.subtract(df.CompetitionOpenSince).dt.days | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
We'll replace some erroneous / outlying data. | for df in (joined,joined_test):
df.loc[df.CompetitionDaysOpen<0, "CompetitionDaysOpen"] = 0
df.loc[df.CompetitionOpenSinceYear<1990, "CompetitionDaysOpen"] = 0 | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
We add "CompetitionMonthsOpen" field, limiting the maximum to 2 years to limit number of unique categories. | for df in (joined,joined_test):
df["CompetitionMonthsOpen"] = df["CompetitionDaysOpen"]//30
df.loc[df.CompetitionMonthsOpen>24, "CompetitionMonthsOpen"] = 24
joined.CompetitionMonthsOpen.unique() | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
Same process for Promo dates. | for df in (joined,joined_test):
df["Promo2Since"] = pd.to_datetime(df.apply(lambda x: Week(
x.Promo2SinceYear, x.Promo2SinceWeek).monday(), axis=1).astype(pd.datetime))
df["Promo2Days"] = df.Date.subtract(df["Promo2Since"]).dt.days
for df in (joined,joined_test):
df.loc[df.Promo2Days<0, "Promo2Days"] = 0
df.loc[df.Promo2SinceYear<1990, "Promo2Days"] = 0
df["Promo2Weeks"] = df["Promo2Days"]//7
df.loc[df.Promo2Weeks<0, "Promo2Weeks"] = 0
df.loc[df.Promo2Weeks>25, "Promo2Weeks"] = 25
df.Promo2Weeks.unique()
joined.to_feather(f'{PATH}joined')
joined_test.to_feather(f'{PATH}joined_test') | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
Durations It is common when working with time series data to extract data that explains relationships across rows as opposed to columns, e.g.:* Running averages* Time until next event* Time since last eventThis is often difficult to do with most table manipulation frameworks, since they are designed to work with relationships across columns. As such, we've created a class to handle this type of data.We'll define a function `get_elapsed` for cumulative counting across a sorted dataframe. Given a particular field `fld` to monitor, this function will start tracking time since the last occurrence of that field. When the field is seen again, the counter is set to zero.Upon initialization, this will result in datetime na's until the field is encountered. This is reset every time a new store is seen. We'll see how to use this shortly. | def get_elapsed(fld, pre):
day1 = np.timedelta64(1, 'D')
last_date = np.datetime64()
last_store = 0
res = []
for s,v,d in zip(df.Store.values,df[fld].values, df.Date.values):
if s != last_store:
last_date = np.datetime64()
last_store = s
if v: last_date = d
res.append(((d-last_date).astype('timedelta64[D]') / day1))
df[pre+fld] = res | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
We'll be applying this to a subset of columns: | columns = ["Date", "Store", "Promo", "StateHoliday", "SchoolHoliday"]
df = train[columns]
df = test[columns] | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
Let's walk through an example.Say we're looking at School Holiday. We'll first sort by Store, then Date, and then call `add_elapsed('SchoolHoliday', 'After')`:This will apply to each row with School Holiday:* A applied to every row of the dataframe in order of store and date* Will add to the dataframe the days since seeing a School Holiday* If we sort in the other direction, this will count the days until another holiday. | fld = 'SchoolHoliday'
df = df.sort_values(['Store', 'Date'])
get_elapsed(fld, 'After')
df = df.sort_values(['Store', 'Date'], ascending=[True, False])
get_elapsed(fld, 'Before') | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
We'll do this for two more fields. | fld = 'StateHoliday'
df = df.sort_values(['Store', 'Date'])
get_elapsed(fld, 'After')
df = df.sort_values(['Store', 'Date'], ascending=[True, False])
get_elapsed(fld, 'Before')
fld = 'Promo'
df = df.sort_values(['Store', 'Date'])
get_elapsed(fld, 'After')
df = df.sort_values(['Store', 'Date'], ascending=[True, False])
get_elapsed(fld, 'Before') | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
We're going to set the active index to Date. | df = df.set_index("Date") | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
Then set null values from elapsed field calculations to 0. | columns = ['SchoolHoliday', 'StateHoliday', 'Promo']
for o in ['Before', 'After']:
for p in columns:
a = o+p
df[a] = df[a].fillna(0).astype(int) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
Next we'll demonstrate window functions in pandas to calculate rolling quantities.Here we're sorting by date (`sort_index()`) and counting the number of events of interest (`sum()`) defined in `columns` in the following week (`rolling()`), grouped by Store (`groupby()`). We do the same in the opposite direction. | bwd = df[['Store']+columns].sort_index().groupby("Store").rolling(7, min_periods=1).sum()
fwd = df[['Store']+columns].sort_index(ascending=False
).groupby("Store").rolling(7, min_periods=1).sum() | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
Next we want to drop the Store indices grouped together in the window function.Often in pandas, there is an option to do this in place. This is time and memory efficient when working with large datasets. | bwd.drop('Store',1,inplace=True)
bwd.reset_index(inplace=True)
fwd.drop('Store',1,inplace=True)
fwd.reset_index(inplace=True)
df.reset_index(inplace=True) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
Now we'll merge these values onto the df. | df = df.merge(bwd, 'left', ['Date', 'Store'], suffixes=['', '_bw'])
df = df.merge(fwd, 'left', ['Date', 'Store'], suffixes=['', '_fw'])
df.drop(columns,1,inplace=True)
df.head() | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
It's usually a good idea to back up large tables of extracted / wrangled features before you join them onto another one, that way you can go back to it easily if you need to make changes to it. | df.to_feather(f'{PATH}df')
df = pd.read_feather(f'{PATH}df')
df["Date"] = pd.to_datetime(df.Date)
df.columns
joined = join_df(joined, df, ['Store', 'Date'])
joined_test = join_df(joined_test, df, ['Store', 'Date']) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
The authors also removed all instances where the store had zero sale / was closed. We speculate that this may have cost them a higher standing in the competition. One reason this may be the case is that a little exploratory data analysis reveals that there are often periods where stores are closed, typically for refurbishment. Before and after these periods, there are naturally spikes in sales that one might expect. By ommitting this data from their training, the authors gave up the ability to leverage information about these periods to predict this otherwise volatile behavior. | joined = joined[joined.Sales!=0] | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
We'll back this up as well. | joined.reset_index(inplace=True)
joined_test.reset_index(inplace=True)
joined.to_feather(f'{PATH}joined')
joined_test.to_feather(f'{PATH}joined_test') | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
We now have our final set of engineered features.While these steps were explicitly outlined in the paper, these are all fairly typical feature engineering steps for dealing with time series data and are practical in any similar setting. Create features | joined = pd.read_feather(f'{PATH}joined')
joined_test = pd.read_feather(f'{PATH}joined_test')
joined.head().T.head(40) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
Now that we've engineered all our features, we need to convert to input compatible with a neural network.This includes converting categorical variables into contiguous integers or one-hot encodings, normalizing continuous features to standard normal, etc... | cat_vars = ['Store', 'DayOfWeek', 'Year', 'Month', 'Day', 'StateHoliday', 'CompetitionMonthsOpen',
'Promo2Weeks', 'StoreType', 'Assortment', 'PromoInterval', 'CompetitionOpenSinceYear', 'Promo2SinceYear',
'State', 'Week', 'Events', 'Promo_fw', 'Promo_bw', 'StateHoliday_fw', 'StateHoliday_bw',
'SchoolHoliday_fw', 'SchoolHoliday_bw']
contin_vars = ['CompetitionDistance', 'Max_TemperatureC', 'Mean_TemperatureC', 'Min_TemperatureC',
'Max_Humidity', 'Mean_Humidity', 'Min_Humidity', 'Max_Wind_SpeedKm_h',
'Mean_Wind_SpeedKm_h', 'CloudCover', 'trend', 'trend_DE',
'AfterStateHoliday', 'BeforeStateHoliday', 'Promo', 'SchoolHoliday']
n = len(joined); n
dep = 'Sales'
joined = joined[cat_vars+contin_vars+[dep, 'Date']].copy()
joined_test[dep] = 0
joined_test = joined_test[cat_vars+contin_vars+[dep, 'Date', 'Id']].copy()
for v in cat_vars: joined[v] = joined[v].astype('category').cat.as_ordered()
apply_cats(joined_test, joined)
for v in contin_vars:
joined[v] = joined[v].fillna(0).astype('float32')
joined_test[v] = joined_test[v].fillna(0).astype('float32') | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
We're going to run on a sample. | idxs = get_cv_idxs(n, val_pct=150000/n)
joined_samp = joined.iloc[idxs].set_index("Date")
samp_size = len(joined_samp); samp_size | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
To run on the full dataset, use this instead: | samp_size = n
joined_samp = joined.set_index("Date") | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
We can now process our data... | joined_samp.head(2)
df, y, nas, mapper = proc_df(joined_samp, 'Sales', do_scale=True)
yl = np.log(y)
joined_test = joined_test.set_index("Date")
df_test, _, nas, mapper = proc_df(joined_test, 'Sales', do_scale=True, skip_flds=['Id'],
mapper=mapper, na_dict=nas)
df.head(2) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
In time series data, cross-validation is not random. Instead, our holdout data is generally the most recent data, as it would be in real application. This issue is discussed in detail in [this post](http://www.fast.ai/2017/11/13/validation-sets/) on our web site.One approach is to take the last 25% of rows (sorted by date) as our validation set. | train_ratio = 0.75
# train_ratio = 0.9
train_size = int(samp_size * train_ratio); train_size
val_idx = list(range(train_size, len(df))) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
An even better option for picking a validation set is using the exact same length of time period as the test set uses - this is implemented here: | val_idx = np.flatnonzero(
(df.index<=datetime.datetime(2014,9,17)) & (df.index>=datetime.datetime(2014,8,1)))
val_idx=[0] | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
DL We're ready to put together our models.Root-mean-squared percent error is the metric Kaggle used for this competition. | def inv_y(a): return np.exp(a)
def exp_rmspe(y_pred, targ):
targ = inv_y(targ)
pct_var = (targ - inv_y(y_pred))/targ
return math.sqrt((pct_var**2).mean())
max_log_y = np.max(yl)
y_range = (0, max_log_y*1.2) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
We can create a ModelData object directly from out data frame. | md = ColumnarModelData.from_data_frame(PATH, val_idx, df, yl.astype(np.float32), cat_flds=cat_vars, bs=128,
test_df=df_test) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
Some categorical variables have a lot more levels than others. Store, in particular, has over a thousand! | cat_sz = [(c, len(joined_samp[c].cat.categories)+1) for c in cat_vars]
cat_sz | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
We use the *cardinality* of each variable (that is, its number of unique values) to decide how large to make its *embeddings*. Each level will be associated with a vector with length defined as below. | emb_szs = [(c, min(50, (c+1)//2)) for _,c in cat_sz]
emb_szs
m = md.get_learner(emb_szs, len(df.columns)-len(cat_vars),
0.04, 1, [1000,500], [0.001,0.01], y_range=y_range)
lr = 1e-3
m.lr_find()
m.sched.plot(100) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
Sample | m = md.get_learner(emb_szs, len(df.columns)-len(cat_vars),
0.04, 1, [1000,500], [0.001,0.01], y_range=y_range)
lr = 1e-3
m.fit(lr, 3, metrics=[exp_rmspe])
m.fit(lr, 5, metrics=[exp_rmspe], cycle_len=1)
m.fit(lr, 2, metrics=[exp_rmspe], cycle_len=4) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
All | m = md.get_learner(emb_szs, len(df.columns)-len(cat_vars),
0.04, 1, [1000,500], [0.001,0.01], y_range=y_range)
lr = 1e-3
m.fit(lr, 1, metrics=[exp_rmspe])
m.fit(lr, 3, metrics=[exp_rmspe])
m.fit(lr, 3, metrics=[exp_rmspe], cycle_len=1) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
Test | m = md.get_learner(emb_szs, len(df.columns)-len(cat_vars),
0.04, 1, [1000,500], [0.001,0.01], y_range=y_range)
lr = 1e-3
m.fit(lr, 3, metrics=[exp_rmspe])
m.fit(lr, 3, metrics=[exp_rmspe], cycle_len=1)
m.save('val0')
m.load('val0')
x,y=m.predict_with_targs()
exp_rmspe(x,y)
pred_test=m.predict(True)
pred_test = np.exp(pred_test)
joined_test['Sales']=pred_test
csv_fn=f'{PATH}tmp/sub.csv'
joined_test[['Id','Sales']].to_csv(csv_fn, index=False)
FileLink(csv_fn) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
RF | from sklearn.ensemble import RandomForestRegressor
((val,trn), (y_val,y_trn)) = split_by_idx(val_idx, df.values, yl)
m = RandomForestRegressor(n_estimators=40, max_features=0.99, min_samples_leaf=2,
n_jobs=-1, oob_score=True)
m.fit(trn, y_trn);
preds = m.predict(val)
m.score(trn, y_trn), m.score(val, y_val), m.oob_score_, exp_rmspe(preds, y_val) | _____no_output_____ | Apache-2.0 | courses/dl1/lesson3-rossman.ipynb | linbojin/fastai |
Dataset Marlon Soybean, CBOT Soybean Futures + ( Global Historical Climatology Network (GHCN) filtered by USDA-NASS-soybeans-production_bushels-2015) Soybean, CBOT Soybean Futures- https://blog.quandl.com/api-for-commodity-data- http://www.quandl.com/api/v3/datasets/CHRIS/CME_S1/ Global Historical Climatology Network (GHCN)- https://www.ncdc.noaa.gov/data-access/land-based-station-data/land-based-datasets/global-historical-climatology-network-ghcn- FTP: ftp://ftp.ncdc.noaa.gov/pub/data/ghcn/daily/by_year/ USDA-NASS-soybeans-production_bushels-2015- https://usda-reports.nautilytics.com/?crop=soybeans&statistic=production_dollars&year=2007- https://www.nass.usda.gov/Data_Visualization/index.phphttps://github.com/aaronpenne/get_noaa_ghcn_data.git Imports | %matplotlib inline
import matplotlib.pyplot as plt
import matplotlib.ticker as mticker
import pandas as pd
import numpy as np
import os
from six.moves import urllib
from ftplib import FTP
from io import StringIO
from IPython.display import clear_output
from functools import reduce
import tarfile
import subprocess
#subprocess.run(["ls", "-l"])
import zipfile
import shutil # move files
import psutil
# Load the Drive helper and mount
from google.colab import drive
drive.mount('/content/drive') | Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount("/content/drive", force_remount=True).
| Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
Defines | ROOT_PATH = "drive/My Drive/TCC/"
DATASETS_PATH = ROOT_PATH + "datasets/"
SOYBEAN_PATH = DATASETS_PATH + "CBOTSoybeanFutures/"
WEATHER_PATH = DATASETS_PATH + "GHCN_Data/"
SOYBEAN_URL = "http://www.quandl.com/api/v3/datasets/CHRIS/CME_S1/data.csv"
USDA_PATH = "datasets/USDA-NASS-soybeans-production_bushels-2015/"
WEATHER_PATH_DRIVE_ZIP = WEATHER_PATH + "data/zip/"
WEATHER_PATH_DRIVE_CSV = WEATHER_PATH + "data/csv/"
FIXED_STATE_FILE = WEATHER_PATH + "fixed_states.txt"
CALCULATED_STATE_FILE = WEATHER_PATH + "calculated_states.txt"
DOWNLOADED_STATIONS_FILE = WEATHER_PATH + "downloaded_stations.txt"
DOWNLOADED_STATIONS_FILE_TEMP = DOWNLOADED_STATIONS_FILE
plt.rcParams["figure.figsize"] = [19,15]
plt.rcParams.update({'font.size': 27})
# Create directories
# and initial files
if not os.path.exists(SOYBEAN_PATH):
os.makedirs(SOYBEAN_PATH)
if not os.path.exists(WEATHER_PATH_DRIVE_ZIP):
os.makedirs(WEATHER_PATH_DRIVE_ZIP)
if not os.path.exists(WEATHER_PATH_DRIVE_CSV):
os.makedirs(WEATHER_PATH_DRIVE_CSV)
if not os.path.exists(DOWNLOADED_STATIONS_FILE):
open(DOWNLOADED_STATIONS_FILE,'a').close()
if not os.path.exists(DOWNLOADED_STATIONS_FILE_TEMP):
open(DOWNLOADED_STATIONS_FILE_TEMP,'a').close()
if not os.path.exists(FIXED_STATE_FILE):
open(FIXED_STATE_FILE,'a').close()
if not os.path.exists(CALCULATED_STATE_FILE):
open(CALCULATED_STATE_FILE,'a').close()
| _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
ERROR: type should be string, got " https://github.com/aaronpenne/get_noaa_ghcn_data.git https://github.com/aaronpenne/get_noaa_ghcn_data/blob/master/get_station_id.py" | # -*- coding: utf-8 -*-
"""
Searches list of stations via user input to find the station ID.
Author: Aaron Penne
------------------------------
Variable Columns Type
------------------------------
ID 1-11 Character
LATITUDE 13-20 Real
LONGITUDE 22-30 Real
ELEVATION 32-37 Real
STATE 39-40 Character
NAME 42-71 Character
GSN FLAG 73-75 Character
HCN/CRN FLAG 77-79 Character
WMO ID 81-85 Character
------------------------------
"""
import sys
import pandas as pd
from ftplib import FTP
import os
output_dir = os.path.relpath('output')
if not os.path.isdir(output_dir):
os.mkdir(output_dir)
ftp_path_dly = '/pub/data/ghcn/daily/'
ftp_path_dly_all = '/pub/data/ghcn/daily/all/'
ftp_filename = 'ghcnd-stations.txt'
def connect_to_ftp():
ftp_path_root = 'ftp.ncdc.noaa.gov'
# Access NOAA FTP server
ftp = FTP(ftp_path_root)
message = ftp.login() # No credentials needed
print(message)
return ftp
def get_station_id(ftp, search_term):
'''
Get stations file
'''
ftp_full_path = os.path.join(ftp_path_dly, ftp_filename)
local_full_path = os.path.join(output_dir, ftp_filename)
if not os.path.isfile(local_full_path):
with open(local_full_path, 'wb+') as f:
ftp.retrbinary('RETR ' + ftp_full_path, f.write)
'''
Get user search term
'''
query = search_term
query = query.upper()
print("> Query: '"+query+"'")
'''
Read stations text file using fixed-width-file reader built into pandas
'''
# http://pandas.pydata.org/pandas-docs/stable/generated/pandas.read_fwf.html
dtype = {'STATION_ID': str,
'LATITUDE': str,
'LONGITUDE': str,
'ELEVATION': str,
'STATE': str,
'STATION_NAME': str,
'GSN_FLAG': str,
'HCN_CRN_FLAG': str,
'WMO_ID': str}
names = ['STATION_ID', 'LATITUDE', 'LONGITUDE', 'ELEVATION', 'STATE', 'STATION_NAME', 'GSN_FLAG', 'HCN_CRN_FLAG', 'WMO_ID']
widths = [11, # Station ID
9, # Latitude (decimal degrees)
10, # Longitude (decimal degrees)
7, # Elevation (meters)
3, # State (USA stations only)
31, # Station Name
4, # GSN Flag
4, # HCN/CRN Flag
6] # WMO ID
df = pd.read_fwf(local_full_path, widths=widths, names=names, dtype=dtype, header=None)
'''
Replace missing values (nan, -999.9)
'''
df['STATE'] = df['STATE'].replace('nan', '--')
df['GSN_FLAG'] = df['GSN_FLAG'].replace('nan', '---')
df['HCN_CRN_FLAG'] = df['GSN_FLAG'].replace('nan', '---')
df = df.replace(-999.9, float('nan'))
try:
'''
Get query results, but only the columns we care about
'''
print('Searching records...')
matches = df['STATION_ID'].str.contains(query)
df = df.loc[matches, ['STATION_ID', 'LATITUDE', 'LONGITUDE', 'ELEVATION', 'STATE', 'STATION_NAME']]
df.reset_index(drop=True, inplace=True)
'''
Get file sizes of each station's records to augment results
'''
#print('Getting file sizes...', end='')
#print(df.index)
#ftp.voidcmd('TYPE I') # Needed to avoid FTP error with ftp.size()
#count=0
#last = ''
#for i in list(df.index):
# count = count + 1
# print('.', end='')
# ftp_dly_file = ftp_path_dly + 'all/' + df.loc[i, 'STATION_ID'] + '.dly'
# #print(df.loc[i, 'STATION_ID'], end='')
# df.loc[i, 'SIZE'] = round(ftp.size(ftp_dly_file)/1000) # Kilobytes
# #print('size: %d KB' %round(ftp.size(ftp_dly_file)/1000))
# actual = " %.1f%% " % ((count/df.index.size)*100)
# if (actual != last):
# clear_output()
# last = actual
# #print("%.2f%% " %((count/df.index.size)*100), end='')
# print('Getting file sizes...')
# print(str(actual) + ' ['+ str(count) + ' of ' + str(df.index.size) + ']')
print()
print()
'''
Sort by size then by rounded lat/long values to group geographic areas and show stations with most data
'''
df_sort = df.round(0)
#df_sort.sort_values(['LATITUDE', 'LONGITUDE', 'SIZE'], ascending=False, inplace=True)
df_sort.sort_values(['LATITUDE', 'LONGITUDE'], ascending=False, inplace=True)
df = df.loc[df_sort.index]
df.reset_index(drop=True, inplace=True)
except:
print('Station not found')
traceback.print_exc(file=sys.stdout)
ftp.quit()
sys.exit()
'''
Print headers and values to facilitate reading
'''
#selection = 'Index'
#station_id = 'Station_ID '
#lat = 'Latitude'
#lon = 'Longitude'
#state = 'State'
#name = 'Station_Name '
#size = ' File_Size'
# Format output to be pretty, hopefully there is a prettier way to do this.
#print('{: <6}{: <31}{: <6}({: >8},{: >10}){: >13}'.format(selection, name, state, lat, lon, size))
#print('-'*5 + ' ' + '-'*30 + ' ' + '-'*5 + ' ' + '-'*21 + ' ' + '-'*12)
#for i in list(df.index):
# print('{: 4}: {: <31}{: <6}({: >8},{: >10}){: >10} Kb'.format(i,
# df.loc[i,'STATION_NAME'],
# df.loc[i,'STATE'],
# df.loc[i,'LATITUDE'],
# df.loc[i,'LONGITUDE'],
# df.loc[i,'SIZE']))
# '''
# Get user selection
# '''
# try:
# query = input('Enter selection (ex. 001, 42): ')
# query = int(query)
# except:
# print('Please enter valid selection (ex. 001, 42)')
# ftp.quit()
# sys.exit()
#station_id = df.loc[query, 'STATION_ID']
station_id = df
return station_id
def get_station(ftp=None, search_term='US'):
close_after = False
if ftp==None:
ftp = connect_to_ftp()
close_after = True
station_id = get_station_id(ftp,search_term)
#print(station_id)
if close_after:
ftp.quit()
return (station_id)
| _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
https://github.com/aaronpenne/get_noaa_ghcn_data/blob/master/get_dly.py |
"""
Grabs .dly file from the NOAA GHCN FTP server, parses, and reshapes to have one
day per row and element values in the columns. Writes output as CSV.
Author: Aaron Penne
.dly Format In (roughly): .csv Format Out (roughly):
------------------------- --------------------------
Month1 PRCP Day1 Day2 ... Day31 Day1 PRCP SNOW
Month1 SNOW Day1 Day2 ... Day31 Day2 PRCP SNOW
Month2 PRCP Day1 Day2 ... Day31 Day3 PRCP SNOW
Month2 SNOW Day1 Day2 ... Day31 Day4 PRCP SNOW
Starting with 5 core elements (per README)
PRCP = Precipitation (tenths of mm)
SNOW = Snowfall (mm)
SNWD = Snow depth (mm)
TMAX = Maximum temperature (tenths of degrees C)
TMIN = Minimum temperature (tenths of degrees C)
ICD:
------------------------------
Variable Columns Type
------------------------------
ID 1-11 Character
YEAR 12-15 Integer
MONTH 16-17 Integer
ELEMENT 18-21 Character
VALUE1 22-26 Integer
MFLAG1 27-27 Character
QFLAG1 28-28 Character
SFLAG1 29-29 Character
VALUE2 30-34 Integer
MFLAG2 35-35 Character
QFLAG2 36-36 Character
SFLAG2 37-37 Character
. . .
. . .
. . .
VALUE31 262-266 Integer
MFLAG31 267-267 Character
QFLAG31 268-268 Character
SFLAG31 269-269 Character
------------------------------
"""
import pandas as pd
from ftplib import FTP
from io import StringIO
import os
ftp_path_dly_all = '/pub/data/ghcn/daily/all/'
def connect_to_ftp():
"""
Get FTP server and file details
"""
ftp_path_root = 'ftp.ncdc.noaa.gov'
# Access NOAA FTP server
ftp = FTP(ftp_path_root)
message = ftp.login() # No credentials needed
#print(message)
return ftp
# Marlon Franco
def disconnect_to_ftp(ftp_connection):
return ftp_connection.quit()
def get_flags(s):
"""
Get flags, replacing empty flags with '_' for clarity (' S ' becomes '_S_')
"""
m_flag = s.read(1)
m_flag = m_flag if m_flag.strip() else '_'
q_flag = s.read(1)
q_flag = q_flag if q_flag.strip() else '_'
s_flag = s.read(1)
s_flag = s_flag if s_flag.strip() else '_'
return [m_flag + q_flag + s_flag]
def create_dataframe(element, dict_element):
"""
Make dataframes out of the dicts, make the indices date strings (YYYY-MM-DD)
"""
element = element.upper()
df_element = pd.DataFrame(dict_element)
# Add dates (YYYY-MM-DD) as index on df. Pad days with zeros to two places
df_element.index = df_element['YEAR'] + '-' + df_element['MONTH'] + '-' + df_element['DAY'].str.zfill(2)
df_element.index.name = 'DATE'
# Arrange columns so ID, YEAR, MONTH, DAY are at front. Leaving them in for plotting later - https://stackoverflow.com/a/31396042
for col in ['DAY', 'MONTH', 'YEAR', 'ID']:
df_element = move_col_to_front(col, df_element)
# Convert numerical values to float
df_element.loc[:,element] = df_element.loc[:,element].astype(float)
return df_element
def move_col_to_front(element, df):
element = element.upper()
cols = df.columns.tolist()
cols.insert(0, cols.pop(cols.index(element)))
df = df.reindex(columns=cols)
return df
def dly_to_csv(ftp, station_id, output_dir, save_dly):
#output_dir = os.path.relpath('output')
if not os.path.isdir(output_dir):
os.makedirs(output_dir)
ftp_filename = station_id + '.dly'
# Write .dly file to stream using StringIO using FTP command 'RETR'
s = StringIO()
ftp.retrlines('RETR ' + ftp_path_dly_all + ftp_filename, s.write)
s.seek(0)
# Write .dly file to dir to preserve original # FIXME make optional?
if (save_dly):
with open(os.path.join(output_dir, ftp_filename), 'wb+') as f:
ftp.retrbinary('RETR ' + ftp_path_dly_all + ftp_filename, f.write)
# Move to first char in file
s.seek(0)
# File params
num_chars_line = 269
num_chars_metadata = 21
element_list = ['PRCP', 'SNOW', 'SNWD', 'TMAX', 'TMIN']
'''
Read through entire StringIO stream (the .dly file) and collect the data
'''
all_dicts = {}
element_flag = {}
prev_year = '0000'
i = 0
while True:
i += 1
'''
Read metadata for each line (one month of data for a particular element per line)
'''
id_station = s.read(11)
year = s.read(4)
month = s.read(2)
day = 0
element = s.read(4)
# If this is blank then we've reached EOF and should exit loop
if not element:
break
'''
Print status
'''
if year != prev_year:
#print('Year {} | Line {}'.format(year, i))
prev_year = year
'''
Loop through each day in rest of row, break if current position is end of row
'''
while s.tell() % num_chars_line != 0:
day += 1
# Fill in contents of each dict depending on element type in current row
if day == 1:
try:
first_hit = element_flag[element]
except:
element_flag[element] = 1
all_dicts[element] = {}
all_dicts[element]['ID'] = []
all_dicts[element]['YEAR'] = []
all_dicts[element]['MONTH'] = []
all_dicts[element]['DAY'] = []
all_dicts[element][element.upper()] = []
all_dicts[element][element.upper() + '_FLAGS'] = []
value = s.read(5)
flags = get_flags(s)
if value == '-9999':
continue
all_dicts[element]['ID'] += [station_id]
all_dicts[element]['YEAR'] += [year]
all_dicts[element]['MONTH'] += [month]
all_dicts[element]['DAY'] += [str(day)]
all_dicts[element][element.upper()] += [value]
all_dicts[element][element.upper() + '_FLAGS'] += flags
'''
Create dataframes from dict
'''
all_dfs = {}
for element in list(all_dicts.keys()):
all_dfs[element] = create_dataframe(element, all_dicts[element])
'''
Combine all element dataframes into one dataframe, indexed on date.
'''
# pd.concat automagically aligns values to matching indices, therefore the data is date aligned!
list_dfs = []
for df in list(all_dfs.keys()):
list_dfs += [all_dfs[df]]
df_all = pd.concat(list_dfs, axis=1, sort=False)
df_all.index.name = 'MM/DD/YYYY'
'''
Remove duplicated/broken columns and rows
'''
# https://stackoverflow.com/a/40435354
df_all = df_all.loc[:,~df_all.columns.duplicated()]
df_all = df_all.loc[df_all['ID'].notnull(), :]
'''
Output to CSV, convert everything to strings first
'''
# NOTE: To open the CSV in Excel, go through the CSV import wizard, otherwise it will come out broken
df_out = df_all.astype(str)
df_out.to_csv(os.path.join(output_dir, station_id + '.csv'))
#print('\nOutput CSV saved to: {}'.format(os.path.join(output_dir, station_id + '.csv')))
def get_weather_data(ftp=None, station_id='USR0000CCHC',output_dir=WEATHER_PATH, save_dly=False):
close_after = False
if ftp==None:
ftp = connect_to_ftp()
close_after = True
dly_to_csv(ftp, station_id,output_dir, save_dly)
if close_after:
ftp.quit() | _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
Fetch Data | def fetch_soybean_data(soybean_url=SOYBEAN_URL, soybean_path=SOYBEAN_PATH):
if not os.path.isdir(soybean_path):
os.makedirs(soybean_path)
csv_path = os.path.join(soybean_path, "soybeans.csv")
urllib.request.urlretrieve(soybean_url, csv_path)
def fetch_weather_data(contains='US', weather_path=WEATHER_PATH_DRIVE_CSV, save_dly=False, how_much=100):
conn = connect_to_ftp()
weather = get_station(conn,search_term=contains) # List all stations from USA
downloaded_stations = ""
with open(DOWNLOADED_STATIONS_FILE_TEMP,"r+") as f:
downloaded_stations = f.read()
count = 0
count2 = 0
total = weather['STATION_ID'].size
amount_of_data = total * how_much/100
last = ''
for station in weather['STATION_ID']:
print('.',end='')
count += 1
actual = "%.2f%% " %((count/total)*100)
actual_partial = "%.2f%% " %((count2/amount_of_data)*100)
if (station+'.csv' not in downloaded_stations):
if (count2 > amount_of_data):
print('download completed: ['+str(count2)+' of '+str(amount_of_data)+'], total = '+str(total))
return True
count2 += 1
print('get ', end='')
get_weather_data(conn, station, weather_path, save_dly)
print('done')
downloaded_stations += station+'.csv\r\n'
with open(DOWNLOADED_STATIONS_FILE_TEMP,"a+") as f:
f.write(station+'.csv\r\n')
else:
print(',',end='')
if (actual != last):
clear_output()
last = actual
print('Getting '+str(how_much)+'% of weather data from GHCN ftp containing \''+contains+'\' in STATION_ID...')
print('PARTIAL: '+str(actual_partial) + ' ['+ str(count2) + ' of ' + str(amount_of_data) + ']')
print('TOTAL: '+str(actual) + ' ['+ str(count) + ' of ' + str(total) + ']')
disconnect_to_ftp(conn)
print('Final: download completed: ['+str(count2)+' of '+str(amount_of_data)+'], total = '+str(total))
return True
# Update the local temp control file
#!echo "$DOWNLOADED_STATIONS_FILE" > "$DOWNLOADED_STATIONS_FILE_TEMP" .
#fetch_weather_data(how_much=0.54) #0.54% of total amount of data
fetch_weather_data()
# Update the control file
!echo "$DOWNLOADED_STATIONS_FILE_TEMP" > "$DOWNLOADED_STATIONS_FILE" . | _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
Check the number of downloaded station's csv file | weather = get_station(search_term='US') # List all stations from USA
print("# of stations in GHCN FTP: ", end="")
print(str(weather['STATION_ID'].size))
print("# of downloaded csv files: ", end="")
!find "$WEATHER_PATH_DRIVE_CSV" -type f | wc -l
print("# of downloaded stations in control file: ", end="")
with open(DOWNLOADED_STATIONS_FILE) as f:
num_lines = sum(1 for _ in f.readlines())
print(str(num_lines))
def force_update_control_file():
directory = os.path.join(WEATHER_PATH_DRIVE_CSV)
with open(DOWNLOADED_STATIONS_FILE,"w") as f:
for root,dirs,files in os.walk(directory):
for file in files:
print('.',end='')
if file.endswith(".csv"):
f.write(file+'\r\n')
force_update_control_file() | .................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................... | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
Get 'US' Stations | newfile = ''
with open(PROJECT_PATH+'ghcnd-stations-us.txt', 'r') as f:
for line in f.readlines():
line_list = line.split(' ')
station = line_list[0]
newfile += station
for word in line_list:
if (len(word) > 1):
if (word[0].isalpha() and word!=station):
state = word
newfile += ','+state+'\n'
break
print(newfile)
with open(PROJECT_PATH+'ghcnd-stations-us.csv', 'w+') as f:
f.write(newfile) | _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
Organize Stations by State | def organize_stations_by_state():
f1='' #stations_not_dowloaded
csv_path = WEATHER_PATH_DRIVE_CSV
with open(WEATHER_PATH+'ghcnd-stations-us.csv', 'r') as f:
for line in f:
station = line.split(',')[0]
state = line.split(',')[1].rstrip()
# Create target Directory if don't exist
if not os.path.exists(csv_path+state):
os.mkdir(csv_path+state)
print("Directory " , csv_path+state , " Created ")
#else:
# print("Directory " , "csv/"+state , " already exists")
if not os.path.exists(csv_path+station+".csv"):
print(".", end='')
f1+=station+"\n"
else:
os.rename(csv_path+station+".csv", csv_path+state+"/"+station+".csv")
with open(WEATHER_PATH+'stations_not_dowloaded.csv', 'w+') as f:
f.write(f1)
!ls drive/My\ Drive/TCC/
sLength = len(df1['TMAX'])
df1['e'] = p.Series(np.random.randn(sLength), index=df1.index) | _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
Fix columns | def fix_columns(df):
for column in df:
if column in ('ID','TMAX','TMIN','TAVG','PRCP'):
pass
else:
#print(' deleting ',column, end='')
del(df[column])
if 'TMAX' not in df:
#print(' creating TMAX... ', end='')
#sLength = sizes['TMAX']
df['TMAX'] = pd.Series(0, index=df.index)
if 'TMIN' not in df:
#print(' creating TMIN... ', end='')
#sLength = sizes['TMIN']
df['TMIN'] = pd.Series(0, index=df.index)
if 'TAVG' not in df:
#print(' creating TAVG... ', end='')
#sLength = sizes['TAVG']
df['TAVG'] = pd.Series(0, index=df.index)
if 'PRCP' not in df:
#print(' creating PRCP... ')
#sLength = sizes['PRCP']
df['PRCP'] = pd.Series(0, index=df.index)
df=df.fillna(method='ffill')
df_ref = load_single_csv(CSV_PATH+'WA/USS0017B04S.csv')
sizes = {'TMAX':len(df_ref['TMAX']),'TMIN':len(df_ref['TMIN']),'TAVG':len(df_ref['TAVG']),'PRCP':len(df_ref['PRCP'])}
def fix_dataframes(folder=''):
root_path = CSV_PATH+folder+'/'
print(root_path)
count=0
count2=10
total=0
for root, dirs, files in os.walk(root_path):
total=len(files)
for file in files:
if '.csv' in file:
station=file.strip('.csv')
#print(station)
path = os.path.join(root, file)
df = load_single_csv(path)
fix_columns(df)
new_path = os.path.join(PROJECT_PATH+'new/'+folder+'/', file)
# Create target Directory if don't exist
if not os.path.exists(PROJECT_PATH+'new/'+folder+'/'):
os.mkdir(PROJECT_PATH+'new/'+folder+'/')
print("Directory " , PROJECT_PATH+'new/'+folder+'/' , " Created ")
df.to_csv(new_path)
if count2 == 70:
count2=0
actual = "%.2f%% " %((count/total)*100)
clear_output()
print('Fixing ',folder,' stations... ',actual,' (',str(count),' of ',str(total),')')
count+=1
count2+=1
print('Done: %.2f%% ' %((count/total)*100))
return True
fixed_states = ""
with open(FIXED_STATE_FILE, "r+") as f:
fixed_states = f.read()
print('Already fixed:',fixed_states)
for root, dirs, files in os.walk(CSV_PATH):
total=len(dirs)
for state in dirs:
if (state not in fixed_states):
if(fix_dataframes(state)):
fixed_states+= state
with open(FIXED_STATE_FILE,"a") as f:
f.write(state+'\r\n')
fix_dataframes('CA')
df1 = load_single_csv('drive/My Drive/TCC/datasets/GHCN_Data/data/csv/TX/US1TXAC0002.csv')
df2 = load_single_csv('drive/My Drive/TCC/datasets/GHCN_Data/data/new/TX/US1TXAC0002.csv')
df1.tail()
df2.tail() | _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
Load Data | def load_soybean_data(soybean_path=SOYBEAN_PATH):
csv_path = os.path.join(soybean_path, "soybeans.csv")
print(csv_path)
return pd.read_csv(csv_path)
def load_single_csv(csv_path):
#print(csv_path)
df = pd.read_csv(csv_path,low_memory=False)
df.set_index(['MM/DD/YYYY','YEAR','MONTH','DAY'], inplace=True)
return df
def extract_zip(dir_name=WEATHER_PATH_DRIVE_ZIP,destination_dir=WEATHER_PATH_DRIVE_CSV):
for item in os.listdir(dir_name): # loop through items in dir
if item.endswith(".zip"): # check for ".zip" extension
print("Extracting "+str(item), end="")
#file_name = os.path.abspath(item) # get full path of files
file_name = dir_name+item # get full path of files
zip_ref = zipfile.ZipFile(file_name) # create zipfile object
zip_ref.extractall(destination_dir) # extract file to dir
zip_ref.close() # close file
print("... OK!")
#os.remove(file_name) # delete zipped file
print("Extraction complete!")
def load_weather_data(weather_path=WEATHER_PATH_DRIVE_CSV,from_zip=False):
if from_zip:
extract_zip()
data_frames=[]
#first=True
directory = os.path.join(weather_path,"")
print(directory)
for root,dirs,files in os.walk(directory):
print(directory+".")
for file in files:
print(".")
if file.endswith(".csv"):
csv_path = os.path.join(weather_path, file)
df = load_single_csv(csv_path)
#Rename Columns
#df=df.drop(columns=['ID'])
#station = file.replace('.csv','')
#for column in df.columns:
# if(column not in ['MM/DD/YYYY','YEAR','MONTH','DAY']):
# df.rename(columns={column: station +'-'+ column}, inplace=True)
#print(station +'-'+ column)
#Append to list
data_frames.append(df)
#if (first):
# data_frames = df
# first=False
#else:
# data_frames = pd.merge(data_frames, df, on=['MM/DD/YYYY','YEAR','MONTH','DAY'], how='left')
return data_frames
#return pd.concat(data_frames, axis=1)
def load_usda_data(usda_path=USDA_PATH):
csv_path = os.path.join(usda_path, "data.csv")
print(csv_path)
return pd.read_csv(csv_path, thousands=',')
| _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
Calculate Mean and Standard Deviation for each state | def save_csv(df,name,path):
#print('Saving DataFrame in ',path)
# Create target Directory if don't exist
if not os.path.exists(path):
os.mkdir(path)
print("Directory " , path , " Created ")
df.to_csv(path+name)
def read_log(file_path):
files_processed = ""
if not os.path.exists(file_path):
with open(file_path,'w+') as f:
files_processed = f.read()
else :
with open(file_path,'r+') as f:
files_processed = f.read()
return files_processed
def write_log(file_path,content):
with open(file_path,'w+') as f:
f.write(content)
def calculate(daf):
print('TMAX',end='')
daf['TMAX_mean'] = daf[[col for col in daf.columns if 'TMAX' in col ]].mean(1)
print(' Ok mean')
daf['TMAX_std'] = daf[[col for col in daf.columns if 'TMAX' in col ]].std(1)
print(' Ok std')
#daf = daf.drop(columns=['TMAX'])
print(' OK\nTMIN', end='')
daf['TMIN_mean'] = daf[[col for col in daf.columns if 'TMIN' in col ]].mean(1)
print(' Ok mean')
daf['TMIN_std'] = daf[[col for col in daf.columns if 'TMIN' in col ]].std(1)
print(' Ok std')
#daf = daf.drop(columns=['TMIN'])
print(' OK\nTAVG', end='')
daf['TAVG_mean'] = daf[[col for col in daf.columns if 'TAVG' in col ]].mean(1)
print(' Ok mean')
daf['TAVG_std'] = daf[[col for col in daf.columns if 'TAVG' in col ]].std(1)
print(' Ok std')
#daf = daf.drop(columns=['TAVG'])
print(' OK\nPRCP', end='')
daf['PRCP_mean'] = daf[[col for col in daf.columns if 'PRCP' in col ]].mean(1)
print(' Ok mean')
daf['PRCP_std'] = daf[[col for col in daf.columns if 'PRCP' in col ]].std(1)
print(' Ok std')
#daf = daf.drop(columns=['PRCP'])
daf = daf.drop(columns=[col for col in daf.columns if col not in ['MM/DD/YYYY','YEAR','MONTH','DAY','TMAX_mean','TMAX_std','TMIN_mean','TMIN_std','TAVG_mean','TAVG_std','PRCP_mean','PRCP_std']])
print(' OK')
daf=daf.fillna(0)
return daf
def calculate_mean(folder=''):
root_path = WEATHER_PATH_DRIVE_CSV+folder+'/'
new_path = os.path.join(WEATHER_PATH+'new/'+folder+'/','')
file_path = new_path+folder+'.txt'
if not os.path.exists(new_path):
os.mkdir(new_path)
print("Directory " , new_path , " Created ")
files_processed = read_log(file_path)
print(root_path)
n=0
count=0
count2=70
count_to_save=0
count_to_reset=0
total=0
already_readed=False
for root, dirs, files in os.walk(root_path):
total=len(files)
for file in files:
if '.csv' in file:
station=file.strip('.csv')
if (station not in files_processed):
path = os.path.join(root, file)
df = load_single_csv(path)
df = df.drop(columns=['ID'])
if not already_readed:
try:
daf = load_single_csv(new_path+folder+'_tmp.csv')
except:
daf = df
already_readed=True
daf = pd.concat([daf,df], axis=1)
if count_to_save == 100:
count_to_save=0
print('saving')
save_csv(daf,folder+'_tmp',new_path)
write_log(file_path,files_processed)
print('saved')
count_to_save+=1
files_processed+=station+'\r\n'
del df
if count2 == 70:
count2=0
actual = "%.2f%% " %((count/total)*100)
clear_output()
process = psutil.Process(os.getpid())
print('RAM usage: %.2f GB' %((process.memory_info().rss) / 1e9))
print('Loading ',folder,' stations in DataFrames... ',actual,' (',str(count),' of ',str(total),')')
count+=1
count2+=1
save_csv(daf,folder+'_tmp',new_path)
write_log(file_path,files_processed)
print('Load done: %.2f%% ' %((count/total)*100))
if("Done" not in files_processed):
daf = load_single_csv(new_path+folder+'_tmp.csv')
daf = calculate(daf)
new_file_name = state+str(n)+'.csv'
print('Saving ', new_file_name)
save_csv(daf,new_file_name,new_path)
print('Done saving ',new_file_name)
write_log(file_path,files_processed+"Done\r\n")
print('Done!')
os.remove(new_path+folder+'_tmp.csv')
else :
print('Already processed.')
return True
calculate_mean('FL')
calculated_states = ""
with open(CALCULATED_STATE_FILE, "r+") as f:
calculated_states = f.read()
print('Already calculated:\n[\n',calculated_states,']\n')
for root, dirs, files in os.walk(WEATHER_PATH_DRIVE_CSV):
total=len(dirs)
for state in dirs:
if (state not in calculated_states):
if(calculate_mean(state)):
calculated_states+= state
with open(CALCULATED_STATE_FILE,"a") as f:
f.write(state+'\r\n')
| _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
Make the Date column as index | soybeans.Date = pd.to_datetime(soybeans.Date)
soybeans.set_index('Date', inplace=True)
soybeans.head()
soybeans.tail()
plt.plot(soybeans.index, soybeans.Settle)
plt.title('CBOT Soybean Futures',fontsize=27)
plt.ylabel('Price (0.01 $USD)',fontsize=27)
plt.gca().yaxis.set_major_formatter(mticker.FormatStrFormatter('%d'))
plt.show()
usda = load_usda_data()
| _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
Filter soybeans by the year 2015: | mask = (soybeans['Date'] > '2015-01-01') & (soybeans['Date'] <= '2015-12-31')
soybeans = soybeans.loc[mask]
mask = (soybeans['Date'] > '2014-01-01')
soybeans = soybeans.loc[mask] | _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
Filter weather by the most productive states | weather = weather.query("state in ('IA','IL','MN','NE','IN','OH','SD','ND','MO','AR','KS','MS','MI','WI','KY','TN','LA','NC','PA','VA','MD','AL','GA','NY','OK','SC','DE','NJ','TX','WV','FL')") | _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
Plot map data | stations = pd.read_csv('US_stations.csv')
#stations.set_index(['LATITUDE','LONGITUDE'], inplace=True)
stations.index.names
#stations.drop_duplicates(subset=['LATITUDE','LONGITUDE'])
stations.plot(kind="scatter", x="LONGITUDE", y="LATITUDE",fontsize=27,figsize=(20,15))
plt.title("Meteorological stations in the USA's most soy producing regions", fontsize=27)
plt.gca().yaxis.set_major_locator(plt.NullLocator())
plt.gca().xaxis.set_major_formatter(plt.NullFormatter())
plt.axis('off')
plt.show()
weather.drop_duplicates(subset=['latitude','longitude']).plot(kind="scatter", x="longitude", y="latitude",fontsize=27,figsize=(20,15))
plt.title("Meteorological stations in the USA's most soy producing regions", fontsize=27)
plt.gca().yaxis.set_major_locator(plt.NullLocator())
plt.gca().xaxis.set_major_formatter(plt.NullFormatter())
plt.axis('off')
plt.show() | _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
Group data by date (daily)Mรฉdia das medidas horรกrias para o avgtemp, mintemp e maxtemp | weather = weather.groupby(['date'], as_index=False)['date','mintemp','maxtemp','avgtemp'].mean()
weather.head()
weather.date = pd.to_datetime(weather.date)
weather.set_index('date', inplace=True) | _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
Join datasets (soybeans + weather) | dtMarlon = soybeans.join(weather)
dtMarlon.head() | _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
Histograms | soybeans.hist(bins=50, figsize=(20,15))
plt.show()
weather.hist(bins=50, figsize=(20,15))
plt.show()
dtMarlon.hist(bins=50, figsize=(20,15))
plt.show() | _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
Time Series | plt.plot(soybeans.index, soybeans.Settle)
plt.title('CBOT Soybean Futures',fontsize=27)
plt.ylabel('Price (0.01 $USD)',fontsize=27)
plt.gca().yaxis.set_major_formatter(mticker.FormatStrFormatter('%d'))
plt.show()
plt.plot(weather.index, weather.avgtemp)
plt.title('2015 USA Weather Avg, Max, Min')
plt.ylabel('Avg. Temp. (ยฐF)');
plt.show()
fig = plt.figure()
ax1 = fig.add_subplot(111)
ax1.plot(dtMarlon.index, dtMarlon.avgtemp, 'g-')
ax1.set_ylabel('Avg. Temp. (ยฐF)')
ax2 = ax1.twinx()
ax2.plot(dtMarlon.index, dtMarlon.Settle, 'b-')
ax2.set_ylabel('Price per bushel (0.01 $USD)')
ax2.yaxis.set_major_formatter(mticker.FormatStrFormatter('%0.01d'))
plt.title('2015 USA Weather Avg and CBOT Soybean Futures')
plt.show() | _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
Missing values for weather in days where we have soy quotes | dtMarlon.query('avgtemp.isnull()')
weather.query("date=='2015-06-05' or date=='2015-06-04' or date=='2015-06-03' or date=='2015-06-02' or date=='2015-06-01'")
soybeans.query("Date=='2015-06-05' or Date=='2015-06-04' or Date=='2015-06-03' or Date=='2015-06-02' or Date=='2015-06-01'") | _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
Filling missing values with method 'ffil'This propagate non-null values forward | dtMarlon = dtMarlon.fillna(method='ffill') | _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
Correlation | dtMarlon.corr()
dtMarlon.diff().corr()
pd.plotting.autocorrelation_plot(dtMarlon) | _____no_output_____ | Apache-2.0 | Create_Dataset_MarlonFranco.ipynb | marlonrcfranco/DatasetMarlon |
numpy- ํ๋ ฌ / ์ ํ๋์ / ํต๊ณ ํจํค์ง- ๋จธ์ ๋ฌ๋์ ์ด๋ก ์ ๋ฐฑ๊ทธ๋ผ์ด๋๋ ์ ํ๋์์ ํต๊ณ๋ก ์ด๋ฃจ์ด์ ธ ์๋ค- ์ฌ์ดํท๋ฐ ๊ฐ์ ๋จธ์ ๋ฌ๋ ํจํค์ง๊ฐ ๋ํ์ด ๊ธฐ๋ฐ์ผ๋ก ๋์ด ์๋ค * ๋จธ์ ๋ฌ๋ ์๊ณ ๋ฆฌ์ฆ์ด๋ ์ฌ์ดํ์ด์ ๊ฐ์ ๊ณผํ, ํต๊ณ ์ง์์ฉ ํจํค์ง๋ฅผ ์ง์ ๋ง๋๋ ๊ฐ๋ฐ์ด ์๋๋ผ๋ฉด ๋ํ์ด๋ฅผ ์์ธํ๊ธฐ ์ ํ์๋ ์๋ค์ง๋ง, ๋ํ์ด๋ฅผ ์ดํดํ๋ ๊ฒ์ด ํ์ด์ฌ ๊ธฐ๋ฐ์ ๋ฐ์ดํ๋ถ์๊ณผ ๋จธ์ ๋ฌ๋์ ์ค์ํ๋ค * ๋ํ์ด๊ฐ ๋ฐ์ดํ ํธ๋ค๋ง์ ํจ์จ์ ์ผ๋ก ์ฝ๊ณ ํธํ๊ณ ํ ์ ์๋ค. ๊ทธ๋ฌ๋ ๋ฐ์ดํ ํธ๋ค๋ง์ ์ฃผ๋ก ์ฌ์ฉํ๋ ํ๋ค์ค๋ ๋ง์ ๋ถ๋ถ์ด ๋ํ์ด๋ฅผ ๊ธฐ๋ฐ์ผ๋ก ๋ง๋ค์ด์ ธ ์๋ค. * ndarray - ๋ํ์ด ๊ธฐ๋ฐ ๋ฐ์ดํ ํ์
- np.array() * ์๋ฃํ ์ ๋ฆฌ - ํ์ด์ฌ (list / tuple) - numpy์ ndarray - pandas์ DataFrame / series | import numpy as np
list_1 = [1, 2, 3]
list_2 = [9, 8, 7]
arr = np.array([list_1, list_2])
arr
print(type(arr))
print(arr.shape)
print(arr.ndim)
# ๋๋ฒ์งธํ์ ๋๋ฒ์งธ ์ด์ ๊ฐ์ 100 ์ง์
arr[[1], 1] = 100
arr
arr1 = np.array([1, 2, 3])
arr2 = np.array([[1, 2, 3]])
print(arr1.shape) # 1์ฐจ์ ํํ๋ก 3๊ฐ์ ์์๋ฅผ ๊ฐ์ง
print(arr2.shape) # 1ํ 3์ด์ ์์๋ฅผ ๊ฐ์ง 2์ฐจ์
# ์ฐจ์์ ํ์ธ
print(arr1.ndim, "์ฐจ์")
print(arr2.ndim, "์ฐจ์") | (3,)
(1, 3)
1 ์ฐจ์
2 ์ฐจ์
| MIT | jupyter/dAnalysis/b_numpy_class/Ex01_dnarray.ipynb | WoolinChoi/test |
์๋ฃํ | # ์๋ฃํ ํ์ธ
print(type(arr))
# ์์์ ์๋ฃํ ํ์ธ
print(arr.dtype)
# ์์์ ์๋ฃํ ๋ณ๊ฒฝ
arr2 = arr.astype(np.float64)
print(arr2.dtype)
list1 = [1, 2, 3.6]
print(type(list1))
# ndarray ๋ณ๊ฒฝ
list1 = np.array(list1)
print(type(list1))
# ์์
list1.astype(np.float64)
print(list1.dtype)
list2 = [1, 2.3, 'python']
print(type(list2))
# ndarray ๋ณ๊ฒฝ
list2 = np.array(list2)
print(type(list2))
# ์์
print(list2.dtype) # U32
# [๊ฒฐ๊ณผ] <U11 : Unicode ๋ฌธ์์ด | _____no_output_____ | MIT | jupyter/dAnalysis/b_numpy_class/Ex01_dnarray.ipynb | WoolinChoi/test |
nparray๋ฅผ ํธ๋ฆฌํ๊ฒ ์์ฑ* arange() : ๋ฒ์๋ฅผ ์ด์ฉํ ๋ฐฐ์ด ๋ง๋ค๊ธฐ* zeros() : 0์ผ๋ก ์ฑ์ฐ๋ ๋ฐฐ์ด ๋ง๋ค๊ธฐ* ones() : 1๋ก ์ฑ์ฐ๋ ๋ฐฐ์ด ๋ง๋ค๊ธฐ | a = np.arange(10)
print(a)
b = np.arange(1, 11)
print(b)
c = np.arange(1, 11, 3)
print(c)
a2 = np.zeros((5, 5)) # ๊ธฐ๋ณธ์๋ฃํ: float
a2
a3 = np.ones((3, 4), dtype='int32')
a3 | _____no_output_____ | MIT | jupyter/dAnalysis/b_numpy_class/Ex01_dnarray.ipynb | WoolinChoi/test |
ndarray์ ์ฐจ์๊ณผ ํฌ๊ธฐ๋ฅผ ๋ณ๊ฒฝ : reshape() | arr = np.arange(10)
print(arr)
print(arr.shape)
print(arr.ndim) # ์ฐจ์์ผ๋ก ํ์ธ ๊ถ์ฅ
arr2 = arr.reshape(2, 5) # 2ํ 5์ด๋ก ์ฐจ์ํฌ๊ธฐ ๋ณ๊ฒฝ
print(arr2)
print(arr2.shape)
print(arr2.ndim)
# -1 ์ ์ฉ
arr = np.arange(20)
print(arr)
arr2 = arr.reshape(5, -1)
print(arr2)
arr3 = arr.reshape(-1, 2)
print(arr3) | [ 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19]
[[ 0 1 2 3]
[ 4 5 6 7]
[ 8 9 10 11]
[12 13 14 15]
[16 17 18 19]]
[[ 0 1]
[ 2 3]
[ 4 5]
[ 6 7]
[ 8 9]
[10 11]
[12 13]
[14 15]
[16 17]
[18 19]]
| MIT | jupyter/dAnalysis/b_numpy_class/Ex01_dnarray.ipynb | WoolinChoi/test |
์ธ๋ฑ์ฑ: ํน์ ๋ฐ์ดํ ์ถ์ถ | #------------------------------------------ (1) ๋จ์ผ๊ฐ ์ถ๋ ฅ
import numpy as np
arr = np.arange(1, 11)
print(arr)
## ์ธ๋ฒ์งธ ์์ ์ถ์ถ ( 0๋ถํฐ ์ธ๋ฑ์ค)
print(arr[3])
## ๋ค์์ ์ธ๋ฒ์งธ ์์ ์ถ์ถ ( ๋ค์์ ์ธ๋ฑ์ค๋ -1๋ถํฐ)
print(arr[-3])
## 1๋ถํฐ 9๊น์ง nparray๋ฅผ ๋ง๋ค๊ณ 3ํ 3์ด 2์ฐจ์ ๊ตฌ์กฐ๋ก ๋ณ๊ฒฝํํ
## ๋๋ฒ์งธ ํ์ ์ธ๋ฒ์งธ ์ด์ ๊ฐ ์ถ์ถ
arr = np.arange(1, 10)
print(arr)
arr2 = arr.reshape(3, 3)
print(arr2.ndim)
print(arr2)
print(arr2[[1], 2])
#------------------------------------------ (2) ์ฌ๋ผ์ด์ฑ (:)
arr = np.arange(1, 10)
print(arr)
# 2๋ฒ์งธ๋ถํฐ 4๋ฒ์งธ๊น์ง์ ์์ ์ถ์ถ
print(arr[1:4])
# 2๋ฒ์งธ๋ถํฐ ๋ง์ง๋ง๊น์ง ์์ ์ถ์ถ
print(arr[1:])
# ์ฒ์๋ถํฐ 4๋ฒ์งธ๊น์ง ์์ ์ถ์ถ
print(arr[0:4])
# 2์ฐจ์ ๋ฐฐ์ด์์ ์์ฑ
'''
1 2 3
4 5 6
7 8 9
'''
arr = np.arange(1, 10)
ndarr = arr.reshape(3,3)
print(ndarr)
## ๊ทธ๋ฆผ์์ 1, 2, 4, 5 ์ถ์ถ
print(ndarr[[0, 1], :2])
## ๊ทธ๋ฆผ์์ 4, 5, 6, 7, 8, 9 ์ถ์ถ
print(ndarr[[1, 2], :])
## ๊ทธ๋ฆผ์์ 2, 3, 5, 6 ์ถ์ถ
print(ndarr[[0, 1], 1:])
## ๊ทธ๋ฆผ์์ 1, 4 ์ถ์ถ
print(ndarr[[0, 1], :1])
## ๊ทธ๋ฆผ์์ ์ ์ฒด ์์ ์ถ์ถ
print(ndarr[:, :])
print(ndarr[::])
print(ndarr)
# 2์ฐจ์ ndarray์์ ๋ค์ ์ค๋ ์ธ๋ฑ์ค๊ฐ ์์ผ๋ฉด 1์ฐจ์์ผ๋ก ๋ฐํ
# 3์ฐจ์ ndarray์์ ๋ค์ ์ค๋ ์ธ๋ฑ์ค๊ฐ ์์ผ๋ฉด 2์ฐจ์์ผ๋ก ๋ฐํ
print(ndarr[1][1])
print(ndarr[1]) # 1์ฐจ์ ndarray
# ์ฌ๋ผ์ด์ฑ[:]๊ณผ ์ธ๋ฑ์ค์ ํ์ ๊ฒฐํฉ
'''
1 2 3
4 5 6
7 8 9
'''
import numpy as np
arr = np.arange(1, 10)
ndarr = arr.reshape(3,3)
print(ndarr)
## ๊ทธ๋ฆผ์์ 1, 2, 4, 5 ์ถ์ถ
print(ndarr[[0, 1], :2])
print(ndarr[:2, :2])
# print(ndarr[[0, 1], [0, 1]])
## ๊ทธ๋ฆผ์์ 4, 5, 6, 7, 8, 9 ์ถ์ถ
print(ndarr[[1, 2], :3])
## ๊ทธ๋ฆผ์์ 2, 3, 5, 6 ์ถ์ถ
print(ndarr[[0, 1], 1:3])
## ๊ทธ๋ฆผ์์ 1, 4 ์ถ์ถ
print(ndarr[[0 ,1], :1])
## ๊ทธ๋ฆผ์์ 3, 6 ์ถ์ถ
print(ndarr[[0, 1], 2:])
# ์ง๊ธ์ ๋๋ฌด ๊ฐ๋จํ๋ค๊ณ ์ฌ๊ธฐ์ง๋ง ์ถํ์ ๋ฐ์ดํ์์ ๋ง์ด ํท๊ฐ๋ ค์ ์ฌ๊ธฐ์ ์๊ฐ ์์๋ฅผ ๋ง์ด ํ๋ค
#------------------------------------------ (3) ๋ธ๋ฆฐ์ธ๋ฑ์ฑ
# ์กฐ๊ฑด ํํฐ๋ง๊ณผ ๊ฒ์์ ๊ฐ์ด ํ๊ธฐ์ ์์ฃผ ์ฌ์ฉ
arr = np.arange(1, 10)
ndarr = arr.reshape(3,3)
print(ndarr)
# 5๋ณด๋ค ํฐ ์์๋ค ์ถ์ถ
print(ndarr > 5)
# 8๊ฐ ์์๋ฅผ 88๋ก ๋ณ๊ฒฝ
ndarr[[2], 1] = 88
ndarr | [[1 2 3]
[4 5 6]
[7 8 9]]
[[False False False]
[False False True]
[ True True True]]
| MIT | jupyter/dAnalysis/b_numpy_class/Ex01_dnarray.ipynb | WoolinChoi/test |
Hello PixieDust!This sample notebook provides you with an introduction to many features included in PixieDust. You can find more information about PixieDust at https://ibm-watson-data-lab.github.io/pixiedust/. To ensure you are running the latest version of PixieDust uncomment and run the following cell. Do not run this cell if you installed PixieDust locally from source and want to continue to run PixieDust from source. | #!pip install --user --upgrade pixiedust | _____no_output_____ | Apache-2.0 | notebook/Intro to PixieDust.ipynb | jordangeorge/pixiedust |
Import PixieDustRun the following cell to import the PixieDust library. You may need to restart your kernel after importing. Follow the instructions, if any, after running the cell. Note: You must import PixieDust every time you restart your kernel. | import pixiedust | _____no_output_____ | Apache-2.0 | notebook/Intro to PixieDust.ipynb | jordangeorge/pixiedust |
Enable the Spark Progress MonitorPixieDust includes a Spark Progress Monitor bar that lets you track the status of your Spark jobs. You can find more info at https://ibm-watson-data-lab.github.io/pixiedust/sparkmonitor.html. Run the following cell to enable the Spark Progress Monitor: | pixiedust.enableJobMonitor(); | _____no_output_____ | Apache-2.0 | notebook/Intro to PixieDust.ipynb | jordangeorge/pixiedust |
Example use of the PackageManagerYou can use the PackageManager component of Pixiedust to install and uninstall maven packages into your notebook kernel without editing configuration files. This component is essential when you run notebooks from a hosted cloud environment and do not have access to the configuration files. You can find more info at https://ibm-watson-data-lab.github.io/pixiedust/packagemanager.html. Run the following cell to install the GraphFrame package. You may need to restart your kernel after installing new packages. Follow the instructions, if any, after running the cell. | pixiedust.installPackage("graphframes:graphframes:0.1.0-spark1.6")
print("done") | _____no_output_____ | Apache-2.0 | notebook/Intro to PixieDust.ipynb | jordangeorge/pixiedust |
Run the following cell to print out all installed packages: | pixiedust.printAllPackages() | _____no_output_____ | Apache-2.0 | notebook/Intro to PixieDust.ipynb | jordangeorge/pixiedust |
Example use of the display() APIPixieDust lets you visualize your data in just a few clicks using the display() API. You can find more info at https://ibm-watson-data-lab.github.io/pixiedust/displayapi.html. The following cell creates a DataFrame and uses the display() API to create a bar chart: | sqlContext=SQLContext(sc)
d1 = sqlContext.createDataFrame(
[(2010, 'Camping Equipment', 3),
(2010, 'Golf Equipment', 1),
(2010, 'Mountaineering Equipment', 1),
(2010, 'Outdoor Protection', 2),
(2010, 'Personal Accessories', 2),
(2011, 'Camping Equipment', 4),
(2011, 'Golf Equipment', 5),
(2011, 'Mountaineering Equipment',2),
(2011, 'Outdoor Protection', 4),
(2011, 'Personal Accessories', 2),
(2012, 'Camping Equipment', 5),
(2012, 'Golf Equipment', 5),
(2012, 'Mountaineering Equipment', 3),
(2012, 'Outdoor Protection', 5),
(2012, 'Personal Accessories', 3),
(2013, 'Camping Equipment', 8),
(2013, 'Golf Equipment', 5),
(2013, 'Mountaineering Equipment', 3),
(2013, 'Outdoor Protection', 8),
(2013, 'Personal Accessories', 4)],
["year","zone","unique_customers"])
display(d1) | _____no_output_____ | Apache-2.0 | notebook/Intro to PixieDust.ipynb | jordangeorge/pixiedust |
Example use of the Scala bridgeData scientists working with Spark may occasionaly need to call out to one of the hundreds of libraries available on spark-packages.org which are written in Scala or Java. PixieDust provides a solution to this problem by letting users directly write and run scala code in its own cell. It also lets variables be shared between Python and Scala and vice-versa. You can find more info at https://ibm-watson-data-lab.github.io/pixiedust/scalabridge.html. Start by creating a python variable that we'll use in scala: | python_var = "Hello From Python"
python_num = 10 | _____no_output_____ | Apache-2.0 | notebook/Intro to PixieDust.ipynb | jordangeorge/pixiedust |
Create scala code that use the python_var and create a new variable that we'll use in Python: | %%scala
println(python_var)
println(python_num+10)
val __scala_var = "Hello From Scala" | _____no_output_____ | Apache-2.0 | notebook/Intro to PixieDust.ipynb | jordangeorge/pixiedust |
Use the __scala_var from python: | print(__scala_var) | _____no_output_____ | Apache-2.0 | notebook/Intro to PixieDust.ipynb | jordangeorge/pixiedust |
Sample DataPixieDust includes a number of sample data sets. You can use these sample data sets to start playing with the display() API and other PixieDust features. You can find more info at https://ibm-watson-data-lab.github.io/pixiedust/loaddata.html. Run the following cell to view the available data sets: | pixiedust.sampleData() | _____no_output_____ | Apache-2.0 | notebook/Intro to PixieDust.ipynb | jordangeorge/pixiedust |
Example use of sample dataTo use sample data locally run the following cell to install required packages. You may need to restart your kernel after running this cell. | pixiedust.installPackage("com.databricks:spark-csv_2.10:1.5.0")
pixiedust.installPackage("org.apache.commons:commons-csv:0") | _____no_output_____ | Apache-2.0 | notebook/Intro to PixieDust.ipynb | jordangeorge/pixiedust |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.