forked from bukped/ai
/
main.go
386 lines (313 loc) · 9.4 KB
/
main.go
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
package main
import (
"encoding/csv"
"flag"
"fmt"
"image"
"image/jpeg"
"log"
"math"
"math/rand"
"os"
"strconv"
_ "net/http/pprof"
"github.com/pkg/errors"
"gorgonia.org/gorgonia"
"gorgonia.org/gorgonia/examples/mnist"
"gorgonia.org/tensor"
"time"
"gopkg.in/cheggaaa/pb.v1"
)
var (
epochs = flag.Int("epochs", 100, "Number of epochs to train for")
dataset = flag.String("dataset", "train", "Which dataset to train on? Valid options are \"train\" or \"test\"")
dtype = flag.String("dtype", "float64", "Which dtype to use")
batchsize = flag.Int("batchsize", 100, "Batch size")
cpuprofile = flag.String("cpuprofile", "", "CPU profiling")
)
const loc = "./mnist/"
var dt tensor.Dtype
func parseDtype() {
switch *dtype {
case "float64":
dt = tensor.Float64
case "float32":
dt = tensor.Float32
default:
log.Fatalf("Unknown dtype: %v", *dtype)
}
}
type nn struct {
g *gorgonia.ExprGraph
w0, w1, w2 *gorgonia.Node
out *gorgonia.Node
predVal gorgonia.Value
}
type sli struct {
start, end int
}
func (s sli) Start() int { return s.start }
func (s sli) End() int { return s.end }
func (s sli) Step() int { return 1 }
func newNN(g *gorgonia.ExprGraph) *nn {
// Create node for w/weight
w0 := gorgonia.NewMatrix(g, dt, gorgonia.WithShape(784, 300), gorgonia.WithName("w0"), gorgonia.WithInit(gorgonia.GlorotN(1.0)))
w1 := gorgonia.NewMatrix(g, dt, gorgonia.WithShape(300, 100), gorgonia.WithName("w1"), gorgonia.WithInit(gorgonia.GlorotN(1.0)))
w2 := gorgonia.NewMatrix(g, dt, gorgonia.WithShape(100, 10), gorgonia.WithName("w2"), gorgonia.WithInit(gorgonia.GlorotN(1.0)))
return &nn{
g: g,
w0: w0,
w1: w1,
w2: w2,
}
}
func (m *nn) learnables() gorgonia.Nodes {
return gorgonia.Nodes{m.w0, m.w1, m.w2}
}
func (m *nn) fwd(x *gorgonia.Node) (err error) {
var l0, l1, l2 *gorgonia.Node
var l0dot, l1dot *gorgonia.Node
// Set first layer to be copy of input
l0 = x
// Dot product of l0 and w0, use as input for ReLU
if l0dot, err = gorgonia.Mul(l0, m.w0); err != nil {
return errors.Wrap(err, "Unable to multiply l0 and w0")
}
// l0dot := gorgonia.Must(gorgonia.Mul(l0, m.w0))
// Build hidden layer out of result
l1 = gorgonia.Must(gorgonia.Rectify(l0dot))
// MOAR layers
// l2dot := gorgonia.Must(gorgonia.Mul(l1, m.w1))
if l1dot, err = gorgonia.Mul(l1, m.w1); err != nil {
return errors.Wrap(err, "Unable to multiply l1 and w1")
}
l2 = gorgonia.Must(gorgonia.Rectify(l1dot))
var out *gorgonia.Node
if out, err = gorgonia.Mul(l2, m.w2); err != nil {
return errors.Wrapf(err, "Unable to multiply l2 and w2")
}
// m.pred = l3dot
// gorgonia.Read(m.pred, &m.predVal)
// return nil
m.out, err = gorgonia.SoftMax(out)
gorgonia.Read(m.out, &m.predVal)
return
}
const pixelRange = 255
func reversePixelWeight(px float64) byte {
// return byte((pixelRange*px - pixelRange) / 0.9)
return byte(pixelRange*math.Min(0.99, math.Max(0.01, px)) - pixelRange)
}
func visualizeRow(x []float64) *image.Gray {
// since this is a square, we can take advantage of that
l := len(x)
side := int(math.Sqrt(float64(l)))
r := image.Rect(0, 0, side, side)
img := image.NewGray(r)
pix := make([]byte, l)
for i, px := range x {
pix[i] = reversePixelWeight(px)
}
img.Pix = pix
return img
}
func main() {
flag.Parse()
parseDtype()
rand.Seed(7945)
// // intercept Ctrl+C
// sigChan := make(chan os.Signal, 1)
// signal.Notify(sigChan, syscall.SIGINT, syscall.SIGTERM)
// doneChan := make(chan bool, 1)
var inputs, targets tensor.Tensor
var err error
// load our data set
trainOn := *dataset
if inputs, targets, err = mnist.Load(trainOn, loc, dt); err != nil {
log.Fatal(err)
}
numExamples := inputs.Shape()[0]
bs := *batchsize
// MNIST data consists of 28 by 28 black and white images
// however we've imported it directly now as 784 different pixels
// as a result, we need to reshape it to match what we actually want
// if err := inputs.Reshape(numExamples, 1, 28, 28); err != nil {
// log.Fatal(err)
// }
// we should now also proceed to put in our desired variables
// x is where our input should go, while y is the desired output
g := gorgonia.NewGraph()
// x := gorgonia.NewTensor(g, dt, 4, gorgonia.WithShape(bs, 1, 28, 28), gorgonia.WithName("x"))
x := gorgonia.NewMatrix(g, dt, gorgonia.WithShape(bs, 784), gorgonia.WithName("x"))
y := gorgonia.NewMatrix(g, dt, gorgonia.WithShape(bs, 10), gorgonia.WithName("y"))
// ioutil.WriteFile("simple_graph.dot", []byte(g.ToDot()), 0644)
m := newNN(g)
if err = m.fwd(x); err != nil {
log.Fatalf("%+v", err)
}
// ioutil.WriteFile("simple_graph_2.dot", []byte(g.ToDot()), 0644)
losses, err := gorgonia.HadamardProd(m.out, y)
if err != nil {
log.Fatal(err)
}
cost := gorgonia.Must(gorgonia.Mean(losses))
cost = gorgonia.Must(gorgonia.Neg(cost))
// we wanna track costs
var costVal gorgonia.Value
gorgonia.Read(cost, &costVal)
if _, err = gorgonia.Grad(cost, m.learnables()...); err != nil {
log.Fatal(err)
}
vm := gorgonia.NewTapeMachine(g, gorgonia.BindDualValues(m.learnables()...))
solver := gorgonia.NewRMSPropSolver(gorgonia.WithBatchSize(float64(bs)))
batches := numExamples / bs
log.Printf("Batches %d", batches)
bar := pb.New(batches)
bar.SetRefreshRate(time.Second / 20)
bar.SetMaxWidth(80)
for i := 0; i < *epochs; i++ {
// for i := 0; i < 1; i++ {
bar.Prefix(fmt.Sprintf("Epoch %d", i))
bar.Set(0)
bar.Start()
for b := 0; b < batches; b++ {
start := b * bs
end := start + bs
if start >= numExamples {
break
}
if end > numExamples {
end = numExamples
}
var xVal, yVal tensor.Tensor
if xVal, err = inputs.Slice(sli{start, end}); err != nil {
log.Fatal("Unable to slice x")
}
if yVal, err = targets.Slice(sli{start, end}); err != nil {
log.Fatal("Unable to slice y")
}
// if err = xVal.(*tensor.Dense).Reshape(bs, 1, 28, 28); err != nil {
// log.Fatal("Unable to reshape %v", err)
// }
if err = xVal.(*tensor.Dense).Reshape(bs, 784); err != nil {
log.Fatal("Unable to reshape %v", err)
}
gorgonia.Let(x, xVal)
gorgonia.Let(y, yVal)
if err = vm.RunAll(); err != nil {
log.Fatalf("Failed at epoch %d: %v", i, err)
}
// solver.Step(m.learnables())
solver.Step(gorgonia.NodesToValueGrads(m.learnables()))
vm.Reset()
bar.Increment()
}
bar.Update()
log.Printf("Epoch %d | cost %v", i, costVal)
}
bar.Finish()
log.Printf("Run Tests")
// load our test set
if inputs, targets, err = mnist.Load("test", loc, dt); err != nil {
log.Fatal(err)
}
numExamples = inputs.Shape()[0]
bs = *batchsize
batches = numExamples / bs
bar = pb.New(batches)
bar.SetRefreshRate(time.Second / 20)
bar.SetMaxWidth(80)
bar.Prefix(fmt.Sprintf("Epoch Test"))
bar.Set(0)
bar.Start()
for b := 0; b < batches; b++ {
start := b * bs
end := start + bs
if start >= numExamples {
break
}
if end > numExamples {
end = numExamples
}
var xVal, yVal tensor.Tensor
if xVal, err = inputs.Slice(sli{start, end}); err != nil {
log.Fatal("Unable to slice x")
}
if yVal, err = targets.Slice(sli{start, end}); err != nil {
log.Fatal("Unable to slice y")
}
// if err = xVal.(*tensor.Dense).Reshape(bs, 1, 28, 28); err != nil {
// log.Fatal("Unable to reshape %v", err)
// }
if err = xVal.(*tensor.Dense).Reshape(bs, 784); err != nil {
log.Fatal("Unable to reshape %v", err)
}
gorgonia.Let(x, xVal)
gorgonia.Let(y, yVal)
if err = vm.RunAll(); err != nil {
log.Fatalf("Failed at epoch test: %v", err)
}
arrayOutput := m.predVal.Data().([]float64)
yOutput := tensor.New(tensor.WithShape(bs, 10), tensor.WithBacking(arrayOutput))
for j := 0; j < xVal.Shape()[0]; j++ {
rowT, _ := xVal.Slice(sli{j, j + 1})
row := rowT.Data().([]float64)
img := visualizeRow(row)
// get label
yRowT, _ := yVal.Slice(sli{j, j + 1})
yRow := yRowT.Data().([]float64)
var rowLabel int
var yRowHigh float64
for k := 0; k < 10; k++ {
if k == 0 {
rowLabel = 0
yRowHigh = yRow[k]
} else if yRow[k] > yRowHigh {
rowLabel = k
yRowHigh = yRow[k]
}
}
// get prediction
predRowT, _ := yOutput.Slice(sli{j, j + 1})
predRow := predRowT.Data().([]float64)
var rowGuess int
var predRowHigh float64
// guess result
for k := 0; k < 10; k++ {
if k == 0 {
rowGuess = 0
predRowHigh = predRow[k]
} else if predRow[k] > predRowHigh {
rowGuess = k
predRowHigh = predRow[k]
}
}
f, _ := os.OpenFile(fmt.Sprintf("images/%d - %d - %d - %d.jpg", b, j, rowLabel, rowGuess), os.O_CREATE|os.O_WRONLY|os.O_TRUNC, 0644)
jpeg.Encode(f, img, &jpeg.Options{jpeg.DefaultQuality})
f.Close()
}
arrayOutput = m.predVal.Data().([]float64)
yOutput = tensor.New(tensor.WithShape(bs, 10), tensor.WithBacking(arrayOutput))
file, err := os.OpenFile(fmt.Sprintf("%d.csv", b), os.O_CREATE|os.O_WRONLY, 0777)
if err = xVal.(*tensor.Dense).Reshape(bs, 784); err != nil {
log.Fatal("Unable to create csv", err)
}
defer file.Close()
var matrixToWrite [][]string
for j := 0; j < yOutput.Shape()[0]; j++ {
rowT, _ := yOutput.Slice(sli{j, j + 1})
row := rowT.Data().([]float64)
var rowToWrite []string
for k := 0; k < 10; k++ {
rowToWrite = append(rowToWrite, strconv.FormatFloat(row[k], 'f', 6, 64))
}
matrixToWrite = append(matrixToWrite, rowToWrite)
}
csvWriter := csv.NewWriter(file)
csvWriter.WriteAll(matrixToWrite)
csvWriter.Flush()
vm.Reset()
bar.Increment()
}
log.Printf("Epoch Test | cost %v", costVal)
}